IS IT UNCONSTITUTIONAL TO PROHIBIT FAITH-BASED SCHOOLS FROM BECOMING CHARTER SCHOOLS?

Abstract

This article argues that it is unconstitutional for state charter school programs to preclude faith-based schools from obtaining charters. The first section describes the “school choice” movement of the past fifty years, situating charter schools in that movement. The current state of play of school choice is documented and the roles of charter schools, private schools (primarily faith-based schools), and public school choice options are elaborated. The second section argues that based on the current state of the law it should not be unconstitutional, under the First Amendment's Establishment Clause, for states to elect to make faith-based schools eligible for charters, and, therefore, the current practice of formal discrimination on the basis of religion against families and school founders who want faith-based charter schools should be deemed unconstitutional by the US Supreme Court. Put differently, this is not the sort of issue in which the “play in the joints” between the Free Exercise and Establishment Clauses should apply so as to give states the option of restricting charter schools to secular schools.

Introduction

The legal issue addressed in this article is whether state laws that preclude religious schools from becoming “charter schools” violate the U.S. Constitution. Currently, all of the forty-three states that have embraced the charter school movement restrict charter schools (described in detail below) to nonsectarian schools. ¹ On the face of it, this deliberate discrimination on the basis of religion seems legally suspect.

But two alternative principles might justify the restriction. The first is that to publicly fund charter schools that are parochial schools might itself be an unconstitutional “establishment” of religion in violation of the First Amendment. The second is that the broad notion of the “separation of church and state” might give states the legal discretion to restrict charter schools to non-parochial schools.

In contrast with others who are supportive of faith-based schools and have examined the issue, ² I take the bold position that the current discrimination against faith-based charter school applicants (and families seeking to send their children to such schools) is unconstitutional.

Before discussing the legal arguments, I present a general picture of the school choice movement over the past fifty years. This picture is critical to my constitutional argument because charter schools arose out of and are deeply embedded in the broad push for “family choice” in American education. ³ Charter schools are very different from, say, alternative and magnet schools created by school districts to serve the district's purposes. Charter schools are (almost entirely) created and run by private actors offering families something other than the regular public school for their children. But the preclusion of faith-based schools from obtaining charters leaves families seeking a religious education particularly discriminated against in their desires to match their values with their children's schooling.

Many Americans favor empowering working-class and poor families with the ability to choose how their children are educated, just as well-to-do families have long been able to do so by either paying for wholly private education or moving to high-priced communities with well-resourced public schools. From this perspective, privately run charter schools can be a sound pathway to that empowerment. And indeed, about three million children are currently enrolled in such schools.

But for families of modest means wanting a religious education for their children, the charter school option is not presently available. Those families are typically forced to (a) scrape together a little money to pay tuition at a very underfunded private religious school; (b) rely on the charity of more established religious schools, sometimes run by a faith other than theirs; (c) home school their children at a great burden to the family; (d) select a charter school that is not really what they prefer; or (e) give up and send their children to a public school where the parents now often find themselves battling with school officials and other parents ⁴ over curriculum content, reading material, dress codes, school celebrations, and more that run contrary to their religious values.

This article is not centrally about why, as a moral matter, or as an expedient matter, or even as a matter of best educating children a family's preference for a faith-based charter school should be respected. ⁵ It is rather, first, a portrayal of the school choice movement and the place of charter schools in that movement, and then second a legal argument as to why the U.S. Supreme Court might very well decide that de jure exclusion of faith-based schools from the charter school schemes of all the states with charter schools is unconstitutional.

School Choice: The State of Play

The School Choice Movement

Starting in the 1960s a range of scholars and other advocates began arguing for government-funded “school choice.” ⁶ They generally depicted the public school system as involuntarily assigning children to attend a specific school, usually the one located nearest to where they lived. The critics often portrayed the system as resting on a myth of the “common school” in which all public schools are understood to be essentially the same thereby making it largely irrelevant which one any child attended. ⁷ But, of course, in the real world public schools have long differed from one another. In terms of the basic education they deliver, some are much better than others. Moreover, the values taught in public schools have long varied from district to district, school to school, and classroom to classroom. On top of that, children differ in their needs and in their parents’ desires for them so that even if any specific school might be well suited for some children, it might not be for others. The upshot, in this stylized presentation of things, is that while some parents were quite happy with what their children were given, others were not.

