The Washington PostDemocracy Dies in Darkness

‘Neovouchers’: A primer on private school tax credits

By Kevin Welner

In a March 1ststatement on the Hoover Institution’s “Education Next” blog, Jason Bedrick of the Cato Institute offers his conclusion that policies he calls “scholarship tax credit (STC) programs” primarily benefit lower-income families. Because so little evidence exists, Mr. Bedrick avoids making many specific claims about the actual distribution of benefits. But he does make some broad claims that I think are worth engaging.

As a policy matter, there are many issues worth considering. There are, for instance, the basic questions applicable to all voucher and voucher-like policies concerning outcomes for students offered vouchers, competition effects, constitutional establishment clause (separate of church and state) issues, philosophical issues about liberty and choice, and societal issues about Balkanization and the role of public schooling in a democracy.

There is no question that these programs, which I have called “neovouchers” (for reasons I explain below), do indeed provide financial assistance to many lower-income families. This is particularly true in states like Florida that means-test the recipients and have safeguards in place to prevent donor self-enrichment. It appears to be much less true in states like Arizona and Georgia that have designed their programs with a much stronger free-market emphasis. In those states, the programs also provide a great deal of assistance to upper-income families. And attempts to amend those laws to better focus on those with greater needs have been thwarted.

Efforts have also been thwarted that were designed to add some transparency to the processes and results of these policies. Most outrageously, following some investigative reporting by the Atlanta Journal Constitution, the Georgia legislature made it a “crime for state officials to release key information about the program.” This is not a healthy response to unwanted news, and it prompted the Society of Professional Journalists to award the Georgia legislature a 2012 Black Hole Award. As the Award announcement pointed out, “Georgia’s law fails to hold anyone accountable for how they divert or spend tax funds. It does not track who is receiving scholarships under the program.”

Among the original findings from the newspaper investigation was the following:

The purpose of the program is to give public school kids a chance to attend private schools by granting them scholarships. But some SSOs encourage families to game the system: Although their children are already in private schools, they enroll them in public schools for the sole purpose of making them eligible for the scholarships. The children never actually attend the public school. The AJC found 81 such cases in one county alone.

Similarly, the School Tuition Organizations supporting Arizona private schools have allowed or even encouraged parents to subvert a provision in the law that is supposed to prevent parent taxpayers from directing a donation to support their own children’s tuition. As I discussed in my 2008 book, “NeoVouchers: The Emergence of Tuition Tax Credits for Private Schooling,” parents of children in the schools have entered into agreements, whereby one set of parents designates a classmate as a beneficiary, while that classmate’s parents return the favor for the first set of parents.

In fact, contrary to the assertion in Mr. Bedrick’s post (“STC Donors Do Not Benefit Financially”), some of these School Tuition Organizations have counseled parents that they might be able to make a profit on their donations, by taking a federal tax deduction on top of their 100% tax credit from the state. (To see some examples, google the phrase, “A taxpayer may also be able to claim a federal deduction for the donation. The taxpayer should consult their tax advisor for specific tax advice.”) A similar problem apparently exists in Georgia, where state senators in 2009 voted down a proposal aimed at preventing such double-dipping.

Again, there is great variation from state to state. After some highly publicized abuses in Florida, for example, that state modified its law to avoid some of the above-described problems. But in none of the states, Florida included, does the law provide the sort of data and transparency needed to answer the question of how neovoucher students do academically or the question of how many students are prompted by the law to attend private school instead of public school. This second question may not seem important, but it is in fact key, since it is crucial for any determination of the effect of the policy on the public coffers. Different people have made different guesses (see the Southern Education Foundation’s estimation approach, discussed in this New York Times article versus the Florida OPPAGA estimate, which I discussed here.

Neither Mr. Bedrick nor I can know if he’s correct in the statement, “Well-designed STC programs such as those in Arizona, Florida, and Pennsylvania actually save states money by decreasing state expenditures more than they decrease state tax revenue.” I wouldn’t be surprised if Florida’s program does, but even that is based on guesswork. And, in contrast, I’d be very, very surprised if Arizona’s program saves the state money – although again, we all are left to merely guess since the laws don’t provide for the data needed to concretely answer the question.

Mr. Bedrick also states, “Every state supreme court to address the matter has agreed” that “STC programs use private money not public money”. I haven’t been following this issue all that closely in recent years, but I’m only aware of the Arizona case. Is there a second state supreme court that has weighed in on this, let alone enough decisions to qualify for an “every state” statement?

The broader claim from Mr. Bedrick is as follows: “Contrary to [Valerie] Strauss’ assertions, scholarship tax credit programs are not the same as vouchers. They differ greatly in terms of their funding mechanisms and administration.” [She didn’t say they were the same as vouchers in this blog post.] In a conventional voucher system, the taxpayer pays taxes to the state, which then distributes vouchers to parents to use as payment at a chosen private school. The school then ‘cashes in’ the voucher with the state. Parents are free to use the voucher with any private school that elects to participate and meets the state’s straightforward criteria. With a neovoucher system, however, the taxpayer never pays (all or most of) the taxes, which are forgiven through the tax credit when the taxpayer instead pays the money to a private non-profit (often called a “school tuition organization” or STO). In most states, this STO is associated with (and generally created by) a small number of private schools. So the money goes from the taxpayer to the STO to the parent to the school, with the taxpayer then recovering the money from the state.

What the Arizona Supreme Court decided in 1999 was that this money-laundering mechanism is sufficient to cleanse the neovoucher – to bring the neovoucher out from under a state constitutional provision forbidding state spending for these purposes. In the process, the STOs and taxpayer donors in most of these states – and thus the private schools – are given some of the choice that would otherwise lie with the parents. That is, when a taxpayer chooses to donate to the (hypothetical) Arizona Pastafarian School Tuition Organization, which only provides neovouchers for a few associated Pastafarian private schools, a parent wishing to go to a different private school cannot receive the resulting neovoucher for that purpose.

This becomes an issue in a given state or district when the wishes of wealthy taxpayers (those who owe enough state taxes to make STO donations and get the resulting tax credit) are not aligned with the wishes of parents. Setting aside hypothetical Pastafarians, imagine if Arizona’s wealthier taxpayers are more likely to support Protestant-aligned schools while its lower-income parents are more likely to prefer Catholic schools. While those parents would be free to choose those Catholic schools under a conventional voucher plan, they are at the mercy of donor preference in a neovoucher plan. (The exception here is again Florida, where the state’s main STO serves a role similar to the role played by the state in a conventional voucher plan.)

This and a few other differences do make neovouchers different from vouchers, but I would disagree with Mr. Bedrick’s assertion that the two approaches “differ greatly”. The neovoucher approach is more circuitous and convoluted, but the purposes and end results are largely the same. Analogously, the government may assist with home purchases through direct subsidies of mortgage payments or through a mortgage interest deduction, and those are certainly different policies – but let’s not kid ourselves into believing that lessons from one cannot powerfully inform the other.

In past neovoucher discussions, I have found myself “aligned” in some way with Cato’s position. We both agree that these policies have political advantages (less likely to stir opposition and more likely to pass in many states) as well as legal advantages (more likely to survive court scrutiny, as we saw with the Arizona law). I also agree with Cato that neovoucher policies are less likely to come with regulations and transparency. But when we get to the policy benefits of all this, I have long disagreed with Cato. I see an unregulated, non-transparent system designed to be confusing and to undermine constitutional restrictions. Cato sees freedom.

Potato, potahto…