For profit schools want to maximize profits. In the short-run this may come from passing all students or graduating them all. This way parents looking to get their child through school will pick them. This may all seem a quick fix for parents whose child is not doing well in public school. The for profit school can sell them passing, grade promotion and graduation with little effort.
This type of incentive complicates the question of whether for profit schools can out-compete public schools. There is substantial evidence that for profit schools and possibly even other voucher supported schools are simply trying to farm parents desperate for their child to pass and graduate.
An in-depth economics review of the economics literature in general on vouchers, but not covering this point is here. https://www.nber.org/system/files/working_papers/w21523/w21523.pdf
A review at Brookings of results shows poor results of voucher programs.
Public schools play two roles. One is to filter students. They limit who can get a credential, such as a high school diploma or receive good grades in math or science courses. Then they turn around and try to educate students to pass those filters. The public and government impose testing and requirements to get certifications or good grades. This includes employers who want the certifications and grades to be meaningful.
For profit schools have an incentive to game the system. They can lure in parents by offering credentials. So instead of teaching up to the test and certification requirements, they simply pass as many students as possible. This is particularly true if they are just recent start-ups. They want to get in bodies to cover their costs. So they just pass them.
Filtering students puts pressure on the schools as well as on the students. The students are pressured to pass. The schools are pressured to help them pass. Maintaining this tension is needed for education to actually educate. Many voucher schools fail at this. They are just going for short term revenue and profit maximization to stay in business.
Because of this moral hazard problem, vouchers can’t just be analyzed as if education was an industrial product. It is also a filtering product. The consumer isn’t just the society. The consumer is a future employer or a future node in the education system. The later stage consumers don’t have much control or choice in a voucher school. Only one consumer is choosing with a voucher, the parent. The employer and the college don’t get to choose the school, at least not right away. They choose many years later by rejecting students from voucher farm schools, i.e. ones just farming the system.
In Milton Friedman terminology, the future employer and the future college are not free to choose in the voucher system. At least not right away. They choose many years later. This time gap creates an enormous moral hazard. Which many voucher school operators fill with short term revenue to cover costs.
Milton Friedman’s voucher proposal came in the 1970s or possibly earlier. He may not have used the type of signaling theory and game theory we have today in economics. Although, those did exist at the time, they were difficult to apply then and still are. They do have a heavy level of math and Milton Friedman was not a mathematical economist, at least compared to some.
Voucher schools should be analyzed in terms of this moral hazard and that employers and colleges get a delayed input into choosing the voucher schools. This delayed choice may make voucher schools fail in practice.
Any attempt at voucher schools has to take into account this type of analysis, both at a theoretical level and empirical level. Any voucher experimental design has to confront that key choosers only make their choices years later. At which stage, it may be too late for the student and parent.
Milton Friedman was thinking of schools in an industrial model where they just produced a product at lower cost. But because schools filter and not just educate, and have a moral hazard problem when they give up filtering, his analysis is incomplete and inadequate. This is why in practice, voucher schools have been disappointments.
Milton Friedman’s vouchers are a failure in mechanism design. They don’t address the moral hazards of voucher schools and the delayed choice of key agents, colleges and future employers, whose choices are made years later.
There has to be a new mechanism design for school vouchers. This mechanism design has to build in the delayed choices of colleges and future employers. Maybe someone has done this analysis and published it. Maybe not. There has to be a new mechanism design before voucher systems can move forward to a new stage of testing and experiment. The Milton Friedman mechanism design leaves out too much to be adequate.
At this stage, economic theorists and others need to build a new mechanism design for school vouchers. This has to deal with the moral hazard of voucher schools to just pass students to farm the voucher programs as a revenue source by attracting desperate parents.
A paper on mechanism design and school choice is School Choice: A Mechanism Design Approach
Atila Abdulkadiroglu, Tayfun Sönmez, located here. It has a different focus than the above, but sometimes it may get close to these issues.
Some of the issues in mechanism design of voucher programs are testing, transparency, and accountability. The voucher schools need to all the same testing as the public schools and maybe more. The testing has to be done by outside parties to be reliable. They have to make the testing results public for transparency. They have to be accountable to future stakeholders like colleges and future employers in some way.
Selection bias in who goes to which voucher school is a major issue to transparency of a voucher school’s performance. Voucher schools have to be able to assess the selection bias in order to make their performance results be comparable. This has to be a mandate on the voucher schools themselves as well as on any voucher program design. Any voucher system experiment has to deal with the issue of selection bias in how students were assigned to which school. At this stage, it does not appear any solution has been adequately structured, tested and verified. Understanding how to do this is part of the job of a school board member, since the same issues apply even within the public schools by themselves. So a good foundation in economic theory and econometrics needs to be part of the skill set of a school board as a whole.
Another way to sum up the Milton Friedman school voucher proposal is that it lacks a detailed analysis of the market microstructure of education. The failure of voucher programs to beat the public schools, and to often fail, shows that in education, market microstructure matters. The market microstructure of the education market is extremely complicated and reflects almost every issue at play in American society. Because the proposal is not based on a detailed analysis of the education market microstructure, the mechanism design of the voucher proposal is inadequate for the actual reality of the market microstructure of the education marketplace.
The experience to date is that the Friedman voucher proposal is malformed to the actual education market microstructure and produces bad outcomes. So large voucher programs have to be put on hold until a detailed education market microstructure analysis exists in the form of many studies and published papers and in practical know-how that is applied by working school boards. Small experimental voucher programs where the parents have to sign a consent that we lack sufficient know-how to guarantee a good outcome can go forward on a limited basis.
