Vlad Kogan: How Politics Hijacks Education Policy and Hurts Kids

School board elections, culture wars, school closures, and more

We’re special here in Virginia. Due to a historical quirk, we hold important elections every year:

- In 2020, we voted for president and U.S. Congress;

- In 2021, we had statewide elections and a governor’s race;

- In 2022, it was the congressional midterms;

- In 2023, it was statewide elections again plus school boards and other local races; and

- In 2024, it was president and U.S. Congress again.

Next week, we’ll elect a new governor and vote in other statewide races.

That’s… a lot. And it doesn’t even count primaries or special elections. It costs time and money to run all those the elections. And, because schools often serve as polling locations, some districts cancel classes the first Tuesday in November (my kids will be off in Fairfax County).

It also has implications for who turns out to vote. In Fairfax, 72% of registered voters turned out for the last presidential election, the last governor’s race had 60% participation, and the last school board race was 41%.

Is that a good thing for schools? What affect do these off-cycle elections and low-turnout votes have on school boards?

Before heading to the polls, I wanted to talk with Vladimir Kogan, a Professor in the Department of Political Science at the Ohio State University. Vlad has a new book out about education politics and elections, and I wanted to pick his brain about the intersection of schools and politics.

What follows is a lightly edited transcript of our conversation.

Chad Aldeman: Your book uses a lot of provocative language. The title is No Adult Left Behind: How Politics Hijacks Education Policy and Hurts Kids. You write about “well-intentioned adults” operating “dysfunctional” schools. For folks who haven’t read the full book, can you briefly summarize your thesis and argument?

Vladimir Kogan: Yeah, it’s funny. The original title was going to be Anything for the Children! and the publisher said, “No way, that’s way too sarcastic.” So No Adult Left Behind was the more toned-down option.

My thesis is that a lot of the problems in K–12 education—the dysfunction—is baked in and hardwired thanks to its governance. We don’t spend enough time thinking about the incentives that governance creates. At a 10,000-foot level, most communities have elected school boards making key decisions about curriculum, teacher evaluations, compensation, and more.

The key disconnect is that school board elections are decided by adults, but the decisions mostly impact learning outcomes and opportunities for students. Many voters don’t have school-age children, so issues parents experience in real life become more abstract, symbolic issues for these voters. That introduces some of the perverse incentives that ultimately get in the way of providing a high-quality education.

Aldeman: What do we know about school boards? How many school board members are there? Let’s start there.

We have somewhere in the neighborhood of 13,000 school boards, and across those boards, about 50,000 school board members. Almost all are elected. In most states, elections are nonpartisan, although this is starting to change. Historically, 85–90% were nonpartisan.

In many cases these elections are held off cycle, meaning they are on the ballot at a different time than congressional or presidential elections. Rather than the voting taking place in November of even years, the voting may take place in different years or a different time of the year. And that has important implications for how many people vote and who those voters are.

The other thing we know is that school board elections are relatively uncompetitive. In many cases, there’s not a contested election, only one candidate is running, and the voters have limited choices. That, again, has implications for the incentives that elected officials face.

The other thing that we know is that, when voters do have a choice, it doesn’t seem like they put much weight on academic considerations. If a district provides a terrible education, incumbents are usually reelected at only slightly lower rates than if the district delivers high achievement growth. That doesn’t provide strong motivation for elected officials to improve the quality of the educational opportunity that they’re providing.

Aldeman: Can you say more about who the voters are in most school board elections? How many people vote, and who are they?

That’s hard to answer nationally because elections are so decentralized. But we have collected data from about 20 states, and we find that election timing is hugely important. When elections are on-cycle with a presidential election, turnout is around 50% of registered voters. During midterms, turnout is around 40%. Off-cycle elections see 10–20%.

The other thing we know is that most people who vote in school board elections do not have school-age children. This is especially true in off-cycle elections, because it turns out one of the strongest predictors of voting in them is age. Elderly people vote regardless of when the election is and so as you move down the gradient from higher to lower turnout elections, you get more and more elderly voters who do not have children living at home.

It’s almost never the case that parents are a majority of the electorate. In California, for example, if a school board election is held on the day of presidential race, about 30% of voters have children in their household. If it’s an off-cycle election, that drops to 20–22%.

Aldeman: Some people might argue this is a good thing—that it means that the most attentive voters are the ones with the most influence. How would you respond to that?

That’s ultimately an empirical claim, and I would like to see some evidence that that’s the case.

