If school vouchers bettered the educational opportunities only of children who use the vouchers to attend private schools or schools in another district, many reformers would be left holding cups half empty. For the animating theory of school choice has always been that it will not only serve as an escape hatch from dysfunctional public schools but also will spark public schools to improve. Thus far this theory remains mostly untested. Through caps on enrollment, chronic underfunding, and legal attacks, voucher programs have been kept artificially small, restraining any influence they might have on local districts. The combined enrollment of all the publicly and privately financed voucher programs in the nation was still only 0.1 percent of public school enrollment in the fall of 2000. The statewide voucher program in Florida affected only two public schools directly.
Despite the limited data, scholars have found creative ways of teasing out the effects of competition on public schools. Elsewhere in this issue, Harvard economist Caroline Hoxby reports that cities with many low-cost private school options tend to elicit better performance from their public schools. Hoxby also finds that urban areas with a large number of school districts, and therefore many options for families choosing where to reside, tend to have higher test scores than cities like Miami, where one school district covers anyone living close enough to work in the city. However, the options of choosing a private school or locating in a suburb with high-performing public schools are mainly closed to low-income students.
Charter schools, by contrast, are tuition free and, in many states, take per-pupil funding away from local school districts. They therefore present a threat-and a real source of competition-to traditional public schools. As of the spring of 2001, the Center for Education Reform estimated that 1,750 charter schools were educating about 520,000 students in 36 states and the District of Columbia, more than seven times the number of students in all the public and private voucher programs combined. And certain cities and states have been so supportive of charter schools that they have nurtured education markets competitive enough to warrant study.
Why Study Arizona?
During the past three years, Kansas City, Philadelphia, and Washington, D.C., have used charter schooling to create somewhat competitive systems at the city level. Arizona alone offers a statewide free market in education, a result of its being the only state that has not placed caps on the number of charter campuses that can be opened. From 1994 to 2000, two state-level boards were each able to issue up to 25 charters annually, and a single charter holder can open multiple campuses. Local school districts could grant a charter to schools operating in other districts. (In 2000 the charter law was amended to remove the 25-charter cap while ending the practice of local districts’ chartering schools in other districts.)
As of the fall of 2001, Arizona’s charter law had spawned about 295 operators with 431 campuses among them. About one in four Arizona public schools is a charter school. Nationwide, only about 1 percent of public school students are in charters. The charter-school share is about 7 percent in Arizona, with some districts losing more than 20 percent of their potential enrollment to charters. One of the main reasons why the charter movement has spread so quickly in Arizona is that per-pupil funding in charter schools is about 95 percent of what traditional public schools receive-which is still not much in Arizona, but it is at least enough to be competitive with traditional public schools.
Moreover, district schools lose state funding equal to 57 percent of their per-pupil funding for every student who leaves the district for a charter school. So traditional schools have incentives to compete for students. In other words, Arizona is a virtual laboratory for researchers looking to study the effects of competition on public schools.
I have studied Arizona’s charter schools as part of a team with widely varying views of school choice. Together we conducted a major teacher survey, and I personally visited 28 charter campuses in Arizona run by 20 different operators and conducted in-person and phone interviews with more than 200 policymakers, district school officials, and charter school teachers, operators, students, and parents between November 1997 and May 2001.
Our findings suggest that competition improves education for all students, including the vast majority who remain in district schools. Yet the changes are evolutionary rather than revolutionary, often less dramatic than many supporters of school choice had hoped. Still, the evidence shows clearly that other states would do well to study and emulate the Arizona model. Eliciting incremental improvement while doing little or no harm is about as good as it gets in the world of social policy.
The effects of competition in Arizona have been muted at least partly by the fact that Arizona’s public school enrollment grew by roughly 20,000 students each year through the mid-1990s, dwarfing the number taken away by charters. Indeed, some Arizona education officials suggest that charter schools enable district schools to manage their growth. In addition, many charters serve the at-risk students who district schools are more than happy to see leave. District schools also tend to ignore some charters simply because some charters are not very good. Finally, a few charter schools offer curricula so unusual that school boards are reluctant to copy them for fear of controversy. A proposal in the early 1990s to create a Waldorf magnet school in Flagstaff offended organized groups of parents and teachers who saw Waldorf curricula as religious or just flaky. Flagstaff and Sedona now have Waldorf charter schools, and local districts have made no effort to compete with them.
