The point of publicly funded schools is to ensure that all children have access to essential education necessary to succeed in life. Today, however, the US public school system is riddled with inefficiencies and inequities that leave millions without the knowledge and skills they need.
Quality schools are a key driver behind decisions about where to live for many parents, but many other families lack the resources to find and evaluate good schools and move near one. This sorting means that poorer children are disproportionately educated in central city public school districts, which, over time, reinforces income disparity.
These points provided the background for the June 7 public discussion, “Lessons Learned: Understanding Education Policy Choices.” A panel of three economists who have studied the results of decades of reform efforts showed the audience that so far, charter and voucher systems designed to give parents and children more choice and better school options have not yet paid off in stronger educational achievement.
One reason that charter and voucher schools haven’t shown strong results that they have been tried with limited funding, on a small scale and in inefficient ways. Decades after these approaches were first introduced, charter schools still account for only 5.4 percent of public school enrollment; only half a percent of the nation’s students are using vouchers.
With the exception of the nation’s oldest voucher program, which was launched in Milwaukee in 1990, these programs have been operated with puny budgets. They often don’t allow flexibility and innovation in the way teachers are hired, evaluated, paid, and fired—a key to true improvement, said Derek Neal, professor of economics at the University of Chicago.
Neal pointed out that any attempt at school improvement rests on setting standards for learning, and then measuring proficiency against those standards. However, methods of measurement and reporting are very hard to verify—and too easy to game. When achievement tests are linked to high-stakes outcomes like school funding and teacher evaluations, they invite coaching and cheating, Neal said.
Carnegie Mellon’s Dennis Epple explained that while we care about the effectiveness of all schools, we should be especially interested in the quality of central city schools because they educate more low-income children who lack other options. He shared evidence that median family income is a better predictor of school achievement than school funding levels.
Epple noted that the research on charter schools show that while they sometimes improve outcomes, “it is fair to say that the typical charter school does not produce better outcomes than the typical city school. The same is true of vouchers.”
Chris Walters of the University of California, Berkeley provided more detail on that point. He has collaborated on evaluations of city and suburban charter schools in the Boston area that use an admissions lottery—the best way of evaluating impact because it’s possible to compare outcomes for students who got into a charter school with those who applied and did not.
Walters and colleagues found that overall, city charter schools did improve achievement scores, but the biggest gains were in “No-Excuses” charter schools following a model similar to the Knowledge is Power Program. These schools focus heavily on reading and math achievement using high behavioral standards; gains at other types of schools were modest.
But Epple noted that the No Excuses model may not be that much better and may not insure results everywhere. He also questioned whether there was selection bias behind the results. Students from families willing to stand in line for three days and nights to sign up for a lottery are likely to be more motivated and encouraged to study hard.
In contrast to the success of some city charters, Walters said that the Boston study found that children admitted to suburban charter schools actually performed slightly worse. And recent results from a study of a Louisiana program that provided vouchers for students to attend private schools found a large negative effect; student achievement declined by .4 standard deviations. “Which leads to the question of why parents are choosing schools that reduce achievement,” Walters said.
One possible answer is that that parents may not have the information they need to understand the impact an inadequate education can have on their children. Walters also suggested parents see something else in the schools than what evaluations are measuring, such as higher safety standards, arts programs, or religious grounding.
Neal said that may also explain why suburban charter schools in Boston that are underperforming standard public schools remain popular: the education quality is still high and the parents may be finding courses or experiences they value. Panel moderator Kevin Murphy agreed: “Parents making choice; I don’t think they wanted to shortchange their kids. There was something there that they wanted.
The private schools chosen to participate in the Louisiana voucher program were all struggling schools with declining enrollment, with per-pupil spending at about half that of public schools. The voucher funding was also low. That could explain the negative results, Neal said. “It’s possible that one way to read those results is that it’s a huge indictment of inefficient use of money in public schools” that spend twice as much. But, he added, “The best take on the voucher systems may be that they are giving people a choice of a place where they feel more at home, so they stay in school more.”
So how could we raise school performance and enhance school choice? Murphy highlighted the familiar favorite solution for economists—competition. If parents were actually provided with more and better information on schools and if there were actually a robust choice system, schools that are not performing well would either improve or be eliminated. “At a minimum, we have to give parents good information about school quality because clearly, over the last 20 years we haven’t given reliable, trustworthy data about what are the good and bad schools,” Neal said.
Such information would have to move beyond test performance and would have to include evaluations of other elements, such as student and teacher safety. Further, Murphy said, there has to be flexibility because different groups in different places need different things. “We can’t think uniformly,” he explained.
However, such approaches would only be possible if the system allowed schools to innovate on the ways teachers are hired, trained, and evaluated and how students are taught and tested. Plus, schools themselves, including private schools that opt in to voucher systems, must be subject to rigorous accreditation, something sadly lacking in many jurisdictions.
Unsurprisingly, all the panelists agreed that operating voucher and charter school systems on the cheap is not going to work. “Maybe we haven’t gotten results is that we’ve never tried vouchers at any here near full funding,” Neal commented. “Let’s face it. We don’t have a lot of evidence of what would happen if we have these systems real funding.”
— Robin Mordfin