In fact, there’s a famous Supreme Court case from the 1960s in New York, Kramer v. Union Free School District, involving a state law that said to vote in school board elections, you either had to be a property owner or to be a parent. The theory for that law was that those are the only groups that actually had good information and that were incentivized and motivated enough to pay attention. That law was struck down, but I think the logic behind the law conflicts with the argument that elderly people who don’t have kids in schools are going to be paying more attention than parents.

There’s a really nice paper by Julia Payson, a political science professor at UCLA. She looked at California, and California is great because you have within-state variation on when school board elections are held. She looked to see, are incumbents doing better if, during their term, academic outcomes improve? She finds that to be the case, but only in on-cycle elections in presidential years. The rest of the time, school board incumbents are not held accountable for academic outcomes.

Aldeman: If voters don’t care about student outcomes, what do they do care about? What are they paying attention to?

That’s a great question. What we can say in the aggregate is that whatever is driving their decisions, it’s not academics.

Now, I have some hypotheses. To a large extent partisan considerations play a big role, even though many of these are nonpartisan elections. I think many voters can figure out who’s the “D” and who’s the “R.” The kinds of things that animate the national dialog about education, whether that’s transgender issues, The 1619 Project, or other issues of the day, those things seem to motivate a lot of voters. Elderly people read the news more. They pay more attention. So they’re probably relying on partisan cues.

Relatedly, that’s one reason why some of the culture war things play an outsized role and why we have so much conflict over the books that are in libraries, or whether there are masks in schools during COVID, because those happen to be kind of the topics that people who are politically engaged are especially incensed and interested in, regardless of whether that translates to academic performance.

Aldeman: Ok, let’s say I’m with you on aligning election cycles. Talk to me about what the potential impacts would be. How much of a benefit would that be for students?

Again, the best evidence we have is from California. We actually have two different studies. One is the Julia Payson study I mentioned earlier. There was another dissertation by an economics student at the RAND Corporation that looks at the late 1990s when California first created its school accountability system, and it also finds that it led to greater accountability for performance, but only during on-cycle elections.

Now we wouldn’t want to make wholesale changes based in California alone, so I would like to see much more careful evaluation. We do have some recent research using national data where we actually don’t find dramatic differences based on election timing. So I’d say the evidence is circumstantial at best.

But I think it’s encouraging because there are other independent reasons to prefer on-cycle elections. Turnout goes up, which many people think is a good thing in of itself. The share of parents who are voting goes up, which is good because those are the voters that have the most skin in the game. And we have some evidence that it leads to greater accountability.

If I was to try to play devil’s advocate, what would be the strongest counter arguments? One is that on-cycle elections tend to have less-informed voters in general, so they might be more reliant on other sources of information. There’s some research suggesting that the incumbency advantage goes up in on-cycle elections, because voters may recognize the incumbent and vote for them by default.

The other possible consequence is the cost of running an election. On-cycle elections may be more expensive because school board races would be running at the same time as state- and national-level contests. However, I don’t think school board candidates are buying a lot of TV time, so I don’t think most of those costs translate. But there is competition for scarce attention. And so we might imagine that when all the races are on the ballot at the same time, maybe voters can’t pay attention as much to any given one of them.

Aldeman: You’re talking about the costs of running for election in a more crowded environment?

Right.

Aldeman: What else do you recommend? What other recommendations do you have for fixing some of the issues with school board elections?

The timing issue is low hanging fruit, because everybody loves it for their own reasons, and I think there’s overwhelming popular support as well in public opinion.

The other thing is more subtle, but policymakers can help make academic performance more salient. One reason why voters don’t put much weight on academic performance it that it’s not necessarily top of mind when they go to vote.

There are ways to change that. To think outside the box, I would love to print school accountability information right on the ballots next to the candidate names. Some states already do something similar for school bonds. When a district wants to pass a bond in Texas, it has to tell voters exactly how much it’s going to cost, to remind them of the financial stakes. Why not do the same thing with academics? We could design the ballot to put performance information right there, helping voters and encouraging them in a very subtle way to think about academic performance. That would be worth trying.

In the modern school choice context, where parents have a lot of choices, it’s hard for parents to really do an apples-to-apples comparison, because it’s hard to separate school quality from the demographics of students who are being served. Both voters and parents hold schools accountable for things largely outside of their control, and that creates some perverse consequences. Making performance information, especially ratings based on growth, more salient would help with this too.

One thing I definitely would not support is adding party labels. That is gaining steam though. Tennessee, a few years ago, introduced partisan elections for school boards. Indiana just recently voted to do so and other, mostly red, states are thinking about that as well.

The motivation makes sense. The idea is that we want to provide information to voters so they can figure out who’s aligned with them. Unfortunately, if you agree with my argument that these national labels are mostly about divisive culture war issues, it’s not obvious that that’s going to necessarily help voters align their preferences on academics.

