Webinar Today at 1pm ET: PISA in Perspective: Cultivating Mathematical Mindsets. Register here.

Cross-posted at Education Week.

Twenty-five years ago, David Osborne, a journalist now based at the Progressive Policy Institute, and Ted Gaebler, a former city manager, wrote a book called Reinventing Government: How the Entrepreneurial Spirit is Transforming the Public Sector that got a lot of attention among city and state officials all over the United States. In it, the authors advanced a set of powerful ideas designed to transform tired bureaucracies into high performance organizations driven by strong incentives to deliver and constantly improve results.

They wanted to do it—as a recent article in Governing magazine put it—by “…fostering competition, abandoning bureaucratic processes, measuring outcomes rather than inputs, and demanding accountability.”  I remember thinking, when I read their book, “Well, this is a breath of fresh air.”

Last year, Osborne came out with another book, Reinventing America’s Schools, which applies his thinking on reinventing government to the schools sector.  In the same Governing magazine article I just quoted, the author summarizes Osborne’s ideas about school reform by focusing entirely on Osborne’s proposals for a particular approach to charter schools.

In essence, the idea is very like the core idea in our 2006 report, Tough Choices or Tough Times.  In that report, we proposed that states empower school districts to stop actually running schools, and, instead, develop specifications for the outcomes they wanted from their schools with metrics for measuring student performance against those specifications, write announcements presenting the terms of competition among organizations that wished to run schools designed to produce those outcomes, make awards to the offerors judged to have the highest probability of producing the best outcomes at prices the district could afford, gather data on  the performance of the winners according to the metrics, dump the contractors that performed poorly and renew the contracts of the superior performers.  This is chartering with a strong, results-oriented accountability system.

In a recent blog, I characterized Osborne’s approach as featuring proposals for more competition among schools and for more autonomy for schools.  I was not alone in doing this.  As I said, the reviewer of his book in Governing magazine had done the same thing.  But Osborne responded to my blog in a note, saying he thought I had misrepresented his views.  He believes, he says, “…that competition between public schools is better for children than monopoly, and I believe many studies have proven that. But it’s far from sufficient. School autonomy, accountability, differentiation of learning models, choice, and teacher and school leader talent are also critical, as I argue in my book. And competition has to be tightly regulated, because you’re so right, education is a public good.”

Perhaps I did not do his views justice.  He is a thoughtful analyst and journalist and he will make you think.  You should read his book.

But that does not mean that I come out where he comes out on system design.  If you have been following this narrative, this is where you should sit up straight.  You might be thinking, “If Tucker says that Osborne’s argument is virtually the same argument that he made in NCEE’s Tough Choices report, how can he now say that he comes out in a different place than Osborne?”

The answer is that I have changed my mind about some very important aspects of system design over the last 18 or so years.  I’ll start with my initial reaction to the No Child Left Behind (NCLB) legislation.  My first reaction to this bipartisan landmark bill was, “Hey, this bill does not just put money on the table in the hope that something good will come of it.  It actually holds the professional educators responsible for the results.”

I had many good friends, some serving in some of the worst-performing districts in the U.S., who, after years of service were still giving it everything they had, working at their jobs not just during the required hours, but late into the afternoon, at night and on the weekends.  I saw the glow in their eyes when they described a student whose life they had quietly rescued from virtually certain wreckage and I saw how those wins had enabled them to go on.  These are the people who are my personal heroes and heroines. At the same time, it infuriated me that these extraordinary people were all too often treated by the organization for which they worked no differently than the teachers who had burnt out years before and were just hanging around until they could collect the best possible pension.  I saw unions that saw their jobs as protecting these teachers rather than the students they were failing.  Because I had spent years analyzing data on the countries with the best student performance, I knew that U.S. student performance was, on average, mediocre. I put down the NCLB legislation and sighed to myself, “Well, it is about time.  Now there will be some reward for teachers who are very good at what they do and schools that are very good at what they do.  Finally, there will be a policy regime that connects the provision of funds to the performance of both the students and the professional educators.”

The design we offered in the Tough Choices report was the logical conclusion of this kind of reasoning.  It shifted the focus from inputs (mainly more money) to results, and said, in effect, how you get the results is up to you, but you had better get them or we will get someone else who will.  Just as in the American economy as a whole, this stance would not only make the system much more efficient, it would unleash untold amounts of entrepreneurship and creativity, in the process upending the bureaucracy and sending the dead wood flying out the door.

