In educating its young, the world’s greatest industrial power cannot keep up with its competition. Consider the findings of the Third International Mathematics and Science Study, which in 1995 tested over a half-million students in 41 countries. Although at the fourth-grade level American children scored passably well, behind the Japanese and Koreans but ahead of the British and Norwegians, by the eighth grade they ranked last among the major industrial nations. Compared with their peers, the longer our children stay in school, the farther behind they fall.
Nor does public education in the U.S. fare any better when considered on its own terms. Inflation-adjusted spending on education in the U.S. increased by 50 percent between 1974 and 1991—it is now at an all-time high of well over $300 billion a year—but this did virtually nothing to boost test scores. Over roughly the same period, according to the National Assessment of Educational Progress (and despite recent news reports to the contrary), elementary-school students showed only slight improvement in math and reading, while those of high-school age made no gains in these subjects and actually slipped in science and reading.
As bad as things look in the round, moreover, they are much worse in schools serving the low-income, minority populations in our inner cities. An analysis published by the New York Times last January, for example, found that New York City students performed worse than their peers in the rest of the state, even after controlling for race and income, and that the differences grew over time. In urban schools, bureaucracy, unionization, political corruption, and social dislocation have conspired in an especially cruel way to create a system seemingly impervious to serious, long-term reform.
In short, if public education in the United States were a business, investors would long since have put their money elsewhere. And in fact something along these lines is now beginning to happen. After well over a decade of distress signals and exhortation, touched off by the 1983 commission that first officially recognized the “rising tide of mediocrity” in American schools, a movement is growing to reconfigure the existing model of public education. In many cities and most states, and over the opposition of every interest group with a stake in the educational status quo, the new byword of reform is “choice.”
At its most fundamental, choice simply entails giving parents and students more options when it comes to publicly supported schooling. Such options, the reasoning goes, will create competitive pressures where today there are only public monopolies with all their attendant problems. Under the discipline of the marketplace, schools will either satisfy their customers or close shop.
So far, nothing quite so radical has happened: choice today remains an experiment rather than a full-fledged alternative. In practice, and at its most modest, the policy has meant allowing public-school students to pick from among schools within an existing district or, sometimes, from those in neighboring districts. In Massachusetts, for example, over 6,700 students from more than 200 districts have chosen to attend a school in another district.
A more ambitious incarnation of choice can be found in charter schools, independent public institutions that are allowed to operate with a relatively free hand on matters like curricula and union rules in exchange for meeting certain standards. The laws authorizing such schools vary a great deal from state to state, with some much more burdensome than others. In the six years since the first of the charter schools was launched, they have been a growth industry: nearly 700 of them, with well over 100,000 students, opened their doors in some twenty states this fall.
Both these forms of experimentation have much to recommend them. According to the Harvard economist Caroline Minter Hoxby, where interdistrict competition among public schools is more prevalent, school taxes are lower, students take more academic courses, and they learn more. And the early reports from charter schools, too, are upbeat.
Most proponents of choice, however, look beyond these modest variations on the traditional public-school system. The true advantages of an educational marketplace can be realized, they insist, only where parents and students are free to spend their school dollars wherever they see fit, including in private and, should they so desire, religious institutions.
Over 30 such programs, serving more than 13,000 students, have in fact sprung into existence since 1990, with financial support coming from both states and private donors. All operate through targeted subsidies that can be used only for tuition and other school expenses. The most common devices for accomplishing this are vouchers and scholarships, but tax deductions and credits are also being used, as under a new law in Minnesota.
For the most part, these more far-reaching experiments have been confined to our inner cities, and the great bulk of the participants comes from minority, low-income, single-parent families. This is no accident: however strongly the urban education establishment may resist change, parents in cities are desperate to rescue their children from dead-end schools. According to a recent survey by the Stanford political scientist Terry Moe, fully 79 percent of the inner-city poor favor vouchers, whereas the comparable figure for whites living in more advantaged communities is (a still-impressive) 59 percent. Another survey, by the Joint Center for Political and Economic Studies, found school choice receiving the support of 65 percent of Hispanics and 57 percent of African-Americans, as against 47 percent of whites.
