Jay P. Greene's Blog

Mike Petrilli seems concerned that I haven’t answered his questions about how to address certain problems without resorting to federally-imposed national standards.  I thought I had.  I said: “The answer is not to have bigger, more centralized regulations.  The answer is to maintain the proper incentives by empowering market forces, which also serve to keep the regulatory framework honest.”

I also thought I answered his questions when I said:

Even though it is messy and imperfect, we need to decentralize power in education rather than centralize it.  We need to do so for the same reason the Constitution decentralizes power — to prevent abuses and tyranny that inevitably arise when power is unchecked and concentrated.  We need to decentralize power in education to allow market mechanisms to operate.  We need to decentralize power to recognize the legitimate diversity of needs and approaches that exist in our educational system.

Benevolent dictatorships are always attractive on paper but the benevolent part never works out in practice.

But perhaps the problem was that I didn’t apply my answers to each of the questions he raised, so I’ll do so here:

Mike asks:   If not through common standards, how else should we address the problem of vague, content-free state standards?

I answer:  Focus political pressure on states with weak standards and assessments.  Using NAEP to shame weak states, as Paul Peterson and Rick Hess have been doing, is helpful.  Choice and competition among states should also help to some degree.  States with lousy standards and assessments will have a harder time attracting (and developing) skilled labor and will attract less capital investment, job-creation, and, as a result, generate less tax revenue.  Lastly, centralizing the standards and assessment process at the national level does nothing to address this problem and may well make things worse, as I’ve been arguing.

Mike asks:  Laughably low cut scores?

I answer:  See the answer to the last question.

Mike asks:  Tests that are poorly designed and can’t possibly bear the weight being placed on them, from value-added demands to merit pay to teacher evaluations, etc.?

I answer:  See above.  Also, there are several high-quality, for-profit testing companies.  States could be urged to contract with one of them.  Frankly, most states already do and most tests are technically reasonable, so I don’t see this as a big problem.  I do see moving the development of assessments to the national level as a big problem because then you are liable to get the Linda Darling-Hammond test focusing on project-based learning, measuring collaboration, etc…

Mike asks:  Small state departments of education that don’t have the resources or capacity to get this technical stuff right?

I answer: Name me a state that does not have the resources to hire a decent commercial testing company.  Even the small state departments of education have more money than Croesus.  And as we know from Caroline Hoxby’s research, the cost of testing is trivial.

Mike asks: Textbook and curriculum and professional development and teacher training markets that are fragmented into fifty pieces?

I answer:  I’ll answer with a question.  What’s bad about having 50 textbooks, curricula, professional development, etc…?  We have more than 50 different restaurants, book publishers, etc… and the expanded choice and competition in those sectors helps improve quality.  It’s odd to hear someone call for a monopoly when the government normally tries to break those things up.

(edited to correct typos)



Over at Flypaper Mike Petrilli has finally tried to address the problems we’ve raised regarding national standards.  Despite Mike’s best efforts, I’m afraid that national standards and assessments still sound like a really bad idea.

I raised doubts about the rigor and soundness of the proposed national standards, citing the fact that many credible experts have denounced them as lousy.  His response is simply to repeat that Fordham has given the standards good grades and thinks the latest revisions have been positive.  This is not a substantive response; it is simply a reiteration of their initial position.

Why should we find Fordham’s grading of the proposed national standards any more credible than that of the experts who have denounced the standards?  The fact that Fordham issued a report with letter grades is just a marketing exercise for Fordham’s opinion.  There is nothing scientific or rigorous about Fordham hand-picking their friends experts to repeat the opinion Fordham already holds — especially when we know from past experience that Fordham might exclude experts or change the grades if it does not come out the way they want.

Yes, the national standards may be better than those in some states, but everyone seems to agree that they are also worse than the standards in some states.  Why should we hurt the excellent standards in MA or CA to improve the standards in AR or MS?  Wouldn’t it be smarter to focus our energies on pressuring states with bad standards to improve them?