Nevertheless, when we conducted a survey asking long-serving teachers to rate their schools on a number of criteria in the spring of 1998 compared with the spring of 1995 (immediately before charter schools opened in Arizona), the results showed clear perceptions of improvement since the advent of charter competition, with the greatest gains in the districts with the most charter schools. Surveys of teachers in Nevada, where there were no charter schools, showed no improvement in teachers’ perceptions of their districts. Further interviews and fieldwork suggest that the Arizona districts hit hardest by competition react in the following ways:
• Improved customer service. The Roosevelt district, an inner-city district in Phoenix that was hit hard by charters, sent letters to local charter parents asking why they had left district schools and explaining how the district would serve them better. Teachers and administrators in the nearby Isaac district visited parents in their homes. The largest district in the state, Mesa Unified, responded with particular vigor. Mesa sends policy staff members to the state department of education to study charter proposals, checking up on the competition. Since 1996, Mesa has conducted customer-service training for staff, developed a “Red Carpet Treatment” for reintegrating charter parents into district schools, and expanded all-day kindergarten. Mesa officials initially denied, but later confirmed, that their actions were motivated by competition from charters.
• Advertising. The Mesa, Kyrene, Tempe, and Madison school districts are among those with active advertising campaigns aimed at enticing students from charter schools and from other school districts. All the district officials interviewed felt that charter schools had forced district schools to do a better job of communicating their strengths to the public. As one put it: “In some cases the charters are terrific. In other cases there is not a lot of substance but the advertising is there. It may be that we in the [district] schools have substance but are not very good at advertising. Maybe now we will get better at it.” A conservative policymaker claims that district schools “started out advertising their wonderfully high self-esteem and ability to deal with all diversities of people in an open and collegial way for the benefit of mankind.” However, in response to competition from charter schools, “Now they’re talking about math. Now they’re talking about phonics. Now they’re talking about reading ability.”
• Providing new curriculum choices. Many school districts facing competition have opened magnet schools, district-sponsored charter schools, or gifted programs in part to compete with charter schools. Often these offer either Montessori or “back-to-basics” curricula, the latter including the Benjamin Franklin program. Mesa had created a Ben Franklin magnet school to respond to the requests of many parents for a more traditional curriculum, but when demand for the school far exceeded the number of available desks, the district refused to expand the program. That’s when parents banded together to form a charter school. Now the district has changed its tune regarding expansion. In the coming decade Mesa plans to open five Benjamin Franklin magnet schools to compete with Benjamin Franklin charter schools. In Mesa, Queen Creek, and several smaller districts around the state, the spread of charters forced district schools to conduct in-service teacher training in phonics or Saxon math, curricula that local charter schools were providing. Queen Creek won back more than a third of the students lost to a charter school when it emphasized phonics and changed district leadership. The operator of an arts academy suggested that her charter served to “protect arts programs all over the district.”
• Changing administrators. I conducted fieldwork in four small, relatively isolated districts that lost more than 10 percent of their enrollment to charter schools. Three of the four districts changed school superintendents in the four-year study period, and the fourth nearly did so-a high level of turnover for rural districts. All four districts changed some of their principals, partly in response to charter competition.