It is hardly surprising that low-income, inner-city families should see school choice as a way to win some of the latitude that their more affluent neighbors already enjoy. But are they also right to think that such programs can turn around the educational fortunes of their children?
Though we still have much to learn about school choice, what we know so far is encouraging. After reviewing the latest research on the largest voucher experiments from around the country, I count myself among the cautious optimists.
The programs now being studied in some detail—in Cleveland, Indianapolis, Milwaukee, New York, and San Antonio—represent just about every possible permutation of choice. Some are funded publicly, others privately. Rules vary among them about whether parents can throw in financial support of their own, and procedures for admission differ as well, with some programs using a lottery and others assigning places on a first-come, first-served basis. New York’s School Choice Scholarships Program allows families to select from as many as 250 parochial and secular schools, while Milwaukee’s state-supported program limits the choice to twenty nonreligious schools. A second effort in Milwaukee, the privately run Partners Advancing Values in Education, offers help to nearly 4,500 students, while San Antonio’s Children’s Educational Opportunity program reaches just under 1,000 children. Funding ranges from the $800 subsidy provided by Indianapolis’s Educational Choice Charitable Trust to the $4,700 scholarship available under Milwaukee’s state-supported program.
How do these varied programs measure up? Four different studies have found that families participating in school choice are more satisfied in virtually every way with their children’s education. In San Antonio, for example, parents using vouchers were much more likely to say good things about discipline in their schools, and to be content with how much their children were learning, than were parents with children in the public system. Similar results turned up in Cleveland, Indianapolis, and Milwaukee.
In yet another sign of satisfaction, a strikingly greater proportion of choice students sticks with their schools. In Milwaukee, for example, it is normal for 20 percent of public elementary-school students to change schools in the course of the academic year, and for fully 3 5 percent to find their way to another school come fall. By contrast, only 4 to 6 percent of elementary-school students in Milwaukee’s state-funded school-choice program leave during the middle of the year, while 77 percent return to the same school after the summer.
Under school choice, teachers also find it far easier to teach. In San Antonio, for instance, where six out of ten middle-school teachers complain that their classroom environment is not conducive to learning, only one out of ten teachers in the school-choice program voices a similar complaint. Unsurprisingly, teachers in these private schools are also far more likely to assign homework and to expect their students to exceed course objectives by the end of the year.
Such positive responses would mean little if students were not learning more. But, by most indications, they are. My colleagues and I have found strong gains, especially in math, among first-year students in Cleveland’s school-choice program; in the San Antonio program, in marked contrast to the declining performance of a public-school comparison group, scores went up over a three-year period between 1991 and 1994.
The most serious problem in evaluating these programs lies in distinguishing the effect of the schools themselves from other, nonacademic factors. This difficulty has provided ammunition to critics like Herbert Grover, a former state school superintendent in Wisconsin, who claims that students in school-choice programs do better because of the “initial advantages they enjoy from their parents.” To some extent, the charge is easily answered: precisely because such programs tend to be limited to low-income, inner-city families, the demographic profile of the students usually matches that of their public-school peers. That being so, success cannot be rationalized by pointing to their parents’ extra financial resources or level of education.
A related “selection” problem raised by skeptics is harder to get around. Because the privately funded programs in Indianapolis, San Antonio, and Milwaukee admit students on a first-come, first-served basis, places necessarily go to families that are quicker, more clever, and better connected. No amount of statistical tinkering, it is fair to claim, can ever fully correct for the encouragement such parents give their children.
But this problem, too, vanishes in those cases where scholarship winners are chosen by lottery, thereby creating two groups of students identical in every relevant way except for the fact that the winners had their names randomly drawn from a hat. Of the school-choice programs that have been operating under such a system, only one, the state-funded program in Milwaukee, has produced test results so far, and these numbers have been at the center of a swirling controversy.