In a world in which children were assigned to public schools based on their address, the only ways unhappy parents could “choose” a school they preferred for their child were either to move into an attendance catchment area of a public school they liked or to opt out of the public school system by sending their child to a tuition-charging private school. In the 1960s families with about 12 percent of America's schoolchildren did the latter. ⁸ Obviously, both of those “choice” options were practically far more plausible for well-to-do families than they were for most lower-income families, especially when many of the most highly desired public schools were located in suburban neighborhoods with substantially higher than average housing costs.

For Catholic families wanting a religious education for their children, however, the tuition burden at parochial schools at that time tended to be light, perhaps most importantly because so many of the teachers in Catholic schools were nominally paid members of religious orders, and also because, at least at the elementary school level, local parishioners made charitable contributions that helped support the parish school. At that time the private school sector in the United States was overwhelmingly dominated by Catholic schools, whose roots go back to the nineteenth century, when Catholic leaders, viewing the “public” schools as Protestant schools, created an elaborate system of schools for Catholic children. ⁹

Opening Up the Private Sector

Many of the critics in the 1960s and 1970s who helped generate the “school choice” movement looked to the private sector to remedy the regime of restricted choice. The central vision of most of them was that if government would offer to fund families, instead of just funding schools, then more families would have a choice between what the public school system offered to them and what would be provided by the private sector. Although the language used by some critics envisioned a system of school “scholarships,” the label that stuck to this policy approach is school “vouchers” largely because that was the term used by the conservative Nobel Prize–winning economist Milton Friedman, who initially prominently championed the idea in 1962 in his book Capitalism and Freedom. ¹⁰

Friedman actually wanted to do away with public schools, privatizing the system entirely. Moreover, he wanted to reduce public funding for education by giving all families vouchers worth much less than was then being spent on public education. He was motivated largely by an ideological commitment to capitalism and competition, confidently predicting that education could be delivered both cheaper and better via the private market. Friedman also noted that while there were public benefits that flowed from having an educated population (thereby justifying some public subsidy), being educated also conferred very substantial private benefits to students, who should pay for those benefits (or in this case their parents should). Under Friedman's approach, there would be a market in elementary and secondary education that would become much more like the markets for food or clothing. And the private sector would become flooded with new schools, many of which, probably most of which, would not be religious schools (especially as existing public schools became privatized).

During the 1960s, however, arguments for school choice and school vouchers were initially tainted by the South's use of these strategies as part of what came to be called the “massive resistance” to school desegregation. ¹¹

Moreover, simultaneously, in the Northeast many Catholic schools found themselves facing growing financial difficulties, and legislatures in several states that had substantial Catholic-school-going populations enacted measures designed to bail out the Catholic schools. These included providing money directly to religious schools, to their teachers, to pay for curriculum materials that were then presented to the schools, and to the parents in the form of small value vouchers. But even the latter were not aimed at financially helping parents. Rather, the legislative assumption was that families could then pay somewhat higher tuition thereby allowing the schools to shore up their finances. Put differently, unlike Friedman, supporters of these aid schemes did not intend to expand choice to more households. ¹²

In any event, the U.S. Supreme Court stepped in and blocked the southern resistance strategy as incompatible with Brown v. Board of Education ¹³ and struck down most of the Catholic school prop-up schemes as invalid aid to religion in violation of the “Establishment Clause” (discussed below). ¹⁴

Despite these early setbacks, a third development in that era is reflected in the writings of a small set of liberal school choice advocates. ¹⁵ They believed that the public schools, especially in urban areas, were failing many children from low-income households (many of whom were not white). Their thinking was that if the government offered substantial-value scholarships (or vouchers) to children in those families, this would create the possibility of enhanced educational opportunities where they were most needed. Absent the creation of new private secular schools, advocates realized that most of these options would be religious schools. This did not trouble them.

Indeed, by the 1970s observers began to point to the fact that many urban Catholic schools already were no longer catering to white Catholics (so many of whom had moved to the suburbs) but instead to low-income African Americans, many of whom were Protestants. Black families who made these choices (typically paying little or sometimes no tuition) were getting something that they thought was better for their children, even if it often meant exposing them to a different form of Christianity. But these Catholic schools could afford to take on only so many low-income families. Moreover, those who viewed this development as a strong positive for school choice feared that in many cities a large share of these schools would eventually have to close as Catholic parishes and bishops had only so much money available for this sort of social justice project. ¹⁶

These liberal advocates of school vouchers for the poor viewed such a reform as way to help the poor via the non-public sector just as had been envisioned by the newly adopted food stamps program (in 1964) ¹⁷ and the newly created Medicaid program (in 1965). ¹⁸ Both of those programs, in effect, provided service-specific vouchers to low-income families.