• Undermining the competition. Charter operators insist that some districts compete using unethical tactics. One operator claims that district officials spread rumors that he is racist, leading many minority applicants to stay in district schools. Another complained about district officials’ alleging he was teaching religion. Some district teachers who decided to teach at or start charter schools reported being ostracized by former colleagues-one says he was slapped and called a traitor. Zoning is also an issue. In 1998 the Tucson city council passed a bill requiring zoning hearings before schools of small size (meaning charters) can open. In 1999, Maricopa County, the state’s largest, imposed zoning restrictions on charter and private schools, but not on district schools, as if a school of 100 students causes more disruption than a school of 1,000. Cities are often reluctant to approve expansion plans for charter schools, and city administrators in Phoenix and Gilbert have been accused of harassing charter operators, in one case even issuing press releases about nonexistent violations of the fire code. In the face of pressure from local district supporters, two housing developments that had planned to provide land for charter schools instead gave it to district schools. Several charter operators complained that district schools told certain students they were expelled and would have to report to charters. This happened just after the 100-day count on which state funding is based, but before standardized testing season.
• Assimilation. At least two districts have made overtures to buy out and absorb their more popular and conventional charter competitors, while ignoring the more exotic charters and those for at-risk students. In the long run, this may prove a popular strategy for districts with the resources to pursue it.
Demand and Supply
Critics of school choice have advanced several arguments that are contradicted by Arizona’s experience with competition. First is the claim that school choice is unnecessary since most Americans are pleased with their public schools. For example, Berliner and Biddle state that it is a “canard that American parents are generally dissatisfied with public schools.” They call demands for education reform a “manufactured crisis” created by influential “reactionary voices” from “the Far Right, the Religious Right, and Neoconservatives.”
In the fall of 1995, 55 charter campuses opened in Arizona, with approximately 7,500 students. In 1996 there were 119 campuses, rising to 222 in 1997. By the fall of 2001 about 431 charter campuses were serving roughly 61,000 students. The charter sector has added about 1 percent of market share (5,000 to 10,000 students) annually. The Arizona Department of Education reports that by their second year, most charter schools have waiting lists. Even if the majority of parents are reasonably happy with their public schools, a growing number are clearly interested in alternatives. As former Arizona superintendent of public instruction Lisa Graham Keegan said:
It takes a great deal of courage for parents to enroll their children in a new, untested charter school when there is a nearby district school whose program is a known quality. Given a choice, relatively few parents have absolute loyalty to the school their child is in right now. There is a huge market for schools that would address kids’ needs in different ways.
Critics are of course right to note that demand for schools does not develop overnight, but this is very different from saying there is no demand for alternatives. Sometimes parents don’t know that they want something different until they see it.
Another popular argument among critics of school choice is that there aren’t enough spaces in schools of choice to absorb all the students interested in leaving traditional public schools (notice how the critiques of school choice tend to cancel one another out). Yet Arizona clearly shows that a range of providers are eager to open new schools if given the opportunity. About 25 percent of Arizona’s charter schools are operated by former district school administrators; 25 percent are run by for-profit firms; 23 percent by former district school teachers; and 23 percent by social workers. A further 13 percent converted from private school status, and 11 percent were started by parents, professors, and charter or private school teachers. (The total is more than 100 percent due to overlap among categories.) Educators and social workers, rather than for-profit management firms like Edison Schools, dominate the Arizona market, most likely because per-pupil funding in Arizona is too low to attract the for-profits, in contrast to that in other states, like Massachusetts and Michigan. As of the fall of 1999, 162 of 216 charter-school operators were running a single campus; only 10 ran 5 or more, signaling a grassroots movement driven mainly by local educators and parents, not distant management companies.
Parents’ reasons for withdrawing their children from public schools are one of the major flashpoints in the school choice debate. Supporters of choice claim that parents look mainly for the best academic opportunity for their children; critics charge that parents will just as often search for a school on the basis of ethnic, religious, or ideological preferences, the quality of the sports program, or how blue the student body’s blood is.
Parents’ motives are difficult to ascertain, but their choice of schools is clearly not being driven by racism in Arizona. While some charter schools emphasize Hispanic, Native American, or African-American culture, so do certain district schools. Casey Cobb of the University of New Hampshire and Gene Glass of Arizona State University found that Phoenix-area charter schools are, on average, 11 percent whiter than the nearest district school. But most charter schools draw from numerous district schools. To compare charter schools with the nearest district school is particularly misleading, since, in order to save money, charters often locate in low-rent areas, but draw their students from surrounding areas. When charters are compared with their entire district, and those charters that converted from private-school status are excluded from the analysis, charter schools turn out to be, on average, 2 percent less white than district schools. Of course, some private schools have converted to charter status while maintaining their old (mainly white) student bodies.