That, I think, is Osborne’s argument, too, but my own thinking has evolved on these key issues of education system design.  Since we produced Tough Choices, two things have happened.  First, this country has learned a great deal about what does NOT work from NCLB and, its successor, Race to the Top.  Second, we have—or should have—learned a great about what DOES work from the countries whose students are greatly outperforming ours.  This is my 370th blog.  The vast majority of them have been addressed either to what went wrong with NCLB and Race to the Top or to what we should have learned from the top performers.  I will not try to summarize them here.  Instead, what I want to do is say just a few words about where I differ now from a model Osborne and I once shared.

Shifting the focus from inputs (mainly money) to results    It is certainly true that getting good results depends in significant measure on being clear about the results that are wanted.  But, as I have pointed out again and again, Americans are not yet willing to spend what it takes to get good assessments of the kinds of complex student performance our students will need in the new economy.

Even if we had used much better tests and used them to cover more of what most of us think of as the core curriculum over the last 18 years, the fact is that there is a growing consensus that our schools should be doing much more than developing the kind of cognitive skills that one grows in conventionally organized classes. That includes developing everything from “grit” and determination to the ability to organize one’s work, to contribute to the team when that is required and lead it when that is the challenge, to say nothing of behaving ethically and being able to understand what the world looks like from the vantage point of others.  Our ability to measure these things, if the standard is the American conception of objectivity, is rather limited.

That is not an insurmountable obstacle if one trusts the judgement of our teachers, but if you are using the very limited measures we are accustomed to now to decide which teachers should be hired and which fired, then you are setting yourself up for failure. There are many lessons here, but the most important for our purposes is: Don’t load up the measurement system on which accountability is based with a greater weight than it can bear. If the only thing the school is accountable for is student performance and educator’s jobs and contracts depend on the outcome, you had better be sure that the measures you will use will capture everything you are after and the results cannot be gamed.  Good luck with that.

Once the goals and measures are in place and the accountability system is implemented, give the schools the autonomy they will need to come up with innovative, creative ways to get there.

This is a simple and powerful idea. It should maximize efficiency and creativity.  But I have come to the conclusion that it won’t produce the expected results.  High-performance mass education systems are just that: systems.  Their parts and pieces fit together like giant jigsaw puzzles.  They have to, because the rules that apply in freewheeling markets don’t apply.  It is OK with us if some families own three high-performance cars and another has no car at all.  It is not OK if some kids get a first-rate education and others get a lousy one that leaves them ready for neither college nor career. We know that students who come from vulnerable families need a lot of help before they arrive at the school house door, from a whole raft of public and private agencies, if they are going to have a decent chance of reading by the end of grade four.  We know that our system of school finance, based largely on local property taxes, typically provides more resources for the students from our wealthiest families than those with the least money.  We know that our teachers are much less well compensated than their peers in the top-performing countries and are often educated and trained in very low-performing teachers colleges.  We know that the students who most depend on school for a chance in life are the ones most likely to suffer from low expectations and a weak curriculum. We have only to look at the top-performing countries to see how these and many other key challenges can be met at a cost little more than we are paying now.

But challenges won’t be met by saying to the schools: Here’s the measure of your success, either meet it or get shut down.  The reality is that these problems are all linked to each other and they require linked solutions.  Implementing those solutions requires capacity that neither the regular public schools nor the charter management companies have.  The failures of NCLB and Race to the Top were partly failures of design but they were primarily massive failures of implementation and lack of implementation capacity at every level of the system.  Only the state can come up with that design and only the state can assure that it is properly implemented.

The hard reality is that our failure to reach the student performance levels attained by the top performers and the kind of equity we see in those systems is not just a failure to get the goals clear and the incentives right.  It is a failure to put in place the key features of the systems they have been using and to develop the capacity in the system to implement those features well.

Some of our biggest and best known charter companies get results by hiring cheap teachers who they work very hard and who do not last very long, burnt out long before their time.  Maybe that is innovative but it does not attract to a career in teaching the kind of teachers this country needs to greatly raise student performance across the board.  Others use a curriculum that is heavily focused on drill and practice that will produce near term gains on state basic skills accountability tests, but will leave the same students high and dry when they go on to serious study that requires deeper understanding of the material.  It is not a formula for getting vulnerable students to high standards.  The whole system needs to be redesigned if that is going to happen, not just the incentive structures.

It is time for the states to create effective designs for high-performance systems and to insist that schools and districts get the money they need to get their students to much higher standards only if they are willing to fully implement the key features of the system designs that the top performers use. A system that puts the full weight of accountability on schools and districts only for student performance, using poor measures that capture only a small part of what we all want for our students, is bound to fail.  Yes, there should be accountability for results but there should also be accountability for implementing system features that we know are essential for top performance and high equity.  But even when schools and districts see the need and are willing, most will lack the know-how and capacity to do it. While accountability is essential, it is just as important for the state to provide the support that schools, universities, districts and communities will need to implement these features well and wisely.