We also have some evidence from elections to other offices that it actually could have the opposite effect. It could crowd out academic performance. If you look at the state level, most states elect their judges, and how they run these elections will vary dramatically. Some states have retention elections, where you just vote yes or no on the incumbents. Some states have nonpartisan elections, like we do for school boards, and some have partisan elections. And some compelling research from Jim Snyder at Harvard shows that when you add the party labels to judicial elections, those play an outsized role. They dominate every other consideration.

There might be other ways of thinking about accountability. Voting in elections is not the only method. Another option is school choice, and we do have evidence from places like Florida that when school choice expands, not only does that impact the students who take advantage of those options, but it impacts the students who stay behind in the traditional public schools. This is what people call competitive effects. Now that students and parents have the option of leaving, it changes the internal politics of districts. It creates some incentive to improve the offerings to try to retain students. I see that as a complement to electoral accountability.

Aldeman: That brings to mind two things here in Virginia. One, Republican Governor Glenn Youngkin worked to increase the transparency and salience of school performance. I supported that work as a consultant. The second thing is that next month I get to vote on whether Fairfax should issue a $460 million school bond.

Anyway, now I’m going to throw a hard one at you. You write that a campaign to eliminate attendance boundaries will almost certainly not succeed. You write about the political power of education NIMBYs, not to mention unions and other community organizations. So how do we get to your vision of a school district that values student performance above all other priorities?

The first step is to acknowledge some of the dysfunction and acknowledge the extent to which our idealistic vision of democracy really doesn’t match how democracy works in practice. It still may be the best system that we have! It may be that we’re stuck with a lot of these pathologies.

Changing the election timing is, I think, feasible. Many states have done it already. I think making performance information more salient, even if you don’t print it on the ballot, is feasible. School choice is expanding.

I don’t want to overstate the extent to which these things will work. I think they’re positive, but probably modest in terms of potential effects.

One of the problems that I think is baked into governance is that for voters who don’t have school-age kids, they still might have skin in the game in terms of their property values. For most Americans, home equity is how they save money. And so it makes total sense that they will care a lot about policies that impact their property values. We also know that homeowners vote at much higher rates. They are especially over-represented in low turnout elections. In principle there’s nothing wrong with that, because, again, they actually have some interest at stake.

The challenge is, once you dig into the real estate literature about what does and does not impact home values, it turns out that how much students learn does not get priced into home values very much. That means homeowners don’t have much incentive to care about school quality, because quality is not being picked up by home values.

What does get picked up is things like raw proficiency rates and the demographic composition of students. If you’re a totally reasonable person who just wants to protect your nest egg and your home equity, you’d advocate policies that exclude disadvantaged, low income, and low-performing students from your district.

Maybe the solution is just to get rid of school attendance zones, and make every school open enrollment? I just don’t see how that works in practice. Because you need some tie-breaking mechanism. Because there’s always going to be more students that want to go to high-performing schools than there are seats. And so even in the world of open enrollment for everyone, geography is still going to be a tiebreaker.

There are some places that have tried to go the whole randomized route. I think the cautionary tale is San Francisco, which, in the context of their school desegregation program, essentially had random assignment of students to schools. Parents hated it because it could mean that their kid was on the bus for an hour a day. No one could predict where their child would go to school, because it was essentially random. And it led to a number of parents leaving the public schools for private schools.

I’m curious what you think. Is that too pessimistic, or are there models where the idea of no attendance boundary works?

Aldeman: I’m optimistic at the margins. Open enrollment seems to be expanding across the country. The Reason Foundation has tracked some of those improvements. I’m a fan of local efforts like the Boston METCO program or the Los Angeles Zones of Choice. Charters are growing, ESAs are growing, so there’s some hope there.

When we’re talking about charters and ESAs, those are alternative providers that are not geographically based. But I’m pretty skeptical about open enrollment, because when you have an open enrollment policy it’s almost always subject to capacity constraints. Schools have to accept students if they have space. The core issue is that the most desirable schools don’t have extra space. Unless there’s a way to expand capacity, I don’t see how open enrollment can dramatically move the needle for that reason.

Aldeman: Let’s talk about school closures. Talk to me about what you see in the data around school closures. The accusation is that school closures disproportionately affect black and low-income families and students. What did you see that in your data?

Just to be clear, when I’m talking about school closures, I’m talking about permanently closing a building, not pandemic-related temporary disruptions.

There’s a longstanding charge, going back at least a decade, that school closures are inherently racist because they disproportionately affect communities of color. To a certain extent it is true that when schools close, those are disproportionately happening in large urban districts that serve students of color. And within those districts, the closures have been concentrated in neighborhoods that have higher minority student populations.

