For immediate release. Contact: press@rfer.org

I left my position at Brightbeam at the beginning of 2026, and since then I have done what people in this field rarely allow themselves to do, which is to stop, read the polling, take meetings, and ask honestly whether my twenty years of advocacy accomplished what I wanted.

The honest assessment was that it had not. That admission fueled my next move, which I know will surprise those who know me well. Today I am announcing the launch of Republicans for Education Reform.

Our friends and colleagues pronounce it “reefer,” which we have decided to accept with good humor.

Why the name

I have never been a Republican or Democrat, a fact that surprises people who assumed my registration from my zip code. I’ve been conservative on some issues, but it was generally assumed I aligned more with the Dems. I've believed the supposed party of working families would eventually notice that working families have children, those children attend schools, and those schools are not working.

Sadly, the party stubbornly pursued lesser priorities. Minimum wage increases. Access to a doctor. Drinking water without “forever chemicals.” Air without particulates. The child tax credit. Prenatal care and infant nutrition. Renewable energy. Loan forgiveness for the wrong graduate programs. Twenty years of moving money from people who earned it to people who need it, and not one serious national conversation about portability.

I stayed as long as I could. At some point, a person has to admit the coalition left him standing on a corner it no longer drives past.

Our cousins at Democrats for Education Reform (DFER) have spent fifteen years being accused of being Republicans in disguise. I watched them absorb that accusation, correct the record, produce their voting histories, and absorb it again. I have come to believe they should have embraced the charge as a point of pride, not as an insult. RFER aims to remove the ambiguity. We think there is dignity in accurate labeling, and we would encourage others in our sector to consider whether the D in their acronym is load-bearing or decorative.

As a leader, I want to pave the road for other “D” reformers to end the ruse and join the “R.”

Our position

We see education savings accounts as lifesaving. They are the only remaining mechanism capable of addressing declining wages, the cost of college, the price of housing, the price of groceries, and the erosion of opportunity for American families.

I want to be precise about that claim, because our critics will say it is broad. It is broad. When a family receives roughly seven thousand dollars in a state-administered account restricted to approved educational expenditures, that family gains something the minimum wage never gave them, which is the experience of choosing. Wage floors are set by legislatures. Conversely, parents direct savings accounts. One of those is freedom, and the other is administration, and after two decades in this work I know which one I would rather explain to a mother in Memphis.

Some economists have raised questions about whether an education account can raise a household’s income. We consider this a category error. The account raises the household’s agency, and agency is upstream of everything, including income, including health outcomes, including the durability of a marriage. We have commissioned research on this and expect results by the fall.

What the data show

Education conservatives have produced a body of polling over the last three years that Democratic strategists have simply refused to read.

The findings are consistent. Minorities support school choice at high rates. So do more important voters. Parents want options. Voters under forty want options. In several battleground states the numbers approach two thirds, and while some methodologists have raised concerns about question wording, sample construction, weighting, sponsorship disclosure and the absence of a control condition, the direction of the finding has not changed across any survey our funders have paid for.

The political conclusion follows. Any Democrat who does not support education savings accounts will lose. We do not have an election result demonstrating this yet, but the polling is very clear, and it was also very clear in the cycles where it did not happen, which suggests a timing problem rather than a theory problem.

A word to Democratic governors

Dem leaders need to get with the program, quickly. Federal money is available now, but the window will close. States that opt into the new tax credit scholarship program will receive it. States that decline will not.

I have heard all the objections about the program being untested. It has no meaningful reporting requirements. No achievement data is attached to it, no auditing regime, no provision for students who are counseled out, no obligation for participating providers to accept students with disabilities, and no mechanism for a state to recover funds from a provider that closes in November.

Set all that aside. You have priorities you cannot fund and this is free money on the table. Take it.

The governors resisting this are virtue signaling an integrity that costs their own constituents real dollars, and they are doing it because of who signed the bill. They hate Trump more than they love money. I understand the impulse. I have felt it. I have also concluded that a state that refuses seven thousand dollars per child to make a statement about an administration has confused its own moral comfort with its children’s interests, and I say that as someone who argued that this exact confusion was the problem with the education establishment.

On the other conversations

I am aware there are other things happening in the country that take the Democrats’ attention from ESAs.

Several people I respect have asked me if RFER will issue a statement on the rollback of civil rights enforcement, or on the graduate lending changes that will price a generation of women out of the credential that gets them into professions, or on the armed federal agents who arrive in unmarked vehicles and leave with somebody’s father, or on the fact that we are funding two theaters of conflict while the exchanges collapse, or on the men currently building a future in which most people are either monitored or unnecessary.

The simple answer is that it isn’t a good time to focus on those things. The world has moved on from those issues.

Our position is that these are adjacent conversations. They poll poorly. They crowd the education message. Every hour a Democratic communications staffer spends on masked agents in Chicago is an hour not spent explaining ESAs to a suburban mother in Bucks County, and there are only so many hours before the midterms.

