Let me say at the outset that what I am about to propose is not going to transform America’s education system. It won’t propel the United States ahead of our international competitors on PISA. It won’t eliminate our stubborn achievement gaps. It won’t do any of these things because, for better or worse, the federal government is incapable of affecting these kinds of sweeping changes. Not for any ideological reasons, but for structural reasons. Uncle Sam is at least three steps removed from the classroom, and all the carrots and sticks in the world won’t allow him to make everything right in our schools.

The WRONG way to think about federal policy in education is to identify the myriad problems plaguing our schools, and then dream up federal solutions, as if Congress could pass a law and magically things would change in the real world (and without any unintended consequences).

The RIGHT way to think about federal policy is to figure out what Uncle Sam is capable of doing, and then doing that well.

Now, I disagree with some friends on the right, like Jennifer and her colleagues at Heritage, that there is basically NOTHING Uncle Sam is capable of doing right in education. For instance, thoughtfully crafted competitive grant programs, like the Teacher Incentive Fund, CAN move the reform ball down the field.

But I strongly disagree with my friends on the left, like Cindy and her colleagues at CAP, who seem to think that Uncle Sam’s capacity to do good is practically unlimited.

Here are Fordham, we embrace the notion of Reform Realism. Washington absolutely should promote reform. But we have to be realistic about the limits of its power and influence. Because while Washington CAN force states and districts to do things they don’t want to do, it can’t force them to do those things well. And almost everything in school reform is only worth doing if done right.

So what does this mean for federal policy around accountability? There are two questions. First, should Washington prescribe a particular school ratings system for all the states (like the rating system we call AYP)? And second, should Washington prescribe particular interventions for failing schools in all of the states? Let’s ask: Can Washington do these things well and without lots of perverse consequences?

To my eye, the answer is clearly no. Nobody will come right out and defend AYP, with its byzantine rules and dichotomous ratings. Yet, if you read the policy papers coming out of CAP, Ed Trust, the Chamber of Commerce, and other left of center organizations, they don’t want to scrap its most onerous parts: a deadline for getting all kids to “proficiency,” Soviet-style annual goals, dozens of boxes to check in order to be considered a good school, etc. They might agree to a tweak here or there, but their Son of AYP would look a whole lot like the AYP we know and hate today.

And what about federally-mandated interventions? This has been a total flop. Not because federal law wasn’t prescriptive enough, but because NCLB’s architects never had a realistic theory of action for how the feds were going to compel recalcitrant states and districts to implement these sweeping reforms. Setting up a public school choice program, or free tutoring program, or intervening in failing schools—all of this is hard under the best of circumstances. Expecting states and districts to do these things under duress and do them well is nuts.

So what would I propose instead? Three words: Transparency, transparency, transparency. What the federal government can do and do well is ensure that schools results—AND finances—are transparent to the public.

It starts by encouraging rigorous standards and tests, so that the test scores that are at the base of any transparency system can be trusted and might mean something in the real world. The Common Core initiative is a huge step forward here.

The feds can also reasonably ask the states to develop ratings systems so that parents and educators have an easy to understand signal about whether their schools are on the right track. But rather than prescribe a rating system from Washington, Congress should simply list the elements that states must include. I would include: Student growth, subgroup performance, graduation rates, and more.

Not everybody is going to be happy with the system every state develops. They might not focus enough on achievement gaps for someone’s taste, or value-added, or science scores, or whatever. So the feds should also ask the states to make public all of the data behind the school ratings so that any independent organization can build their own ratings system. Education Trust, for example, could develop an achievement gap index that would allow viewers to compare all schools in Common Core states, in an apples to apples way. GreatSchools would build its own.

As for intervening in schools that aren’t measuring up, here’s where we’re in need of a big wallop of humility. Let’s face it: nobody really knows how to turn around failing schools; a recent Fordham study found that just 1 percent of low performing schools reach turnaround status five years later. So it doesn’t make any sense to me to pretend that the federal government can waive a wand and make these interventions happen and happen well.

If Congress can’t help itself, and wants to intervene in the very worst schools, I would propose a competitive grant program for states and/or districts that really want to do the work.

So there you have it.

Will my plan fix all that ails America’s schools? Nope, but neither will anyone else’s federal program. I like to think that I should get points for being honest about it.

-Mike Petrilli

Note: These were my opening comments during last week’s Fordham Institute panel, “Is it Time to Turn the Page on Federal Accountability in Education?” Video of the event is available here.