The Education Gadfly Weekly

Set aside for a moment the debate about whether Elon Musk’s “Department of Government Efficiency” (DOGE) is an honest effort to cut waste, fraud, and abuse from the federal government. Let’s ask a different question: What would a serious effort to get more bang from our education buck look like?

First let’s state the obvious: We would want to start at the local level because, like money in banks, that’s where the spending is. You could get rid of the entire federal role in education and only save 10 percent of the cost of our schools, given that 90 cents of each dollar comes from the state and local levels.

It’s also undeniable that we spend most of our education budget on staff—77.5 percent according to NCES. What’s perhaps not well known is that fewer than half of those employees are classroom teachers. Figuring out how to dramatically reduce the number of administrators and other non-teachers employed in our schools—as well as central offices—is a worthy goal.

­

Chart via Corey DeAngelis.

And yes, we could also wonder whether we might get by as well or better with fewer teachers (and larger classes). As Checker Finn remarked a few years ago, we could be paying every teacher six figure salaries if we had been willing to keep our class sizes moderate rather than pushing to make them smaller and smaller over the past half-century.

We’ve also known forever that paying teachers more for getting a master’s degree doesn’t add any value. It’s just a giveaway to the higher education industrial complex.

If we were willing to tackle just these three issues alone, we could achieve major savings without breaking too much of a sweat. Consider these back-of-the-envelope calculations:

1. Imagine that we reduced the number of non-teachers in our school system back down to the level we had in 2000, an era when our schools were making dramatic gains in student achievement. That would mean going from about 3.5 million non-teachers today to fewer than 3 million (the number back then). At a very conservative estimate of $41,000 per staffer (salaries plus benefits), that would save about $20 billion annually.

2. Now do the same when it comes to teachers and compute how much we could save by returning to 2000-era class sizes. That would mean increasing class sizes, on average, to about twenty-one for elementary schools and twenty-four for secondary, up from about nineteen and seventeen today.[1] Do the math and it means we need about 2.25 million teachers instead of the 2.79 million we have now, a reduction of roughly 539,000. At an average cost of about $87,000 per teacher (salaries plus benefits), that would save about $47 billion annually. 

3. Now let’s see what we can save by eliminating those master’s degree supplements. They are worth, on average, about $5,000 per teacher, so that amounts to about $11.25 billion per year.

Boom. We just saved more than $78 billion, roughly 9 percent of the total cost of our education system. We could get there via attrition, and it would barely cause any pain.

Perhaps we could call it LOLCATS: Local Officials Lead Cuts via Attrition to Teachers and Staff. (Yes, I spent a lot of time on that!)

Now, as my friends on the right will tell you, I’m a big squish, so I don’t necessarily think we should take all those savings and just, say, give them back to taxpayers. I’d prefer to invest some of them in higher salaries for teachers, especially great teachers willing to serve in high-need schools, like Washington, D.C., and Houston are doing. To attract the best teachers to the toughest schools, the salary supplements need to be at least $10,000–$15,000 a year.

Let’s say we gave every single one of our 2.25 million remaining teachers a $10,000 raise, and another $15,000 raise on top of that for 1 million teachers in high-poverty schools. That comes to $37.5 billion—just about half of the amount we saved by modestly raising class sizes, reducing non-teaching staff, and eliminating the master’s degree pay bump.

What’s tough, of course, is getting from here to there. Given union politics, few districts will make these changes without state policy mandating them to do so. Perhaps the place to start is by forcing such changes on chronically low-performing districts—but I’m certainly open to other ideas.

—

Between staff cuts and grant and contract terminations, DOGE appears to have saved less than $2 billion in K–12 education spending so far. With my back-of-the-envelope exercise, I saved north of $78 billion (at least in theory). The main point is simple: In primary-secondary education at least, Washington’s not where the waste is.

For generations, the traditional public school has been a bedrock institution in American life. But recent developments suggest that this model is undergoing a stress test like never before, exacerbated by the recent cratering of student achievement. The pandemic accelerated two key trends: a broad societal shift in literacy habits and a structural transformation in K–12 education through the rise of private school choice. Taken together, these two forces raise profound questions for policymakers about the future direction of American education and its role in shaping civic life.