In surveys conducted in 1998 by Arizona State University’s nonpartisan Morrison Institute, the vast majority of parents said they chose charter schools for academic reasons. Parents most often based their decisions on a school’s curricula, teaching methods, class size, and academic standards. Most parents believe that their children are doing better in charter schools. Likewise, a 2000 survey sponsored by the Arizona State Board for Charter Schools reported that academics clearly are the primary concern of charter school parents. The same study found that 61 percent of Arizona charter parents gave their schools an A+ or an A. Comparable surveys of Arizona parents with children in traditional public schools found only 38 percent grading their schools A+ or A.
Fieldwork and quantitative analyses conducted by me and Scott Milliman of James Madison University find that when superintendents and principals are out of touch with significant numbers of teachers and parents, district schools experience high rates of defection to charter schools. For example, a survey of district elementary-school teachers found a strong, negative correlation between teacher morale (as measured by their response to the statement, “I feel I am treated as a valued employee) and the market share of charter elementary schools. In other words, as teacher morale plummets, charter school enrollment rises.
What alternatives do charter schools offer? On the secondary level, as of the spring of 2001, 64 percent of charter campuses were for at-risk students. Local district schools often welcome the arrival of this type of charter school. As one Arizona Department of Education official put it, such charters allow districts “to avoid a dropout, which is a statistic the districts do not want.” Charter high schools are often started by social workers wishing to provide small communities for troubled students. As of 1997 their median enrollment was 65 students, compared with 871 students in district schools.
Relatively few charter secondary schools serve “mainstream” students. At the high-school level, parents demand science labs, football stadiums, and swimming pools, amenities that few charters can afford. At the elementary- and middle-school levels, parents want academic programs and a safe environment, and here charter schools are able to compete for conventional students. Robert Stout of Arizona State University and Gregg Garn of the University of Oklahoma found that 47 percent of Arizona charter elementary-school students attend content-centered schools, usually advertising “back to basics” or Core Knowledge approaches that are considered old-fashioned by district schools. Thirty-five percent attend child-centered schools, usually Montessori-based. The remaining schools offer a wide range of approaches, including bilingual, arts-based, and Waldorf programs.
The most striking difference between charter elementary and district elementary schools is their size. Charter schools, with a median size of 110 students, tend to provide small learning communities compared with district elementary schools, whose median size is 590. Charter school principals typically-and intentionally-know all the parents. One teacher-operator said that she initially wanted a school of 200 children, but after a year decided that “130 was actually about right.” Another teacher-operator wants enrollment to grow, but by adding one small campus at a time every other year rather than by expanding existing schools. The idea is to maintain school quality and to provide small communities where each child can be a “big fish in a small pond.” This is difficult for district schools, which are encouraged to build large schools because of the incentives contained in state funding rules.
Nearly everything done in a charter school has been done at some time at some district school. Yet it is cold comfort to teachers and parents who desire Core Knowledge or Montessori education to know that in some distant county, a district school has what they want. Arizona’s free market allows teachers and parents to find or create schools that suit their preferences without moving hundreds of miles, going to expensive private schools, or spending ten years lobbying the school board.
In the clunky, incremental manner of real-world social systems, school choice is improving public education in Arizona. It provides outlets for teachers and parents who are unhappy with existing district schools, and it forces district schools to improve their outreach, provide popular curricular options, and, in some cases, change leadership. It encourages talented teacher-entrepreneurs to realize their dreams, keeping them in the field when they might otherwise have left.
-Robert Maranto is an assistant professor of political science at Villanova University and coeditor, with Scott Milliman, Frederick Hess, and April Gresham, of School Choice in the Real World: Lessons from Arizona Charter Schools (Westview, 2001).
Last updated July 19, 2006