A research team headed by John Witte of the University of Wisconsin took a first look at test results from Milwaukee and reported that students with vouchers got no discernible boost from attending private schools: their math and reading scores were no different from those of the control group. But when my colleague Jay P. Greene of the University of Texas and I looked at Witte’s data, we quickly realized he had drawn the wrong comparison. Instead of measuring students in the school-choice program against those who had lost out in the lottery, he had measured them against a sample of Milwaukee public-school students from much more advantaged backgrounds who had performed significantly better on earlier tests. In brief, Witte had compared apples to oranges.
Returning to the raw data from Milwaukee, and focusing on the right two groups, Greene and I discovered that, after two years in the school-choice program, students were just a small step ahead of their peers. In a short time, however, they began to pull away, and by the end of the fourth year they were performing 11 percentile points better in reading and 6 points better in math. Modest though these results may appear at first glance, they are nevertheless impressive. If such progress were duplicated among minorities nationwide, it would reduce by as much as half the current disparity between their test scores and those of whites.
New York’s School Choice Scholarships Foundation, which has enrolled its first students this fall, will offer a still better opportunity for gauging the effects of choice. To the astonishment of the program’s organizers, over 20,000 applicants took part in the New York lottery. In addition to the 1,300 scholarship winners, a control group of another 1,300 is being selected from the same pool. Over the next four years, I will be working with David Myers of the firm Mathematica to see if New York gets the same pay-off from school choice that Milwaukee has so far earned.
As most supporters of school choice acknowledge, it is premature to talk about the wholesale dismantling of our 150-year-old public-school system, where some 90 percent of American students still receive instruction. School choice has not yet had a chance to prove itself. It needs to be undertaken gradually, experimentally, and with a focus on the places that are in the greatest need.
To opponents of the experiment, however, even this is too much. The many interests served by the current system have done everything in their power to quash the modest pilot programs and to keep others from getting off the ground. To take but one illustrative episode: in an unguarded moment during last year’s presidential debate, Bill Clinton, ever on the lookout for an issue popular with voters, declared, “If a local school district in Cleveland or any place else wants to have a private-school choice plan like Milwaukee did, let them have at it.” The next day, presidential advisers scrambled to reassure the indignant national teachers’ unions that the President had been misunderstood; he held firm in his unqualified opposition to public funding for any scheme of private-school choice.
Justifying such resistance has become harder in the face of clear evidence that progress is being made in these schools. Deprived of the argument that school choice will do nothing for academic achievement, opponents have resorted to other grounds for not giving it a chance.
The first and simplest objection is a constitutional one. According to the critics, publicly-funded vouchers or scholarships that can wind up in the hands of parochial schools (Catholic, Jewish, Episcopalian, or whatever) amount to an “establishment” of religion on the part of government and hence are in flagrant violation of the First Amendment.
Next, and in a related vein, opponents of school choice maintain that it threatens our deepest democratic values. The schools in a country like ours teach more than the three R’s; they also aim to instruct students in their responsibilities as citizens in a free, diverse society—a task to which the public schools have been traditionally and uniquely well-suited. In this respect, school choice may end up acting as a mighty solvent of social solidarity, encouraging the further balkanization of American life and subsidizing educational enterprises run by who-knows-what sorts of extremists—from neo-Nazis and black nationalists to witches and ecozealots. As Michael Kelly, then editor of the New Republic, put it:
Public money is shared money, and it is to be used for the furtherance of shared values, in the interests of e pluribus unum. Charter schools and their like . . . take from the pluribus to destroy the unum.
Finally, foes of school choice ask, what about those left behind? Choice programs, they assert, rob the public schools of their brightest students—students indispensable to the academic progress of their classmates. In so doing, they shift public attention away from the urgency of our educational problems, tempting policy-makers (in the words of the American Federation of Teachers) with “promises to ‘save’ a handful of students” instead of doing what needs to be done to improve the achievement “of all of our youngsters.”