It is true that the Edublob dominates the standards and assessment process in many states, but the existence of choice and competition among the states places constraints on their ability to impose nonsense through that machinery.  If the standards and assessment process is centralized at the national level, the Edublob will be able to impose nonsense on everyone with no “exit power” to constraint them.

Rather than rely on market mechanisms to constrain nonsense, Mike places his trust in devising national political systems that he thinks can develop and maintain good national standards and assessments.  In particular, Mike thinks that it is “more likely that the good guys will stay in charge at the national level, where all of this stuff will operate under the bright lights of the national media, than in the states, where decisions get made behind closed doors.”  The national government also regulates off-shore drilling and the financial system.  How well did those bright lights work at ensuring a sensible regulatory framework?

The hard reality is that regulation tends to be captured by the regulated industry (unless there are competing, well-organized interests, which in education there are not).  Education regulations, like national standards and assessments, are at least as likely to be captured by the Edublob as the oil industry is to capture off-shore drilling regulations or the banking industry is to capture financial regulations.

The answer is not to have bigger, more centralized regulations.  The answer is to maintain the proper incentives by empowering market forces, which also serve to keep the regulatory framework honest.  I’m not advocating against all regulations.  I’m saying that there need to be market checks and balances to keep regulatory frameworks reasonable.  If we centralize the standards and assessment process, we have eliminated some of the few market checks and balances we have in education.  The fact that Linda Darling-Hammond is part of the leading bid to develop national assessments to go along with these national standards should make clear the dangers of nationalizing this process.

And make no mistake.  The Obama administration has signaled that it intends to link federal money to adoption of a Linda Darling-Hammond test or whatever other nonsense this centralized process may produce.  Just because Mike thinks  ”the Administration erred and gave national standards opponents an opportunity to raise concerns about federal overreach” doesn’t mean that they aren’t going to do precisely what they have declared they will do even if he thinks it is mistaken.

But the most telling comment of Mike’s faulty thinking on national standards was when he asked: ”Does Jay oppose voucher programs because they might get hijacked by shady for-profit providers who just want to make money off the backs of poor kids?”  The fundamental difference between the potential for “hijacking” of national standards and assessments and the “hijacking” of a voucher school is the mechanism by which one can control (or hijack) them.

Voucher schools are controlled primarily by the market choices of parents.  You can’t “hijack” a voucher school because parents can choose to go to another school if they dislike what the school tries to do.  But you can “hijack” national standards and assessments because they are controlled politically and not by market forces.  People who dislike what the national standards and assessments do are still compelled to send their children to schools operating under that national system.  You don’t need parental or even popular buy-in to hijack national standards and assessments.  You just have to be better politically organized and motivated to dominate the process by which those standards and assessments are developed and maintained.

This all leads to my question that Mike never answered:

“If there really were one true way to educate all children, why stop at national standards? Why not have global standards with a global curriculum?

We would oppose global standards for the same reasons we should oppose national standards. Making education uniform at too high of a level of aggregation ignores the diversity of needs of our children as well as the diversity of opinion about how best to serve those needs. And giving people at the national or global level the power to determine what everyone should learn is dangerous because they will someday use that power to promote unproductive or even harmful ideas.”

The reason Mike and other supporters of national standards and assessments don’t advocate for global standards and assessments (even though the logic for doing so is essentially the same as national standards and assessments) is that they imagine that they’ll be the ones controlling the national process.  Someone else would dominate the global one and that would have to be bad.

As much as I like Mike, I don’t want him or (more likely) the Edublob dominating national standards and assessment, which would have profound effects on how every classroom in the country operates.  Even though it is messy and imperfect, we need to decentralize power in education rather than centralize it.  We need to do so for the same reason the Constitution decentralizes power — to prevent abuses and tyranny that inevitably arise when power is unchecked and concentrated.  We need to decentralize power in education to allow market mechanisms to operate.  We need to decentralize power to recognize the legitimate diversity of needs and approaches that exist in our educational system.

Benevolent dictatorships are always attractive on paper but the benevolent part never works out in practice.