For a nice overview of the counterproductive incoherence of Republican education efforts over the decades, check out this piece by Frederick Hess and Andrew Kelly of the American Enterprise Institute. And for a sense of how confused conservatives remain when they write pieces telling other conservatives how to have “good” federal education policy, read the same piece. Its history section gives you the first part, and, unfortunately, its other sections give you the rest.

Hess and Kelly — who are generally pretty sharp — furnish a terrific overview of what happens when you talk “local control” of government schools and decry federal micromanaging, but can’t stop yourself from spending federal money and love “standards and accountability.” Basically, you get a great big refuse heap of squandered money, red tape, educational stagnation, and political failure.

Having laid all that out pretty nicely, you’d think that Hess and Kelly would reach the logical conclusion: Conservatives should obey the Constitution and get Washington out of education. But they don’t. Instead they give precious little thought to the Constitution, and make policy prescriptions that fundamentally ignore that government tends to work for the people we’d have it control. You know, concentrated benefits, diffuse costs; iron triangles — basically, the big problems Hess and Kelly decry at state and local levels.

Start with their constitutional argument (such as it is):

The federal government does have a legitimate role to play in schooling — and it always has. From the Land Ordinance of 1785, which set aside land for the purpose of building and funding schools, through Dwight Eisenhower’s 1958 investment in math and science instruction after the launch of Sputnik, the federal government has recognized a compelling national interest in the quality of American education.

Wow! What a sweep of history! What a great many years this covers!

The thing is, most of those years see essentially no federal education activity, and the first year mentioned — 1785 — precedes the Constitution. Why very little activity between 1785 and 1958? Because relatively few people thought the national government had any role to play in governing education. That’s why neither the word “education” nor “school” (or, for that matter, “compelling national interest”) is in the Constitution, and even a commission created by Franklin Delano Roosevelt said that there is no constitutional federal role in education. Washington does have jurisdiction over District of Columbia schools, and a responsibility under the 14th Amendment to prohibit discrimination, but that’s it.

OK, to be fair, it could also be argued that Congress can constitutionally pass education provisions like those in the Land Ordinance. But not because Washington has power over education. No, because as you see in Article IV, Section 3 of the Constitution, it has power over territories. Of course, states and districts are not territories, and the 10th Amendment reiterates what is clear from the granting of only specific, enumerated powers to Congress:

The powers not delegated to the United States by the Constitution, nor prohibited by it to the States, are reserved to the States respectively, or to the people.

In an effort to deal with the very clear failures of federal education policy — but without the clarity of following the Constitution — Hess and Kelly offer three things they think the Feds can and should focus on: transparency, research, and “trust-busting.” However, all three ignore the fundamental political reality that school systems tend to be controlled by their employees because the employees have the most at stake. And what are their incentives? Same as mine and yours: to get compensated as generously as possible and have no one hold them accountable for their performance. The result, of course, has been oodles of money spent without meaningful academic improvement.

Start with transparency. Lauding No Child Left Behind for having improved “citizens’ ability to gauge and compare school quality,” Hess and Kelly argue that the Feds “should require that states collecting federal school funds measure and report detailed data on school quality and educational costs in a consistent, uniform way.” They caution, however, that Washington shouldn’t prescribe standards or curricula, but should measure “schools’ return on public investment.”

Talk about conflicted! How exactly do Hess and Kelly expect the Feds to both stop short of mandating curriculum and standards, and provide an accepted measure of specific schools’ return on investment? Even if you could thread the needle for a while, it is very hard to imagine Washington not eventually narrowing acceptable measures down to a single curriculum and test so that results could be uniform and distilled into soundbites. And what would likely happen even if the standards started off rigorous and the testing tough? The employees who would ultimately be held accountable would simply move their dumbing-down pressure from state and local levels to the federal level, where policy was now being made. Nothing would be solved, and there would be huge added problems of a new, monolithic standard that could in no way effectively serve greatly diverse kids, as well as the quashing of competing curricular ideas.

Next we’ve got the research argument, which is predicated on the well-known contention that the incentives for private-sector investment in “basic” scientific research — which doesn’t offer immediate returns — are too weak to provide for optimal amounts. The Feds, therefore, have to step in with oodles of grant money. Hess and Kelly would like to extend that model to education research.

Education, however, is not particle physics, the kind of research we generally imagine as “basic.” The need for major equipment investment is not nearly as great for education, nor are the likely benefits of, say, studying the effects of flash cards as far distant as the industrial applications of string theory. Moreover, if we had broad school choice, with schools able to seek profit without penalty, educators would have every reason to invest in research and find better, more efficient ways to teach kids. Finally, there is good evidence that science funding is just as likely to translate into rent-seeking benefits to the scientists as scientific benefits to the public.

Oh, and the constitutional basis for education research spending? Hess and Kelly don’t even try to offer one.

Finally, we have “trust-busting.” Here even Hess and Kelly’s examples illustrate how mistaken their ideas are. While sensibly calling for a reduction in calcifying federal rules and regulations, Hess and Kelly also argue that the Feds should be able to set up “private” entities to compete with the public-school monopoly. They cite as an example the American Board for Certification of Teacher Excellence, which provides a teacher preparation and certification process separate from those established by states. They also note that ABCTE has been “generally neglected.” Which is probably a good thing. After all, think of two other “private” federal creations: Fannie Mae and Freddie Mac. Seems there can be big problems when Washington creates supposedly private entities.

Then there’s this: “A related role for federal lawmakers is to help lift the burden of bad past decisions and troubling policy legacies that hinder reform-minded state and local leaders.” Hess and Kelly argue that Washington should create a form of bankruptcy that lets states and districts render null and void labor contracts and other obligations that make it hard for them to do business. And what example do they use of organizations that have been crippled by “legacy” contracts? General Motors and Chrysler.

Of course, thanks to the federal government, GM and Chrysler didn’t go through normal bankruptcy, did they? No, they went through processes jury-rigged to favor politically important special interests. That lesson should be screaming at Hess and Kelly: Give Washington power over something and they won’t use it for the common good. They will use it to reward the politically powerful, the very state and local problem Hess and Kelly are trying to solve!

The simple reality is that the federal government is no less subject to special-interest control — the ultimate result of the basic political problem of concentrated benefits and diffuse costs — than state and local governments. Except, that is, that Washington is even more distant from the people than state and local governments, and if people don’t like their state or local schools they can at least move.

There is, really, only one solution to the basic government problem of concentrated benefits and diffuse costs, and we do no one any favors by denying it: We need to end government control of education. We need a free market, in which educators are free to teach as they see fit and are held accountable by having to earn the business of paying customers.

The federal role in getting to this, thankfully, is simpler than what must be done at state levels, where constitutional authority over education actually exists. All that Washington has to do is obey the Constitution and get out of education. And yes, Rick and Andrew, that is what the Constitution requires.