First among these is the uncomfortable truth that we may be headed toward a “post-literate” society in which deep literacy is no longer a given. States have been working in earnest to counter this shift through implementation of the science of reading in the early grades, but it’s an uphill battle: Reading rates among both children and adults continue to drop, while online consumption—particularly short-form video—has exploded. As technology reshapes how students consume and interact with information, schools appear increasingly outmatched by the rapid pace of change in the digital world.

What could this mean for America’s students in the long run? Twenty years ago, The New Yorker published an article titled Twilight of the Books, offering insight into what a post-literate landscape might look like:

In an oral culture, cliché and stereotype are valued, as accumulations of wisdom, and analysis is frowned upon… There’s no such concept as plagiarism, and redundancy is an asset that helps an audience follow a complex argument. Opponents in struggle are more memorable than calm and abstract investigations, so bards revel in name-calling and in “enthusiastic description of physical violence.” Since there’s no way to erase a mistake invisibly, as one may in writing, speakers tend not to correct themselves at all. Words have their present meanings but no older ones, and if the past seems to tell a story with values different from current ones, it is either forgotten or silently adjusted. 

Amid today’s coarse politics and unchecked presentism, this description feels eerily familiar. It also underscores the stakes in how schools navigate the roles of technology and artificial intelligence in the classroom.

At the same time, the stunning speed at which states have expanded options beyond geographically-assigned, district-run public schools carries significant implications that have yet to be fully appreciated. Currently, nearly 40 percent of American schoolchildren are enrolled in alternatives to their neighborhood schools, and that figure is poised to surpass 50 percent as Texas moves closer to providing Lone Star families with the means to opt out of traditional public education altogether. While expanding private school choice could help address the aforementioned decline in literacy, it will also force a reckoning in how we ensure access to educational excellence in an increasingly diverse and decentralized environment.

Empowering parents with greater authority in selecting the best avenue for their children’s education is inherently attractive, but it brings tradeoffs. School choice promises a more personalized experience but also pulls on the fundamental tension between liberty and community—and it could widen the gap between haves and have-nots. Much like the shift from network TV to streaming, unfettered school choice risks balkanizing the educational ecosystem and eroding what little social cohesion remains.

More importantly, today’s fervor around private school choice as an end in itself obscures the most critical consideration: academic outcomes. More options do not necessarily translate to higher achievement. Complicating things further is the divide among choice advocates on accountability. Some argue that publicly subsidized private schools should face the same (or similar) testing requirements as public schools, while others see such mandates as government overreach. What’s more, there’s an argument to be made that America’s decentralized approach to education governance is at odds with a nearly unrestricted school choice system. Robert Pondiscio alluded to this issue in a recent essay called The Last Days of Public School:

Most countries fund education pluralistically, supporting a mix of public and private schools adhering to a national curriculum and standards. The US, by contrast, is shifting toward a system where public funds follow individual choices with few common guardrails—the US has neither national standards or a common curriculum.

The “few common guardrails” Pondiscio describes have largely been set by states, not the federal government, leading to significant variation in accountability measures like testing requirements and fiscal oversight. A push toward a system with even fewer constraints would likely widen these gaps. Yet, there has been little serious discussion about these potential—and not insignificant—downsides.

These concerns come from the perspective of an unapologetic champion of the science of reading, as well as a staunch supporter for school choice in all its forms. But being pro-choice does not mean overlooking the complexities of expanding educational options, just as being pro-science of reading requires acknowledging the uphill battle in reversing the stark declines in literacy rates.

The degradation of deep literacy and the rise of private-school choice are unfolding largely outside the spotlight of today’s most heated education debates. While states cannot reverse the digital shift, they must prioritize foundational literacy—especially early reading—across all schools, with clear measures to track progress. Similarly, as private school choice expands, it must be accompanied by rigorous academic standards and accountability provisions to ensure that all educational options meet high expectations for student achievement. Moreover, these reforms should focus on underserved communities to prevent deepening educational inequalities. The decisions made today will shape the trajectory of American education and determine how effectively schools serve future generations.