I have instructed our staff not to comment on things unlikely to change. This is discipline, and discipline is what our side of this fight has always lacked.

Besides, the families most affected by all of it will benefit most from an education savings account. We say this frequently, and no one has yet explained to us why it is wrong.

The socialism question

The rise of the left flank inside the Democratic Party has produced a lot of hand-wringing among people who should know better.

We read it differently. Voters just demonstrated an appetite for candidates who promise to take something away from the powerful and hand it directly to ordinary people. That is the exact emotional structure of an education savings account. The money leaves an institution and lands in a household. Democrats who are frightened of their own socialists are frightened of the only constituency in their party that already understands our product.

Funding and governance

Our goal is not to be saddled by the pretense of Democrat affiliation, but to open the door for Democrats to support Republicans.

To that end, RFER will be supported by a small number of family foundations, two donor-advised funds, a payments fortune, a logistics fortune, and one gentleman who has asked not to be named and whose interest in K–12 policy began in 2023. Our board is being assembled. Several seats remain.

I want to be transparent about what we won’t do here. We will not disclose grant amounts. We will not disclose the model legislation that has been written for us. We will not conduct a study of academic outcomes in states with mature account programs, because those studies already exist, and reading them has not changed anyone’s position, including mine.

Joining

If you have spent the last two years quietly deciding that the coalition you joined is not the coalition you are in, we would like to hear from you. You do not have to announce anything. Most of your colleagues have not.

Come work with us as we work for Republicans.

As I predicted, the quality-blind, data-free school choice sprint has borne bad fruit.

The Tennessean is reporting just seventeen percent of students using Tennessee’s Education Savings Account vouchers met or exceeded the state math standard this year. Twenty-four percent met it in English. Among students in Tennessee public schools, the figures were 40 percent and 42 percent.

This is the fourth year of the program and the fourth year voucher students have trailed the peers they left behind in “failing” public schools. Their scores fell from last year in both subjects. Math proficiency among voucher students dropped in every participating county while the statewide average kept climbing. In Hamilton County, voucher students have lost seven percentage points in English over three years, sliding from 32.2 percent proficient to 25.2 percent, while the public schools there have inched up to 40.3 percent. Shelby County was the one place where voucher students had outperformed the local district, and this year that edge began to close.

It is also the last year anyone will be able to make the comparison at all. Tennessee lawmakers voted this spring to end the TCAP requirement for voucher students. Participating private schools can now choose any nationally normed exam they prefer, from a market of tests that vary in rigor and in price.

The Tennessee House approved the ESA program in 2019 by one vote. Some of the Republicans who supplied that margin agreed to it on the condition that participating students sit for the state exam, so the legislature could see what it had bought. That condition is now gone. One of the senators who helped remove it called the original testing requirement a mistake. Accountability should never get in the way of ideology.

I spent years making the case for school choice. I made it on stages, in print, in rooms where the money was, and to Black and brown parents in cities where the assigned schools had been failing their children for three generations. I won’t disown that work because there is no form of liberation predicated on a lack of options. I would make some of the same arguments again tomorrow.

But I need my friends to be clear on why I argued so forcefully for school choice. Before I was a choice advocate I was a student achievement hawk, and that is the part that will never change. The only question I have ever cared to ask about an education policy is whether it produces the effect it promises. How are the children? Can they read? Do they finish algebra with a deep understanding? Do they arrive at eighteen able to support themselves, enroll somewhere, train for something, vote with comprehension, calculate loan interest, read a lease before signing it, start a business? Choice mattered to me as a means to that end, period, hard stop, end of story. It earned my support because my fellow advocates swore student achievement was the point.

Until it wasn’t. Until data returned showing that test scores were more of a witness against school choice than an accomplice.

Something in me changed. My rhetoric is different on purpose. Clearer. I used to speak almost entirely in the language of parental power and parental rights. That language is useful and often true. It is also incomplete, because the child in the arrangement is a person with interests of their own, and those interests are not always identical to what the adults around them want.

Our own law understands this. The concept of educational neglect is real, and it rests on a plain idea, that adults can deprive a child of capacities so basic that the deprivation follows the child through an entire adult life. A parent who keeps a child from learning the alphabet, for example, has not made a lifestyle choice. They have created an adult who cannot read a job application, a ballot, a prescription label, or a court summons, and who will spend decades paying for a decision they had no part in making. We do not let parents waive a child’s right to that floor, and we should not let a state program waive it either by declining to check whether the floor is there.

Tennessee’s education commissioner reports that 99 percent of ESA families are satisfied with the program. I believe that number. Parents have described getting a child out of a bullying situation, or into a school that finally returned their calls. Those things are real and they matter to a family in ways a proficiency table will never capture. But I agree with State Sen. Raumesh Akbari who said parent satisfaction doesn’t get a child into college or help them get a job. A satisfied parent and an undereducated seventeen-year-old can occupy the same house. The child is the one who carries the outcome forward for sixty years.

