Both New York City schools and Washington, DC schools pilot programs that pay students to learn. Mexico and Brazil have programs that pay families whose kids go to school. This is, by far, the most radical, and the most promising solution to the educational underperformance of American kids. In my view, these programs are not radical enough, because the sums paid to kids are nowhere near to what the public spends on education.
In the fiscal year 2006, school districts in the United States spent an average of $9,138 per student. New York State’s average is almost $15,000, although the City’s expenditures are a little lower, at about $12,000 in the 2004/2005 academic year (the last available year). We’re talking very serious money. Just imagine that the money would be paid not to schools but to the families directly, if the children can demonstrate learning. If the family has the skills and motivation to teach their own kids, they pocket the whole $15,000 a year. If they cannot, they may hire a tutor or sign up with a school of some sort, and share the learning income with those schools or tutors. Pass a test – get paid.
This would put the incentives to where they belong – with students themselves. We can provide all kinds of incentives to teachers, but if children are not motivated, it is not going to work. Education is not consumption; it is hard work, and it benefits us all more than it benefits about half of all K-12 students. This is why we’re subsidizing it in the first place. We should treat learning as any other job: if the society as a whole benefits from basic education, and is willing to pay money for it, why not pay directly to those people who actually need to learn? For the educational system to become more efficient, we must stop paying for attempting to teach, and start paying for proven learning.
Is it possible to implement? You bet. ETS and other testing agencies have decades of experience in administering tests on mass scale. Those tests do not need to be primitive testing of facts; they can include sophisticated measures of thinking, writing, and computational skills. Testing centers can be cheaply set up wherever there is internet access; each kid can have an account that keeps track of tests and of money paid to the student.
Will it leave behind poor children? Not at all; in fact, it will put resources into the hands of poor parents, and allow them to find the best educational solutions – with or without schools. We can also index payments in such a way that children with disadvantages such as poverty, non-native speakers, or with disabilities – receive higher payments than those kids with advantages. That will attract more talented teachers and tutors to poorer neighborhood, and create incentives for better specialized services for kids with learning disabilities.
Hyper-vouchers are the way to go. America has a history of radical, bold innovations. It has not been a part of our educational system for a while, but perhaps now is the time.