Professor John Coons (my mentor) and I sought to occupy ground between Friedman and those focused only on the poor. ¹⁹ We were happy to support income-based scholarship plans aimed at the poor. But we also proposed a plan through which all families would be offered an option to go to schools other than public schools. In our proposal, the scholarships would be worth almost as much as was then being spent on public schools. Our universal scholarship plan assumed several regulatory features, however. Participating schools would have to either provide lottery access among applicants or make available a substantial share of their places for low-income families. Schools accepting vouchers would not be allowed to charge families extra tuition that would price out low-income families. Reasonable transportation assistance would have to be provided. Independent counselors trained to help families choose would have to be made available. Schools would have to provide due process rights to students in their charge. And while faith-based schools could participate in the plan, they could not compel students to profess a commitment to the faith of the school. ²⁰ We envisioned the creation of many new private schools, both faith-based and secular. Unlike Friedman, we assumed that public schools would continue as before albeit with fewer students as more families opted for vouchers. Friedman opposed all of our regulatory controls, and rather than setting the value of the scholarship as we did, at 85–90 percent of what was spent in public schools, he proposed 50 percent. ²¹

Despite powerful rhetoric from libertarian Friedmanites on the right, ²² for half a century proposals based on his vision had almost no political traction. Indeed, twice when his idea was put to a vote in California it was very soundly defeated. ²³ In 2015, however, Nevada adopted a broad voucher-like plan (termed an “educational savings account”) in the Friedman tradition, which could be a breakthrough for that camp if it is legally upheld. ²⁴ Similarly, over that same period Coons's and my universal scholarship plan with a heavy finger on the scale in favor of the poor has not been embraced anywhere in the United States (although it is, in effect, the regime in many other economically developed nations). ²⁵

By contrast, the liberal approach to enhanced private school choice for low-income families has made modest headway in a few states. Milwaukee, Cleveland, and the District of Columbia are home to the most well-known private school voucher programs, with Milwaukee having initiated the idea in 1990. ²⁶ The Cleveland plan has been expanded to the rest of Ohio, and newer, broadly similar, voucher plans have been more recently adopted in Indiana, Louisiana and North Carolina. ²⁷ Together by now these voucher plans serve more than 100,000 children who come from lower-income households. ²⁸

Voucher plans aimed at low-income families were politically adopted for the most part through the combined efforts of African American Democrats in cities and Republicans across the state (who generally favored a smaller role for government and often sought to reduce the budgets and power of teachers’ unions). ²⁹ The voucher plans for low-income families were attacked in court as violating the Establishment Clause, ³⁰ but in the Cleveland case (Zelman v. Simmons-Harris, discussed below) this legal argument was rejected by the U.S. Supreme Court even though most of the children using vouchers attend religious schools. ³¹

Because only a small share of low-income families attending public schools in cities like Cleveland, Milwaukee, and the District of Columbia are whites, these plans work very differently from the southern state schemes that had sought to enable more whites to flee to all-white segregation academies. So, too, unlike the plans providing modest aid designed to hold together the Catholic school system on the East Coast in the late 1960s and early 1970s, these vouchers-for-the-poor plans were clearly meant to facilitate the actual movement of children from public schools into private schools. Hence, this sort of reform can be sharply distinguished from the invalidated publicly funded school choice plans of prior years.

More recently, a new political strategy for public funding of the private school choice by lower-income families has come into play. ³² This is the tax-credit school scholarship plan pioneered in Arizona and now most robust in Florida. Under these schemes, taxpayers make contributions to nonprofit organizations that in turn help lower-income families pay for the private school education of their children. Donors are given a tax credit for their contribution (often a 100 percent credit, which makes their donation costless to them). The recipient organizations consolidate these contributions and award scholarships to be used at private schools.

In Florida in 2016–17 more than 90,000 children were receiving such scholarships in amounts of up to nearly $6,000 a year, for an aggregated scholarship total of more than $500 million that year. ³³ By now, nearly twenty states have such programs in operation and together they are serving approximately 250,000 children. ³⁴ Again, a huge majority of the beneficiaries attends private religious schools.

These tax-credit programs also have been attacked in court on a variety of grounds. Most importantly for our purposes, the U.S. Supreme Court refused to address the constitutionality of the Arizona plan, which those suing claimed violated the Establishment Clause, on the ground that plaintiff taxpayers had no standing to make such claims. ³⁵ This holding seemingly completely isolates such programs from legal attack in federal courts.