The state’s top education leader says we should focus on a real bright spot that might get missed. The share of ESA students scoring in the lowest category has dropped substantially, from 51.6 percent below expectations in math in the program’s first year to 37 percent by 2024-25. Okay, that’s movement and it’s not nothing. Still, movement out of academic hell into a statistical purgatory the state calls “approaching expectations” is not what reformers sold through the years. Tennessee’s own third grade retention law treats “approaching” as insufficient for a public school child, serious enough to hold a nine-year-old back a year. A standard that severe for one group of students and optional for another is a two-tier system with public money on both sides of the line.

I find it hard to write with any detachment. For twenty-five years, the reform movement I belonged to built its entire public case on test data. We used proficiency rates to argue that districts were failing. We used them to justify closing schools, replacing staffs, restructuring governance, evaluating teachers, assigning letter grades to buildings, and telling parents in Memphis and Milwaukee and Newark that the system holding their child was measurably broken. We told the country that the numbers were the truth and that people resisting the numbers were protecting adults at the expense of children. I said versions of that myself. I meant it. I’ll say it again whenever I see it as truth. Even if it’s my friends running from it.

Nowhere in those twenty-five years did we accept the argument we are now making on our own behalf. Not once did we allow that a school’s curriculum was too distinctive to be measured, or that the test was misaligned with what students were really learning, or that satisfaction surveys should substitute for achievement data. When districts said those things, we called it excuse-making and we were often right.

Josh Cowen at Michigan State, who has tracked these programs across states for years, notes that this is now the pattern rather than the exception. Voucher students post weak results, and instead of fixing the results the states retire the instrument. Louisiana produced numbers in 2023 that should have stopped the movement cold, with 14 percent of voucher students hitting state targets against 24 percent of low-income public school students. The response nationally was expansion.

Tennessee’s larger program, the Education Freedom Scholarships, has never required the TCAP at all. It sends about $7,300 per student statewide this year, with no income test on most of the awards and no common measure, and a leading candidate for governor has promised to grow it until every family who wants a voucher gets one. The ESA program, at roughly $10,100 per student, was the smaller, older, more tightly regulated version, and the accountability rules that once distinguished it have now been sanded down to match. The direction of travel is toward more public money and less public knowledge, on purpose.

I know how the expansion happened. It did not happen because state legislators studied the achievement literature and were persuaded. It happened because an enormous amount of concentrated private money moved into statehouse races, primaries, and advocacy campaigns, much of it from donors who are simultaneously funding the dismantling of independent media, independent courts, civil rights enforcement, and the machinery that lets ordinary people hold power to account. Vouchers are one line item in a much larger project of moving public functions into private hands where no one can audit them. Ending student testing requirements comes from the same instinct, applied to a smaller surface.

I am sad about this in a way I have not fully worked out. Some of the people driving it are people I know, people whose children are fine and whose careers are fine and whose certainty has outrun their evidence. I am not going to spend my remaining years arguing with them. They are too invested to hear it right now, and a movement in the middle of a winning streak does not audit itself.

So I am writing this down for whoever comes later, the people who will eventually study how a generation of intelligent, credentialed, well-funded adults spent a quarter century insisting that outcomes were the only thing that mattered and then, at the exact moment their own outcomes came due, voted to stop counting.

The first people who should say this out loud are the honest ones still inside school reform, and I know they exist because I worked beside them. They understand that a political victory built on suppressed evidence is a loan against the movement’s future, and that the bill arrives eventually, with interest, in the form of every reasonable person concluding that the critics were right about us all along.

There is a straightforward way out. Any school taking public money for a child’s education submits that child’s learning to the same measure the public schools submit to, publishes the result, and lets families decide with full information. That was the original bargain. It is still available. Tennessee had it, in writing, as the price of a one-vote majority, and gave it away.

Being wrong is survivable. Arranging things so that no one can tell you are wrong is a different kind of failure, a moral one, and it is the one that costs children.

The mood of Democrats is colored by resistance and urgency, not compliance and concession.

There is an emerging genre of political commentary that has become, in recent years, nearly impossible to distinguish from wishful thinking dressed up as strategic advice. It says Democrats are losing ground with working-class voters; working-class voters care about education; Republicans have seized the education issue; therefore, Democrats must embrace Republican education policies or face irrelevance.

The argument arrives with the confident tone of someone who has done the math, even when the math is largely vibes laced with desperation for philanthropic funding.

This genre is not widespread. It lives mostly in education policy circles, where a cohort of aged centrist Democrats, some of them genuinely well-intentioned, others with more transparent financial and institutional interests, have spent the better part of a decade insisting that the party’s salvation lies in embracing school choice, softening its relationship with teachers’ unions, and generally signaling to voters that Democrats are not, in fact, the party of the teachers’ lounge. The argument has a certain seductive logic. Democrats’ advantage on education in polling has collapsed. Ergo, they must have been doing something wrong. Ergo, whatever Republicans are doing must be closer to right.

This motivated logic works in grant applications, but not in a single room where real Democrats gather.

Read more