Dylan Matthews is a great writer. You should read his argument against equality of opportunity.

But in his most recent piece on education, he greatly overstates his case.

Dylan argues we should:

1. Federalize education spending.
2. Federalize standards and assessments.

Dylan gets three things wrong:

Federalism is Anti-Fragile

Dylan thinks it would be a good idea to federalize education spending and standards + assessments. The risks here are obvious: all are eggs are in one basket and we can’t experiment. I assume Dylan likes the idea of federalizing spending and standards because he thinks this will create more equitable spending and increase the rigor of our standards.

But just because this is what Obama would do does not mean it’s what the federal government would do, either now or in the future.

What if the federal government cut education spending by 20%? Of what if it banned the teaching of evolution? Or what if it banned the teaching of climate change?

Each of these is plausible.

The point is this: in a federalized education system, one bad federal government decision could really mess up education for all of our children.

It is fragile.

The Connection Between Funding and Outcomes is Hardly Air Tight

I recently wrote about how, once you correct for income, most states perform about the same on NAEP.

These states also happen to spend vastly different amounts of money on public education.

Dylan cites one study that shows a connection between spending and outcomes. It’s an interesting study but it’s hardly a slam dunk case.

Big picture, the connection between spending and outcomes is much murkier than Dylan indicates.

50 States Allows for Experimentation and Innovation 

This is another reason federalism is important: fifty different spending patterns allows us to better test the connection between any given policy and student achievement.

So too with numerous other policies. Innovations like charter schools were pioneered by states and then other states adopted them often after watching the original pilots unfold over time.

Yes, the government can run pilot programs. But these pilots will always be connected to the federal political ecosystem, which is inherently far less diverse than the political ecosystems of the fifty states.

In Sum

I personally think that federalizing education spending and standards is a bad idea.

Yes, the federal government can and should use its education spending power to influence states, and I think a case can be made for opt-in national standards.

But having the federal government fully controlling the primary levers of public education would radically shift education in our country, I think for the worse.

Arguments for such a drastic overhaul should be couched in humility, a deep understanding of research, and a thorough exploration of the obvious objections.

I think Dylan failed to do this.

But I’m eager to hear his response. He’s a thoughtful person and talented writer, and perhaps he can make a more expanded case for having the federal government control education spending and standards.