Adding together the number of children participating in the voucher programs and the tax-credit scholarship programs, that still amounts to less than 1 percent of the nation's school children. ³⁶

Opening Up the Public Sector

Over time more and more children have also been given options within the public school system. Hence, even as it was a bit mistaken fifty years ago to describe the public school system exclusively as a system of assigned schools, it is decidedly unnuanced to describe it that way today.

To be sure, many families remain stuck in a single school that is not serving their children well, with no practical alternative available to them. But many families using public school today do have choices, and they exercise them.

First, many school districts have created distinctive “magnet” or “alternative” schools that families may select. ³⁷ These schools were often initially created as part of efforts to reduce racial and ethnic isolation in public schools. In many places the motivation behind this option was to retain in urban districts white families who might otherwise flee to the suburbs.

Second, in recent years a number of urban districts have broken up large, unsuccessful public schools into a number of smaller schools, with families given choices among them, in effect creating several alternative schools inside of existing buildings. ³⁸

Third, some states adopted programs promoting attendance across school district lines. These plans have been motivated not only to promote racial balance but also to allow rural families to opt for better-resourced city high schools offering far greater access to Advanced Placement courses. In some places, children are allowed to enroll in out-of-district schools located where one parent works. Other sorts of inter-district transfer arrangements also exist (sometimes aboveboard, with sending and receiving districts approving the placement, and sometimes “illegally,” with families using false or dubious addresses in order to enroll their children in a preferred school). ³⁹

Fourth, some school districts have opened up their traditional neighborhood public schools to school choice. ⁴⁰ In its weak form, seats that are unfilled by children from the local catchment area become available to other children living in the district. In its strong from, there are no longer any catchment areas in the district, so that where one lives in the district gives one no priority access to any school. Instead, district schools are made available to all families. These latter regimes, in some places termed “controlled choice” programs, ⁴¹ often have been run with a finger on the scale designed explicitly or indirectly to promote school integration along racial or economic lines. Regardless of the motivations, these programs broadly embrace the “school choice” mantra but within the conventional public sector (for example, Berkeley, California). ⁴²

It is estimated that nationwide 10 to 15 percent of school-age children attend school by family choice through these four mechanisms. ⁴³

Charter Schools

In the 1970s and 1980s, some critics of the monolithic public school system called for the creation of a new type of school choice that they promoted as being within the public sector, by which they most importantly meant that these would be new, publicly funded schools that would be attended only by children whose families selected them. Ted Kolderie and Joe Nathan from Minnesota are often credited as playing the key roles in launching this idea. ⁴⁴ They called the new type of school they were promoting a “charter school.” ⁴⁵

They strongly supported the core principle of family choice. But because the “voucher” idea was being commandeered by those on the Right and touted mainly on the basis of the abstract benefits of capitalism, these new supporters of school choice sought to distance themselves from the Friedman wing.

Through their efforts, the Minnesota legislature embraced the idea, and the first charter school was launched in 1992. Since then the overwhelming majority of the states have followed suit so that by the 2015–16 school year there were almost 7,000 charter schools operating nationwide, located in forty-three states plus the District of Columbia. ⁴⁶ Together their enrollment was estimated to be about 2.9 million students (a substantial number, yet still just a bit shy of 6 percent of the nation's schoolchildren). ⁴⁷ About half of these charter school enrollees live in just five states: Arizona, California, Florida, Ohio, and Texas; and when New York, Michigan, and Wisconsin are added, these eight states together account for nearly two-thirds of the charter school students. ⁴⁸

Although they are often explicitly called “public charter schools” both in legislation ⁴⁹ and, at least some of the time, in common discourse, charter schools are clearly not the same as traditional public schools.

To be sure, charter schools have many “public” characteristics. First, they are publicly funded. Second, they may not charge tuition. Third, they must admit all who apply and if there are more applicants than spaces, admission must largely be by lottery. Fourth, because they are “chartered” by public bodies and in that way answerable to the public, these schools may be said to be publicly accountable.