However, that is mostly because those are the places where school enrollment has declined the most. So even if you were deciding to close schools based solely on enrollment, if you have a school that is operating at half of capacity, and you don’t look at race at all, that’s mechanically going to impact heavily minority schools.

And that’s essentially what I find. Once you account for all the factors at the same time, race does not play a huge role. The number one predictor in which school gets closed is enrollment.

The second predictor is student achievement levels. In some ways that’s encouraging, because it suggests that academic considerations are playing a role, but it’s also discouraging in that it’s not actually a reflection of school quality. Instead, it’s mostly isolating the demographic composition of students. School quality as measured by learning gains from year to year doesn’t seem to predict which schools close.

Aldeman: That reminded me of a study about school closures in Newark, NJ. The study found they had a positive impact, overall, but the district didn’t close its absolute lowest-performing schools in terms of growth. If they had done that, they would have improved student performance even more. So it’s just a good reminder to focus on student growth when considering which schools to close.

Nationally, we don’t see that growth is a consistent predictor at all of which schools are closed.

If our goal is to improve academic opportunity, growth is the best metric we have to tell us if a school is better than an alternative. And right now districts, in the aggregate at least, are not looking at that at all to decide what schools they close.

Now, some of that may be in tension with the enrollment and financial concerns. A district trying to balance its budget may be constrained about how much it can consider growth if it needs to respond to enrollment declines. That is, the lowest-growth schools may not necessarily be the smallest or most underenrolled schools.

Aldeman: Let’s talk about culture wars. When you went to study this, how did you define what is a “culture war?” And what impact do they have on schools and students?

A “culture war” is one of those you-know-it-when-you-see-it type of things. The way I define it is simple. I outsourced that work to the Cato Institute. Cato has a great data set called the Public Schooling Battle Map. Anytime there’s a controversy involving schools that gets covered in the media, they put it in their database. It includes controversies involving dress code, school mascots, and of course curriculum decisions. Do we teach evolution? Do we teach the 1619 Project or the 1776 Commission version of history? It’s those kinds of issues.

What I do is to ask, when a school district becomes embroiled in one of these controversies, what happens to the academics in that community? The model I have in my mind borrows from the literature on policing. One of the things we see is that when activists protest based on well-intentioned concerns around racial disparities in policing, it can backfire. Crime might go up because police officers pull back, or maybe citizen cooperation goes down. Some people call this the “Ferguson Effect” in policing.

I looked to see if there was a similar “Ferguson Effect” in public education involving culture wars. If there’s a big blow up about curriculum, or about school mascots, and the superintendent gets fired, and the deputy gets fired, and the community is up in arms, and the parents are screaming at each other, that could filter down to the classroom and disrupt learning.

And that’s essentially what I find. There is some evidence, particularly for race and ethnicity-related controversies and free speech controversies where students bear some negative consequences of attending school at a time that a big controversy is happening in their district.

The challenge is that culture wars are bipartisan, and when we think about “culture wars” we often associate them with people we disagree with. On the flip side, what we view as totally legitimate and reasonable concerns on our side are seen as culture wars on the other side.

I’ll give you two concrete examples that are not in the book but are more recent. Montgomery County, Maryland had LGBT members of their community, so they wanted to assign books that have LGBT characters. As an example, there was a book called Uncle Bobby’s Wedding with two men getting married. And for many people, there’s nothing wrong with that. It’s not like the district was pushing same-sex marriage. But some parents objected and wanted to opt out, and the case went all the way up to the Supreme Court.

On the other hand, Tennessee and a few other states have recently passed a mandate that all human development classes have to show an AI video called “Meet Baby Olivia” that shows the fetal development in real time. It’s mostly scientific, mostly objective, but it begins with a statement that “life begins at conception” which, to half of the country, may sound totally reasonable, but to others it may sound like indoctrination.

Those two controversies illustrate what I mean by culture wars. I think they illustrate why half of the population sees absolutely nothing wrong with something while the other half sees schools trying to push a particular perspective onto kids.

There is kind of an arms race component. If people you disagree with are going to push their agenda onto schools, then you feel like you can’t just unilaterally back down. But I’m in favor of disarmament. I totally understand that everybody has good intentions, but using K-12 education as the arena to fight these battles may not be the right place to do it, because it’s going to have these downstream, disruptive impacts. It’s not to say that these perspectives are illegitimate, but public schools are not the right place to litigate and to fight them out. To the extent that there are contentious ideas, it’s not great to drag the schools into those battles.

To hear more from Vlad or to dive in more on his research, give him a follow and consider buying his book No Adult Left Behind.