Skip to content Skip to navigation

John Merrow, Taking Note: Pay teachers what they are worth (think six-figures)

Wednesday, August 19, 2009

Economists, whether liberal or conservative, don’t think about education the way most educators do, and that’s healthy. My friend Eric Hanushek is in the conservative camp, as his affiliation with the Hoover Institution at Stanford indicates. Rick has been interested in education–no, strike that–in doing something to improve education, for many years. He’s active on a number of fronts, particularly in Texas and with the Koret Task Force on K-12 Education. Professor Hanushek has a new book out, but, because he manages to sneak in two plugs in our interview, I won’t repeat the title here.

The Interview

Before we turn to No Child Left Behind, tell me your take on the current so-called “Race to the Top.” Secretary Duncan has an unprecedented amount of discretionary money, $5B, to give away. States seem to be falling all over themselves promising to do what Washington wants. Is this good?

I absolutely think the Secretary is doing the right thing, and I am actually encouraged by the positive reactions of the states. He has chosen particularly important issues to take to the states: developing systems for ensuring that there are effective teachers in every classroom; encouraging more competition in education through expanding charter schools; and developing good data systems that allow for reliable evaluation of programs and teachers. These are central elements of the funding and policy proposals in my recent book (Schoolhouses, Courthouses, and Statehouses), so I am thrilled that the Secretary is putting the force and the funding of the federal government behind these ideas. The essential unifying idea is that we should provide strong incentives to improve student performance – and each of these policy thrusts fits into that overall structure. I applaud the Secretary and the President for their forceful leadership in these substantive matters. Moreover, he has done this in a way that respects the states’ central role in education, while encouraging their movements in productive directions.

The Department says there will be winners and losers, but will that fly politically? Educators are accustomed to getting money based on formulas, not in a competition. Can you imagine the political pressure Arne Duncan is going to be under?

There is no doubt that the Secretary has taken a courageous position, because many resist the idea that policy should intrude on the way we have always done things. And his are not the positions that have been championed by the educational establishment. But, while there are political difficulties with standing firm, I think of the issue more from the viewpoint of what happens if he does not succeed. I frankly worry for the nation. I have come to believe, based in part on my own analyses, that improving the skills of our students is extraordinarily important for country. Simply put, our economic future depends upon having a very well educated population, and our students currently are not competitive internationally. There are indeed intense pressures to back off on the demands on our schools, but I am hopeful that he stands firm and that he maintains the leadership for true reform measures.

If Arne Duncan asked you for one piece of advice, what would you say?

My advice to him is simply to know his position. He leads the federal government’s education policy, and the federal government has a set of things that it is good at doing and a set that it is not. The things that the federal government is best able to do include: 1) Helping to define what we should be doing as a nation in terms of education; 2) Providing support for disadvantaged groups (as we do now); 3) Ensuring that we have a strong and effective research and evaluation program; and, 4) Providing incentives for states and localities to develop and use effective programs.

On the other hand, the federal government is not very good at running schools. This includes trying to set class sizes, trying to define what all failing schools should do, and the like. (This also feeds back into the discussions of NCLB, because right now the states decide what we should be doing and the federal government declares what the schools should do if they are not performing well. This division of effort is essentially opposite of where the federal government and the states have their comparative advantage).

You are an economist and a student of education. Is public education’s economic model–a teacher and 25 or so students–sustainable? Can you describe what sort of efficiencies technology could bring?

I am not sure if your question is tongue-in-cheek or not. The current public education model is to make this number ever smaller. Indeed we are currently at pupil-teacher ratios that are less than 16-to-1. This quest has been very misguided. I believe it is sustainable, but I also believe it is quite mistaken. We need to put much more emphasis on the quality of teachers. To do this, we can and should be paying highly effective teachers substantially more than we do today. But part of the package is not paying high salaries equally to effective and ineffective teachers. There is a clear bargain here. Pay effective teachers what they are worth (think six-figure salaries) but also have them teach somewhat more kids. That model is easily supported. What is not supported is paying large salaries to both effective and ineffective teachers and also reducing class sizes.

I suspect you are also thinking of some larger issues like the use of technology. On that score I am somewhat optimistic. It is remarkable to see some of the computer-based materials that are coming along. They offer the possibility of a better and cheaper way to do some — but not all– of the things we do today. The trick, as always, is getting the good things introduced into our schools. To me, that is a matter of getting the incentives right. We have to have ways that encourage teachers and schools to introduce new technologies when they are the best way to do things. We have not done a good job at getting these incentives right. Along these lines, we have to recognize that nothing in schools today really pushes good choices based on benefits and costs. This clearly has to change.

You’ve spent a lot of time thinking about the financing of schools. Are we figuring out how that should be done?

School finance has been one of the things we have done very badly, and our finance decisions have reinforced if not created some of our current problems. My new book with Al Lindseth tries to describe both problems and solutions. Fundamentally, we have developed a system that absorbs ever-increasing resources without getting higher achievement. As the title suggests, part of the problem has been the role of courts. The judiciary is heavily involved in educational policy decisions and, frankly, has not been very effective. Both the judiciary and the legislatures have tried to solve problems in our schools by putting in more and more resources, but they have done this in ways that keep the existing incentives and institutional framework.

