Defining Educational Expenditure

A recent conversation with fellow supporters of educational vouchers raised a question that I had not adequately considered: What counts as an educational expenditure? Since one of the benefits of a market is the opportunity to explore new and different ways of producing things, voucher money should be available to be spent on educating children whatever the mechanism not limited to the present model of schooling. That could include conventional private schools, home schooling, microschools, learning pods, online education, correspondence courses, and other alternatives that have not occurred to me. That raises the problem of how to define an educational expense in order to decide whether it qualifies to have voucher money spent on it. Consider some relevant concerns:

A: Since the money is coming from the taxpayers for education it should be spent on education.

B. Parents know their child better than the public-school authorities and are more likely to want to act in his interest, less likely to sacrifice it for the interest of teachers or school administrators,  so their choices should have priority.

C. The structure of the public schools, large government run organizations, pushes them towards standardized ways of teaching, away from flexibility. That is especially a problem in a time of rapid technological change, when both what should be taught and how may be changing in not yet known ways.

I see, broadly speaking, three alternative approaches:

1. Educational expenditure is anything parents say is educational

That makes the voucher a payment to the parents for not sending their child to a public school, with no control over how the money is spent. It works in a society, or parts of society, where parents can be trusted to spend money on education if they say they do.

2. Educational expenditure is an expenditure on something classified as educational — a school, a tutor, textbooks, not a new car or groceries, probably not a trip to Disneyland.

This requires someone other than the parents to decide what qualifies. It reduces the flexibility of the system. The definition of what is educational risks being either too restrictive, limited to whatever forms of education are currently common and widely accepted — experiential education (for an example see Come to Oars by Muriel Curtis) can be very educational but probably won't qualify since it is not teaching academic material — or barely restrictive at all if whoever is approving expenditures can't be bothered to check what they were for.

3. Educational expenditure is expenditure that results in education as measured by some objective test.

3a. Applied to institutions. A private school counts if the average score of its graduates is higher than (say) the 25th percentile of public-school graduates; any kid going to such a school can use the voucher to pay for it.¹

3b. Applied to kids. You can collect the voucher if, at the end of the school year, your kid scores above the 25th percentile of public-school students of his age.

All of these alternatives have problems.

The solution most voucher supporters imagine is private schools competing with public schools, with the private schools possibly being regulated by the government to make sure they are educating kids, something along the lines of 3a above. This is the simplest approach, other than alternative 1, but:

It limits education to something like a school. That eliminates applying the voucher money to home schooling. It may eliminate microschools. It eliminates an educational strategy that combines two or more sources of education — apprenticing with a carpenter while taking some classes at the local community college and learning other things online and a few with a local college student as tutor.

Whoever controls the test controls what kids get taught, what the subjects are, how much of each gets taught, and what the right answers are. If the test is minimal — can you read a newspaper, write a note to someone, add, subtract, multiply and divide² — it can be satisfied, for most kids, with a very minimal education. 

Suppose you want to test more than that, make sure the school is doing as good a job as the public schools it competes with. One problem with constructing a test of whether an 18-year-old has gotten a reasonable education is that people vary in ability, interests and intended profession; there is practically nothing beyond the minimal, base competence in the traditional reading, writing and arithmetic  that it is important for everyone to know. Many, probably a majority, get on fine with no math beyond arithmetic, many with no reading of serious literature, no ability to write an essay. Do you give a low test score to the math genius who neither knows nor cares when the Civil War happened, what it was about, or who won? Do you fail the gifted writer whose mathematical ability is limited to being able to add up a column of numbers and usually get the right answer? Testing a moderate level of all skills that public-schools teach is a recipe for a cookie-cutter education, spending most of the time teaching the same thing to everyone, including those who already know it and those who never learn it. Part of my opposition to this sort of approach comes from the fact that I don't think either the public schools or most private schools are well designed to produce educated people competent in thinking for themselves. I discussed my reasons and the educational model I prefer in an earlier  post.

Another problem is that control over what the right answers are, applied to education beyond the basic, gives the people creating the test the ability to have everyone taught whatever they believe. Think about climate change, gun control, explanations for different outcomes by race or gender, anything reasonable people disagree about. Teach the approved answer or your school doesn’t qualify for vouchers. Many of the people on one side of such an issue believe that there are no reasonable people on the other side and so would include agreement with their position in their definition of education.

The individual version of alternative 3, 3b, has additional problems due to the wide range of families and kids. A test that at least a quarter of public-school students can pass will be trivial for bright kids with educated parents, leaving them free to spend most of the voucher on whatever they want. Does the parent have to spend the extra five thousand on educational expenses along the lines of approach 2 or can he pocket the money, perhaps accumulate a few years of it to buy the kid a car when he gets old enough to drive? Sufficiently unintelligent kids on the other hand, especially ones with behavior problems, will never be able to pass the test so get no voucher, although they can still go to the public school.

In practice, the closest you can probably come to satisfying concern A, making sure the money is spent on education, is a mix of approaches 2 and 3, requiring private schools to produce graduates at least as well educated by some simple objective measure as the graduates of the less successful public schools and requiring money not spent on a school to be spent on educational expenses, broadly defined.

The alternative is to give up on concern A and treat the voucher as a payment to parents for not sending their child to a public school, trusting them with control over how the money is spent. That is the best solution for responsible parents who care about their kids, probably the great majority, but leaves sufficiently irresponsible parents free to neglect the education of their children and spend the money on themselves.

---

Past posts, sorted by topic

A search bar for past posts and much of my other writing

1. The advantage of this sort of criterion is that it makes it difficult for opponents of vouchers to block their use by setting too high a standard. One could apply that approach in the opposite direction by cutting the funding of any public school that fails to match the criterion.

2. The traditional Three R’s: Reading, wRighting and ‘Rithmatic.