On the other hand, as clearly envisioned by their inventors, most charter schools are in key respects “private” schools. They are usually owned by private, nonprofit, tax-exempt organizations and not by the local school district. ⁵⁰ While most charter schools are “free-standing,” a growing number now have contracts with independent companies who manage the schools. Today about 40 percent of students attending charter schools are enrolled in those operated by either nonprofit or for-profit management companies (with each type having about 20 percent of the market). ⁵¹

The mission of the charter school is privately set (such as a college prep school, a road to certain technical vocations, a fine arts school, and so on). The charter school curriculum is privately determined at least to the extent that fully private schools can control their curricula. The pedagogical style is up to the school to determine (does the school drill its students on skills, emphasize group inquiry, employ “master” teachers, and so on). The teachers are privately hired (in some places without necessarily having formal teaching credentials required of public school teachers).

When a charter school negotiates with a school district for its charter, it has private legal representation. Private lawyers represent the school in its legal dealings with others. People injured on charter school grounds claim damages under ordinary tort law not under government tort claims acts.

Yet, the “charter” (in effect, a contract) that the school signs with its public sponsor can, and sometimes does, mandate certain aspects of its operation, and if there is a serious breach of the terms of the charter the public sponsor can cancel its sponsorship and shut the school down (unless a different public sponsor is found). ⁵² Generally speaking, the teachers in such schools have rights most analogous to those in private schools. As to whether the teachers are unionized, this varies from state to state and charter school to charter school. Some states allow charter school teachers to belong to public school teacher unions (and retirement plans). But by 2012 it appears that only 7 percent of charter schools had unionized teachers. ⁵³ Moreover, in recent New York and Pennsylvania cases, the National Labor Relations Board ruled that charter schools are not public schools, but private corporations and subject to its jurisdiction (instead of being governed by state public employee collective bargaining laws). ⁵⁴

As for student rights, things are more complicated. Sometimes the terms of the school's charter (or perhaps state law) requires students to be given due process rights before being expelled or suspended, just like public schools. On the other hand, charter schools are permitted to, and do, demand behaviors from students, which, if not performed, can and sometimes do lead to dismissal. In this respect these charter schools are much more like private schools than traditional public schools. ⁵⁵

Initially some teachers’ unions and their leaders supported the idea of charter schools, ⁵⁶ and in the early days many envisioned that groups of existing public school teachers would be prominent creators of charter schools. Yet, early union support for charter schools was probably most importantly motivated as a strategy to politically cut off the voucher movement. ⁵⁷ Over time, teachers’ unions have become increasingly hostile to non-unionized charter schools, ⁵⁸ arguing that they are not really public schools and are harming our real public education system. Although President Obama and his long-term Secretary of Education Arne Duncan were strong supporters of charter schools, ⁵⁹ the two major national teachers’ unions denounced Duncan and the president for their support for what the unions now characterize as a corporate and marketplace driven approach to schooling. ⁶⁰

Most charter schools are chartered by the local school districts in which they are geographically located and tend to serve local children. But in some states other public bodies are allowed to sponsor charter schools. These most importantly include county boards of education, public universities, state boards of education, and specially created state-level chartering organizations. ⁶¹

State laws differ in the discretion that charter school authorizers may exercise in approving or turning down a charter applicant. In states with what charter school supporters call “strong” charter school laws, any applicant that meets the basic filing requirements for obtaining a charter must presumptively be granted one (subject only to having a coherent educational plan and an adequate business plan for the school). ⁶²

To be sure, some states have numerical caps on the number of charters that may be issued. But when the cap actually bites it appears that the typical practice is to charter qualified applicants in order of application with subsequent applicant schools put on hold until either the legislature raises the cap or existing charter schools drop out of the system creating new room under the cap. ⁶³

Chartering bodies officially are “gatekeepers” and generally view the fact that a proposed school has attracted families who promise to enroll their children by itself as insufficient to award a charter. The charterers also want the school actually to succeed in educating its pupils.

Hence, before giving out a charter they want to know about how the school will be run, what it will teach, and how. Yet, at the same time, one of the main ideas behind the charter school movement is that it will produce new ways of teaching and learning. Hence, many chartering bodies are eager to encourage experiments. Besides, as many charter schools have been formed in communities in which the public schools are seen to be badly failing many children, it is difficult to resist chartering a new school that has enthusiastic parental support.

Once chartered, these schools are generally free from all (or at least most) of the state regulations applicable to public schools; they are much more like private schools in this respect. ⁶⁴

Charters typically have a duration of a fixed number of years and then must be renewed, although in practice, apart from financial mismanagement (or worse) or manifest educational failure, most charter schools have had their charters readily renewed if their enrollment is robust.