If we are concerned about student performance, as I think we should be, we simply have to change the incentives by rewarding good performance. Right now we do the opposite. We are either indifferent to results or more problematically perverse in our incentives. Two quick examples illustrate the problem. Many states (including my own California) have programs designed to help failing schools. They provide substantial additional resources to schools that produce low achievement. But if the schools figure out how to do better and indeed get higher achievement, the extra resources are taken away. The incentives here are just the opposite of what we want.

The second example is that we pay all teachers (with the same experience and college degrees) the same — regardless of effectiveness, of needs (such as knowing math and science), and of willingness to work in the most difficult schools. So it is not surprising that teachers tend to get worthless master’s degrees and tend to move to schools where the job is not so hard. In our book, we describe the elements of a “performance-based funding” system. The essence of it is easy to grasp: be clear about what we want to produce, provide the basic resources to do the job, measure performance, and reward schools and individuals that prove to do a good job. This is not rocket science, but it is also far from what we have today.

It is very important, particularly when we face economic constraints such as those today, that we spend our school funds as best as we can. To me, this involves always focusing on what we want — higher achievement.

On balance, did No Child Left Behind do more harm than good? What’s been its greatest benefit? Biggest down side?

In my mind, there is no doubt that NCLB has been a net positive. NCLB of course mainly continued policies of test-based accountability that 44 states had already put in place in 2002. But it elevated student achievement, particularly by our most disadvantaged students, to a national policy objective. This focus on student outcomes is by far the most important aspect of NCLB, because I personally believe that improving student achievement should be at the very top of our national policy agenda. The knowledge of our students has direct ramifications for the quality of our labor force in the future, and our economic competitiveness strongly depends on this.

Is this an unqualified endorsement then?

It appears that NCLB has been working, especially for minorities and disadvantaged students who have performed better over time. This does not say, however, that NCLB is perfect. It has gained a bad reputation, partly for wrong reasons and partly for right reasons. Some of the grumbling about NCLB simply reflects the fact that many people do not want to be evaluated and held accountable. But the measuring stick in NCLB is also not the best. NCLB evaluates the overall test performance of students, which reflects not just schools but also families, friends, and neighborhoods — things not controlled by schools. It would nonetheless be straightforward to switch to accountability based on the growth in student achievement over time, something that is much closer to the contribution of schools. Indeed 19 states are currently authorized to use “growth models” to show that they are making appropriate progress toward universal proficiency. This change should be encouraged and strengthened.

A related point on measurement is that we do overly concentrate on just basic skills. A strength of NCLB is that it asked states to focus on having everybody meet some minimum standard of proficiency. This kind of commitment was never made previously, and by itself is important. Nonetheless, we are also concerned about higher levels of performance. We should not neglect students just because they have met minimum standards. But, again, this kind of change is easy to accomplish. We could for example reward performance improvements at all levels — basic to advanced.

In sum, test-based accountability as codified by NCLB is a very good idea. It has shown to be a productive policy. But, our early attempts to put this into law can be and should be improved.

NCLB not only focuses on student achievement but also mandates having a “highly qualified teacher” in all classrooms. Has this been productive?

Much of the work that I have done has focused on teacher effectiveness. From this research I have concluded that teacher quality is the most important factor in determining how well a school will do. That having been said, despite its good intentions, I think that the highly qualified teacher part of NCLB is actually its worst part. The reason is simple. We have huge measurement problems that make prior interpretations of this requirement hollow at best and harmful at worst. Teacher quality is not captured by typically discussed characteristics of teachers such as master’s degrees, teaching experience, or even certification – things that states typically monitor. Requiring such things unrelated to student performance dilutes accountability and detracts from things that would make them more effective. Fortunately, however, test-based accountability produces the student achievement data needed to assess the value-added of teachers, a more appropriate focus of policy concerns.

Just about everyone seems to be endorsing national or common standards. Is this a reason to cheer, or should we be worried–the logic being that, when everyone is in favor of something, the rest of us should watch out?

We know that discussions of national standards often lead to very emotional political responses (on both sides). The argument for some agreement is that our schools are really embedded in a national labor market, where students from Tennessee may well end up working in Texas or Oregon. Clearly specifying the skills that our citizens will need to be productive members of society is an important element of policy. Simply put, you have to know what you are trying to do before you can decide what to do.

The variations in state standards have been one of the problems with NCLB. When each state decides what is “proficient,” there is wide variation. At the same time, one always worries about the federal government in matters like this, because the federal government has no particular expertise in developing standards. And, we could end up with some really whacky ideas out of a federal political process around standard setting.

I personally think there is a middle ground. There are advantages to having standards set by groups larger than individual states, because this would likely avoid some of the problems found in the politics and processes of the separate states. We have recently seen a voluntary movement of states to join work toward common standards. The federal government could support such efforts, perhaps by paying for the process or the testing of states that join into consortia.

There is one other aspect of setting standards for larger groups of states. We could get considerable improvement of testing and assessment for these larger groupings. Right now, we require each state to develop its own test (reflecting its own standards). This does not make sense and clearly contributes to some of the poor tests that we currently employ.