Each of these arguments is in fact rather thin. To begin with the supposed constitutional problem: the Supreme Court, it is becoming increasingly clear, has no objection in principle to letting public dollars go to religious schools as part of a wider educational program and so long as neither a particular faith nor religion itself is given a special advantage. In Mueller v. Allen (1983), for example, the Court held that an educational tax deduction passed constitutional muster even though parents could use it for secular or parochial schooling. And just this past June, in Agostini v. Felton, the Court decided that it did not violate the First Amendment for state-supported teachers to provide parochial-school students with the same remedial instruction available in the public schools. The key in this case (as in Mueller) was that benefits were being provided “without regard to the sectarian [or] nonsectarian . . . nature of the institution.”1
The constitutional objections to school choice also fly in the face of long-accepted policy when it comes to public support for higher education. Since 1973, for example, federal Pell grants have been available to low-income college students going either to secular or religious schools. More recently, Congress passed and President Clinton signed into law a tax credit that can be used at any secular or religious college, from San Francisco State to Notre Dame or Concordia. By what reasoning is comparable aid to younger children to be deemed unconstitutional?
As for the anti-democratic threat posed by private schools, there is no evidence that such a thing exists. To the contrary, private schools seem to be doing a much better job than public schools in teaching the virtues of American citizenship. According to a recent study, based on data from the U.S. Department of Education, private-school students, far from shunning involvement with the wider community, are considerably more public-spirited than their public-school counterparts, volunteering more often and more regularly in community affairs. And far from sowing division in society, not only are private schools more racially integrated than neighboring public schools, but their students enjoy more cross-racial friendships and engage in fewer race-related fights.
True, public funding could indeed call into being a new and more destructive strain of private or quasi-private school, and has already done so in a number of places where choice has been insufficiently regulated. But the answer is to see to it that taxpayer dollars do not flow to every self-styled educator who sets up shop; appropriate financial and educational standards are a must. Public oversight, however, is not the same thing as public operation; there is no reason for the government to become a dairy, so to speak, just because it regulates milk.
And what of those children left behind in the public schools? Will they not suffer? In the first place, white, middle-class students have already fled in large numbers to the suburbs or the private schools. Bright but less-advantaged students have similarly deserted the most desperate urban schools, opting for the magnet programs and classes for the gifted that are now available in big-city systems. The sad fact is that many of the students who remain in the public schools have already been left behind.
But more to the point, perhaps, are two other considerations. One is that most programs of school choice do not, as a matter of course, attract the best and the brightest students, for the simple reason that most parents are reluctant to switch schools when their children are doing well in them. The other is that even if some better students do decamp for private schools, the effect on those remaining is unlikely to be all that serious. Research confirms what we know from common sense: children learn primarily from parents and teachers, not from other children. In the expert view of sociologists Christopher Jencks and Susan Mayer, the academic impact of peer groups is small, inconsistent, and by no means straightforward: in some cases, students thrive by absorbing facts and ideas from their more capable peers, but in others they get positively discouraged when they cannot keep up.
Finally, far from drawing attention away from our inner-city schools, school choice subjects them to just the sort of scrutiny they sorely need. Why indeed do they spend their money so inefficiently? What is it, precisely, that prevents them from establishing standards for achievement and behavior as high as those of private schools? Why do their students perform so poorly year after year? One can only conclude that defenders of our failing public schools are alarmed by school choice because it serves as a ferocious standing indictment of what they themselves have wrought.
In the imaginings of its most ardent supporters, by contrast, school choice, if widely implemented, could go a long way toward transforming our cities. Existing private schools would expand, and new ones would be called into existence. The public schools, finally feeling the heat of competition, would begin to address their most serious problems. Young parents with school-age children would forgo moving to the suburbs. Other parents would give up their suburban homes to live closer to their work in the city. As test scores rose among low-income minority students, they would become far more employable upon graduation, sparking a return of businesses to the inner city. Property values there would begin to rise, and racial integration would take an upward swing.
It is a lot to expect. But if even part of this rosiest of scenarios is plausible, school choice is an experiment we cannot afford to pass up.
1 Lower state courts in Ohio and Wisconsin, relying in part on their own state constitutions, have ruled against publicly-supported school-choice programs that include religious schools. Both cases are still being litigated, however, and may well be affected by the Supreme Court’s most recent ruling.