AEI’s Rick Hess and Stanford’s Linda Darling-Hammond—two folks who don’t always see eye to eye—have a New York Times op-ed that decries federal micromanagement in education, then lays out four things they think Washington should do.

If only they’d stopped at lamenting micromanagement.

Let’s take their four should-do’s in order:

First is encouraging transparency for school performance and spending. For all its flaws, No Child Left Behind’s main contribution is that it pushed states to measure and report achievement for all students annually….To track achievement, states should be required to link their assessments to the National Assessment of Educational Progress (or to adopt a similar multistate assessment). To shed light on equity and cost-effectiveness, states should be required to report school- and district-level spending…

This sounds great, but the key is in the doing, and there is precious little evidence Washington can force real transparency. NCLB is exhibit A: Yes, the law required states to break out data for all students and numerous subgroups, but the underlying information was essentially a lie, with states setting very low performance thresholds and calling it “proficiency.” And despite what many NCLB supporters will tell you, when you break down NAEP data—as I have done—there is little support for the notion that traditionally underperforming groups, or anyone else, have done better with NCLB than without it.

How about requiring common standards, both for academics and spending?

Even if you started with excellent, challenging academic standards, they would quickly be gutted at the behest of teacher unions, administrator associations, and probably even parents if many kids and schools didn’t meet them and were punished as a result. We’ve seen it many times, and there’s nothing about being federal that inoculates government against concentrated benefits and diffuse costs; the people most directly effected by a policy having the greatest political power over it. And financial data? As Adam Schaeffer has found, there are countless ways to hide the truth about district finances, and there’s little reason to believe that Washington will be either willing or able to sustainably force clarity.

One last thing: Where in the Constitution is the federal government authorized to demand “transparency”? Nowhere.

Second is ensuring that basic constitutional protections are respected.  No Child Left Behind required states to “disaggregate” assessment results to illuminate how disadvantaged or vulnerable populations…were doing.  Enforcing civil rights laws and ensuring that dollars intended for low-income students and students with disabilities are spent accordingly have been parts of the Education Department’s mandate since its creation in 1979.

Here there’s a slight connection to the Constitution: under the Fourteenth Amendment Washington has the duty to ensure that states and districts do not discriminate. But the presumption underlying what Darling-Hammond and Hess argue—that test data can reveal discrimination—is dubious. Can and should disparities in group scores really be laid exclusively at the feet of schools, districts, and states? Aren’t myriad factors involved in academic outcomes, many of which are outside the control of government?

Third is supporting basic research. While the private market can produce applied research that can be put to profitable use, it tends to underinvest in research that asks fundamental questions. When it comes to brain science, language acquisition or the impact of computer-assisted tutoring, federal financing for reliable research is essential.

We hear this one a lot, and in theory it makes some sense: people won’t risk their money on research that has no discernable payoff. The problem is few people ever contemplate the full cost of government funding “basic” research, or the unintended consequences.

The main concern is that putting money into things with no discernable payoff might yield just that—no payoff. So we hear about successes—government got us to the moon!—but rarely about how much has been lost in failed efforts. People don’t shy away from funding basic research just because they’re shortsighted. It’s also because they factor in risk.

Then there’s this: while we would like to think that all scientists are superhumanly selfless, they are not. They are as self-interested as the rest of us. Perhaps that’s why Austan Goolsbee—yes, Obama administration Austan Goolsbee—found in 1998 that much government R&D funding translated not into more breakthroughs, but higher wages for researchers.

What about the presumption that private markets wouldn’t put money into “brain science” or new tutoring techniques? Highly dubious. Education companies would have strong incentives to invest in research that could make them more efficient and effective because that would increase their profit margins.  The problem is, it is almost impossible to run for-profit schools in the United States, which can’t meaningfully compete against “free” government schools. In Chile, however, we see burgeoning evidence that profit can lead to greater scale—which is crucial for research—and better outcomes.

Of course, there’s nothing in the Constitution authorizing the feds to finance research.

Finally, there is value in voluntary, competitive federal grants that support innovation while providing political cover for school boards, union leaders and others to throw off anachronistic routines.

Again, sounds good, but as Hess and Darling-Hammond themselves admit:

The Obama administration’s $4.35 billion Race to the Top competition tried to do some of this, but it ended up demanding that winning states hire consultants to comply with a 19-point federal agenda, rather than truly innovate.

It’s easy to say that Washington should enable district and union leaders to ignore political concerns, but federal policy is as much government policy as state and local, and government at all levels is a creature of politics. Government and politics cannot be separated, and to expect one governmental level to be above politics while the others are below it is, to say the least, extremely optimistic. And again, there’s no constitutional authority to issue education grants.

Darling-Hammond and Hess are right that Washington has meddled far too much in education. They are on thin ice in asserting that different meddling will work much better.