Editor’s note: This was first published on the author’s Substack, Governing Right.

Shrinking, or even closing, the U.S. Department of Education (ED) would certainly shift power out of Washington. That would be a good thing.

But we shouldn’t assume that an invisible hand will then suddenly materialize and orchestrate all the fixes American education needs. Reforms of systems are important, but reforms of substance don’t always follow.

There’s an historical fact to bear in mind: American schooling was far from perfect prior to the 1979 creation of ED. There were numerous reports and indicators showing that things were badly off track in the 1970s. The panel that penned the famous “A Nation At Risk” report was commissioned in 1981; they were responding to conditions decades in the making, not caused by a federal agency approved eighteen months earlier. ED didn’t singlehandedly cause our education struggles; ending ED won’t end them.

I believe we don’t need much federal activity on education. But I do believe we need national energy.

What that means in practice is a smaller educational footprint for a Washington-based agency but a strong national voice and commitment to students and the institutions that serve them. America should be galvanized to make its schools the best in the world. We should know what the problems are. We should mobilize our state and local leaders to tackle them.

That kind of national leadership, however, requires great knowledge of and experience with education policy and education institutions. It also takes a track record of consensus-building.

President Trump should ask Lamar Alexander to convene the nation’s governors to make school improvement a domestic-policy priority over the next four years.

This would be good for America’s schools and students. It would be good for our politics. It would also help Alexander finish the job he started with the Every Student Succeeds Act—a different decentralization movement a decade ago.

Alexander the Great Reformer

It’s hard to overstate Lamar Alexander’s education-leadership credentials. He was a two-term governor who prioritized school reform. He was the president of his state’s flagship university. He was U.S. Secretary of Education. Then he was a U.S. Senator who gave up a major party leadership position to chair the chamber’s education committee—the perch from which he ultimately led the passage of ESSA, the statute that stopped Uncle Sam from meddling so much in America’s schools.

Alexander knows about limiting federal power, but as a former state official, he also knows that after limiting federal power, you have to do the difficult state, local, and institutional work of making schools better.

ESSA wouldn’t have happened without Alexander. The law scaled back elements of the No Child Left Behind Act while preserving its important features related to content standards and assessments. The law also rapped the knuckles of the Obama administration, especially former secretary Arne Duncan, for its overreach (NCLB waivers, Race to the Top, etc.). Federal education policy is better because of ESSA.

But no one who follows schools closely would argue that American public K–12 education is dramatically better today than it was prior to ESSA. Test scores have fallen to appallingly low levels. Chronic absenteeism is a tragedy. Schools have been engulfed in culture-war melees. We’ve not come to grips with the failure of Covid-era practices and policies. The lowest performing students have fallen even farther behind. Public frustration is high and growing, evidenced by the rapid expansion of choice programs.

And I’ve not even mentioned higher education.

My point is that decentralization alone is not enough. We need America’s governors, legislators, state superintendents, state school boards, college presidents, local board members, and local superintendents to step up. They need to use the power they have to improve schooling.

Alexander has the gravitas, skill, and knowledge to catalyze that movement.

The reform consensus of this decade

Every decade or so, a new education consensus emerges and helps improve American schooling. In the 1950s, Sputnik and other Cold War traumas encouraged new investments in math and science. In the 1960s, the federal government passed the massive Elementary and Secondary Education Act to address the needs of low-income students. In the 1970s, communities fought back against court-ordered busing and the deterioration of student learning and school culture. In the 1980s, “A Nation At Risk” spurred the accountability movement. In the early 1990s, vouchers and charters were created to expand differentiation and choice. In 2002, President Bush signed NCLB into law to push schools and districts to improve. In the mid 2010s, ESSA got Uncle Sam to back up.

We need a new consensus for the 2020s, and it needs to be straightforward: America’s schools must return to focusing on student learning.

