Choice and Markets: Theory and Practice
And nothing could be more American than markets. We know that the way to drive quality up and price down is to create a market and let competition in that market do its magic. Choice and markets are like man and woman, a boy and his dog—they go together. And we also know that bureaucracies will never be as efficient as markets, because the bureaucrats run monopolies; they know that they will never lose their jobs and they have the political power to keep their salaries high and benefits generous, whether they are doing their jobs well or not, so they have no incentive to meet the needs of their own customers or to keep the price of the service down. We know these things.
So it seems obvious that, if we want to improve student performance and drive down the skyrocketing costs of education, we need to create effective markets in education and lubricate those markets with choice.
But do we really know all that? What I have just stated is the theory behind some of the most popular education reform strategies in the United States today. But there is no evidence, in the field of public education, anywhere among the advanced industrial countries, to back those propositions up. In fact, the evidence points in the other direction. Wherever these theories have been turned into policy in the field of education, they have not produced the advertised results. They have neither raised student performance nor lowered costs at the scale of a state, province or nation. The record actually shows that they can even make things worse.
A growing number of countries are surpassing the United States in student performance and are spending less per student than the United States. Not one has used choice and market incentives to do it. Some, like New Zealand, have embraced choice as a value and have developed policies that provide widespread choice for parents and students among public schools. But there is no evidence that these choice and market mechanisms have improved student performance overall and the research that has been done appears to show that there was greater inequality in student performance after such systems were installed than there was before they were introduced. In the United States, some studies show a slight advantage in student achievement for charter schools over public schools after accounting for differences in the background of the students and others show a slight advantage for regular public schools, but, when you look across all such studies, there is no clear advantage for either side.
How can this be? Why should the theories that underlie the very basis of the capitalist approach work so well in so many industries but fail to produce their magic in our schools?
First, we need to look under the rug of the theory. The theory says that, given choices, consumers will seek to maximize the value they are looking for in the product or service the industry offers. The education theorists assume that the value that the consumers will maximize for is student achievement. The whole theory underlying the voucher and charter movements requires that assumption. But the evidence leads elsewhere. Parents are first and foremost looking for schools they regard as safe. Safety trumps everything else. Next, everything being equal, they are looking for schools that are close to the student's home. Beyond that, different things matter to different people. Many, at the secondary level, are looking for schools with the strongest possible competitive sports programs. The quality of the academic program at the school often comes way down the list in the United States. I'll add to this list in a moment, but the point here is that there is no reason to believe that increasing choice will lead to schools being chosen on the basis of their academic program and therefore no reason to believe that choice will lead to improved student achievement. You can advocate increased choice because you value choice in a free society, but you cannot advocate choice because it will improve student performance. It won't, either in theory or in practice.
What really calls into question the idea that parents first and foremost seek schools for their children that maximize their academic achievement is what happens when the authorities try to close schools with abysmal student performance. The theory holds that, with sufficient choice available to parents and students, good schools will drive out bad ones as parents, informed about the academic records of the available schools, select good ones and abandon ones with poor academic performance records. But, as any school superintendent knows, the best way to assure yourself a short tenure is to try to shut down poor-performing schools. Communities across the country rise up in anger when an administration proposes to shut down its poor-performing schools and those who are angriest are the parents of the students currently in those schools. According to the theory, that cannot happen, but it does, all the time.
That is because most parents, apart from the factors I mentioned above, look for teachers who seem to care about their kids, places where their children are comfortable and where people know them. They want a school with a friendly staff and a principal who will solve the problems that parents bring to the principal's office. Apparently, when they have all this in a school that is close to home and seems safe, they will take that any day over another school that might have higher test scores, but is an unknown on these other points. Education reformers may want parents to make choices on the basis of student test scores, but they don't. And that blows a giant hole in the theory.
