Last week I was critical of a New York Times op-ed by AEI’s Rick Hess and Stanford’s Linda-Darling Hammond. Yesterday, Hess graciously replied to my critiques, basically saying that it would be good if we could get the feds out of education, but since that’s highly unlikely, lets see how Washington can help.

That’s a modest and sensible stance, and I don’t think Hess is “endorsing big government.” (At least relative to most edu-analysts—admittedly a lopsided scale.)  But even if you accept that few in Washington are willing to boot themselves out of schools—and few are—it’s still critical to explore whether or not the things you’d have them do would be of net benefit.

Like last time, we’ll take the four proposals in order, this time based on Hess’s rebuttal. But first, one pet peeve:

Hess writes that he’d be happy to end “two centuries” of federal education meddling, noting that it all started with “the Continental Congress’s Northwest Ordinance of 1787.” I don’t know if this was his intent, but that factoid is usually invoked to suggest that even the Founders believed the federal government should advance education. This is not an impression that should be given: the Constitution is very clear in ceding Washington no authority to govern education outside of federal lands and civil rights enforcement. That the states have jurisdiction over education was, in fact, explicitly acknowledged as recently as the 1940s by a commission overseen by none other than Franklin Delano Roosevelt. And, while there was some federal education activity largely during and after the Civil War, it was not until the 1960s that Washington got heavily involved.

On to the four points:

First, when it comes to transparency, states have a collective action problem. There is both the problem of providing parents, taxpayers, and voters with meaningful transparency and the fact that state officials in each state have an incentive to manipulate performance results to their own advantage. More standards accounting and linking results to NAEP is a case of the feds providing a public good that only Washington is equipped to provide.

It’s true that state officials have a big incentive to manipulate performance results so that they stay out of trouble with voters and, especially, the teachers, administrators, and others who would be held accountable. The problem is that once you connect real consequences to NAEP—currently there are none—it will become a target for manipulation just like state tests and standards. Don’t attach consequences, however—including having no consequences attached to the state tests you’d audit with NAEP—and there’s no real impetus for schools to change. At best, then, this is a very limp proposal, and that’s before you get into big questions about whether the public really knows what NAEP assesses, whether one set of tests is a useful measure of education, and others I’ll save for another day.

Second, when it comes to basic research, the market tends to underprovide. Basic research is a public good…and is tough to monetize. The result is that, while the private sector is terrific at funding applied research, it tends to invest little in basic research.

As I mentioned last time, I hear this a lot but rarely see meaningful evidence to support it. And by “meaningful” I mean research looking at both the successes of government-funded basic research and the costs. Is it a net gain? Does the private sector steer clear of much of it because it’s an unjustifiable risk? Does it steer clear because government enables end users to rent-seek? There might be such research, but I’ve not seen it cited by those who assert that government must fund basic research. And then there’s the research I have seen that shows much of the funding translates not into innovation, but higher researcher salaries.

Third, even hard-charging state officials get tangled in decades of entrenched rules, regulations, and practices. The feds can help untangle this status quo by supporting officials seeking to throw off anachronistic routines but who must find ways to persuade skeptical constituents or union leaders to go along.

This if great if the goal is to clear out federal regulations, but state and local? There’s nothing in the Constitution authorizing Washington to manipulate state and local education systems, and perhaps more importantly: Why should anyone think it will work? The overwhelming, long-term track record for Washington is to add efficiency killing rules and regulations, and do the bidding of teachers unions and other school employees. Plus, why should we assume that the feds are able to pick the right routines to throw off or add on?

Finally, the federal government is obliged to ensure that constitutional guarantees of equal protection are observed. That said, this will ideally be pursued far less prescriptively than is the case today.

Here we agree, if Hess means Washington must stop clear state discrimination. And I guess one out of four ain’t bad.