In 1989, President George H.W. Bush convened the nation’s governors at a summit in Charlottesville, Virginia, to energize school reform. It worked. A bipartisan group of leaders from across the nation set about toughening standards, improving curricula, developing tests, crafting accountability measures, piloting choice initiatives, and much more. It is possible to create a bipartisan push for school improvement.

This is never easy work. It will be especially tough today. It will require a blend of individual leadership, state and local policy, institutional reform, social entrepreneurialism, culture change, and more.

No one person can do this alone. But it sure would help if there was a concerted national effort to get it started. And no one is better suited to it than Lamar Alexander. I hope he gets the call.

On this week’s Education Gadfly Show podcast, Steven Wilson, senior fellow at the Pioneer Institute, joins Mike to discuss his new book The Lost Decade: Returning to the Fight for Better Schools in America, which argues that the push for so-called antiracist education derailed reform and harmed marginalized students. Then, on the Research Minute, Amber examines a study on how educators divide their attention during virtual tutoring—and how achievement, gender, race, and English learner status influence those interactions.

Recommended content: 

• Steven Wilson, “The Lost Decade: Returning to the Fight for Better Schools in America,” Pioneer Institute for Public Policy Research, 2024.
• Wilson, Steven F. “The Promise of Intellectual Joy,” June 4, 2019. 
• Robert Pondiscio, “After a “lost decade,” let’s restore high expectations for students,” Thomas B. Fordham Institute (December 12, 2024).
• Michael J. Petrilli, “11 thoughts about the massive layoffs at the U.S. Department of Education,” Thomas B. Fordham Institute (March 11, 2025).
• Frederick M. Hess, “Defunding the teacher trainers,” Thomas B. Fordham Institute (March 6, 2025).
• Qingyang Zhang, Rose E. Wang, Ana T. Ribeiro, Dorottya Demszky, and Susanna Loe, Educator Attention: How computational tools can systematically identify the distribution of a key resource for students, Annenberg Institute (March 2025)

Feedback Welcome: Have ideas for improving our podcast? Send them to Stephanie Distler at [email protected].

Cheers

• The Center for Assessment breaks down the four major reasons that annual statewide tests are important and necessary. —The Center for Assessment
• Secretary of Education Linda McMahon makes her first school visit, to a charter in the Bronx, emphasizing the importance of school choice. —Chalkbeat
• Texas lawmakers and charter school leaders are pushing back against the $870,000 salary for the Valere Schools superintendent, citing it as excessive. —The Texas Tribune
• The Trump administration’s streamlining of the Department of Education laudably aims to cut bureaucracy and refocus federal efforts, but desperately needs more transparency. —Education Next
• London’s Michaela Community School is a highly disciplined and academically rigorous free school that deserves admiration for the exceptional achievement of its disadvantaged students s. —Thomas Chatterton Williams, The Atlantic

Jeers

• The Trump administration’s mass layoffs at the Department of Education have gutted its statistics and research division, leaving only three staffers to oversee national education data, raising concerns about security, oversight, and the future of the Nation’s Report Card. —The Hechinger Report
• AI-powered cheating is surging in schools, with students using tools like ChatGPT and Gemini to complete assignments. —The Wall Street Journal

• Reflecting on his own Hasidic education, this author argues that, while cultural preservation is important, public officials should ensure that all children gain basic skills to ensure that they have the freedom to choose their future. —Zalman Rothschild, The Washington Post
• Sociogenomics challenges the “nature vs. nurture” debate by revealing that genes and environment constantly interact in a feedback loop, influencing educational and other life outcomes while also raising concerns about how genetic screening could exacerbate inequality. —Dalton Conley, The New York Times
• A great look at the friendship that likely explains Justice Barrett’s recusal from the big religious charter schools case. —The 74
• Concerns about declining cognitive abilities due to passive digital consumption are growing—but society may adapt through cognitive specialization, AI-driven skill enhancement, and potential future advancements in genetic intelligence. —James Pethokoukis, Faster, Please!