But there is another, deeper reason that the market theory is problematic. Consider why one school produces students with higher test scores than another. A famous 1965 U.S. Government report authored by a team headed by University of Chicago sociologist James Coleman found that the one of the most important factors explaining student performance was the socio-economic background of the other students in the school. Think about this for a moment. The finding essentially means that one of the most important things you can do for your kids if you want them to get a better education is to move into a community in which they will be surrounded by other kids who have the wealthiest and best-educated parents. Parents, of course, know this, so when they can, they move to the school districts serving the wealthiest and best-educated parents they can afford to be associated with. School choice is actually severely restricted in the United States even where it appears to be available. Poor kids cannot choose to get their schooling in rich kids' school districts. They can't even choose to get their schooling in districts that are only slightly richer than their own. What kind of choice is that, when we know that the parent's education background makes such a big difference in education outcomes?
This point about the influence of the customers on the quality of the service in the education arena is important for another reason. Think, for a moment about another industry with which we are all familiar: the grocery business. Go out to the wealthy suburbs and you will find a wide variety of grocery stores, everything from Costco to Whole Foods, Safeway to the local convenience store. But go to the inner city and it may be impossible to find any grocery store at all. Why? Because the big chains can't make money there. But, you say, that isn't true in education. Why? Because state and federal categorical programs provide extra money for the poor and minorities, per pupil expenditures are sometimes higher in many big city school systems than in some of the nearby suburbs. But so are the costs. Our inner city systems have high concentrations of handicapped students, homeless students, and students who live in homes where English is not spoken. It is much more expensive to educate those students than the students in the suburbs. My point is that high quality education providers are less likely to offer their schools on the scale that is needed in places where there are high concentrations of people who live in poverty. Even though there is money there, the costs are so high as to make it very difficult to offer a high quality service and still break even.
Most important, though, is the point made in the previous paragraph. What these schools lack are other students whose parents have levels of education and high levels of income. They lack one of the most important resources a school can offer. Why is that so important? Because the charter management companies that so many people are relying on to solve our school problems will grow and prosper only if they can show that they are successful. But their success depends more than anything else on successful parents, and schools serving the poor are not schools with successful parents. Precisely because this is a market in which the characteristics of the customer play a key role in the effectiveness of the product, this is a market that will not attract anywhere near enough competent providers to the communities that need competent providers the most.
Ah, you say, but what about the best known of the charter management companies? Don't they show that I am wrong, that competent providers will seek out the communities that most need competent providers? Actually, I don't think so. How many of the people who are likely to read this blog are likely to enroll their own children in the Green Dot schools? Very, very few, I would guess. They are good enough for other people's children but not your own, I will wager.
But this point cuts another way. Suppose for a moment that we set up a voucher system in which students have the right to cash in their vouchers in any public school system in the state. This, by the way, is very close to what the Romney campaign is actually proposing, at least for the expenditure of federal education funds. We know what is likely to happen from the research that has been done on systems in other countries that have been structured that way. Parents with the most education and highest incomes take maximum advantage of the available choices. They are more likely to have two-parent families with one parent not employed outside the home, so they have the time and the education to obtain and analyze the data on school performance and to go and visit the schools they are interested in and to sit in the back of classrooms to make their own judgments about what is on offer. They are more likely to have their own cars to get their kids to a school outside their own neighborhood or to be able to pay to get someone else to ferry their kids to and from school. And they are more likely to place a higher value on education than those on the other side of town who are having trouble feeding their families and putting a roof over their heads. Conversely, the low-income parents on the other side of town are more likely to be single-parents working two or three jobs to keep things together, with no time to find out what schools are available, to analyze the data on their performance, or to figure out the rules for getting into schools that are not nearby. In fact, they are less likely to know they have any choice at all. They are less likely to have their own car or the money or the time to get their children to a school more than walking distance from their home. So their kids are much more likely to be caught in a low-performing school. That school will be even less attractive to the best teachers who have a choice of schools in which to teach, so that school will end up with the most inexperienced and least competent teachers, who drag down the performance of that school, as will the increasing concentration in that school of students who are very difficult to educate.
In the end, the choice system and its market incentives will not improve average student performance, but it will, over time, work to make good schools better and bad schools worse. Is that what we want?
Any way you slice it, the market theory does not hold up in the light of either serious analysis of the premises on which it is based or the data on what happens when it is put into practice in education. If you want choice systems because you think there ought to be as much choice as possible in a democracy, go for it. But if you want choice and market-driven policies in education because you think they will raise student performance overall and lower the cost of provision, think again.