To give a little feel for the variety of charter schools, I note that for the school year 2016–17, the Oakland Unified School District's Office of Charter Schools lists about three dozen schools chartered by the district. ⁶⁵ In addition there are about a half dozen more charter schools that are located in Oakland but chartered by the Alameda County Board of Education. ⁶⁶ In 2015–16 the schools chartered directly by the district enrolled nearly 12,000 students, as compared with 37,000 students enrolled in the public schools operated by the Oakland Unified School District, making Oakland a very robust charter school community. ⁶⁷

Some of the charter schools in Oakland are part of a national chain of charter schools; such as the KIPP Bridge Charter School. ⁶⁸ Seven of the schools are Aspire schools. ⁶⁹ Some of the schools, like Lighthouse and American Indian, hold separate charters for different grade levels (such as K–8 and 9–12). Many of the schools serve only limited grade levels, as do ASCEND K–8 and Oakland Unity 9–12. Many have different emphases, for example, Oakland School of the Arts (including dance, visual arts, and theater), Oakland Military Institute, Conservatory of Vocal/Instrumental Arts, Yu Ming School (bilingual), and Bay Area Technology School (BayTech) (science, technology, engineering, and math). Some charter schools emphasize drill and high test scores, like Oakland Charter Academy-Amethods; others emphasize portfolio assessment and securing four-year college acceptance for all graduates, like ARISE. There are pedagogical differences among the schools as well, for example, Urban Montessori.

Across the nation a number of charter schools have failed and voluntarily gone out of business (some in scandalous ways). Some have had their charters revoked (or not renewed) and could not find a new sponsor and hence had to close their doors. For example, about 200 charter schools in operation in 2012–13 (about 3 percent) did not reopen the next year. ⁷⁰ Charter school critics point to the disruption and the lack of appropriate educational progress often endured by students in charter schools that close. Charter supporters point to this churning as inevitable, and in a sense desirable, arguing that failing charter schools will and should disappear.

As in Oakland, elsewhere in the country a number of charter school operators now manage more than one school, often running a chain of schools in the region or even nationally. Like fast-food franchises or supermarket chains, this holds out the promise of continued rapid growth. And so far, the charter school system has rapidly grown and is far more robust than the publicly funded systems facilitating private school choice (that is, voucher plans and tax credit scholarship plans combined). Between 2006–7 and 2012–13 charter school enrollment nationwide roughly doubled. ⁷¹ At that rate it would take perhaps a decade from now for charter schools to capture 15 percent of the overall market.

One potential source of new charter schools could be existing private schools. But in many states, including California, a functioning private school may not simply convert to a charter school. ⁷² There are a number of possible justifications for this restriction, the most important of which is that the charter school movement was sold in the political process as a way of giving families using public schools a new choice option. And, on that assumption, the claim could further be made that the charter school plan merely shifts the way public money is being spent. In effect, dollars would follow the child from the traditional local public school to the charter school with no new cost to the public.

From the start, this description of new charter schools has been somewhat naïve on the funding side. First, there is the question of just how many dollars the pupils shifting to charter schools take with them. In practice, this is less than is spent per pupil in regular public schools, more than $3,000 per pupil less per year, according to a 2014 report. ⁷³ Hence, and especially because charter schools often have to pay rent or interest to cover the cost of their facilities (something not required of public school principals), those running charter schools have consistently claimed that they are shortchanged and that the public is inappropriately saving money via its inadequate funding of charter schools. ⁷⁴

Second, at the same time, conventional public school supporters often respond that charter schools are unfairly draining off public school money. The main issue here is what share of the costs in public schools is fixed and what share is variable. Unsurprisingly, public-school supporters claim that nearly all costs are fixed, arguing, for example, that losing two children from every grade saves schools virtually no money. Charter school supporters tend to argue the opposite, claiming that if, say, 350 children disappear from the public schools (that being the average enrollment in charter schools), this surely saves the salaries of many classroom teachers. The truth surely lies somewhere in between these claims.

Third, even in states where existing private schools cannot become charter schools, charter schools are drawing a significant share of their enrollment from families who either previously enrolled them in private schools or would have so enrolled them were the charter school option not available to them. Indeed, perhaps as many as one-third of the children attending would otherwise be enrolled in private schools were there no such thing as charter schools. ⁷⁵ These children come from families who prefer the private school to the conventional public school but prefer the charter school to either of the others. This influx of private-school students surely costs the public taxpayers money, but this is not what opponents of charter schools seem to have in mind when they talk of draining funds from public education.

The American charter school system turns out to be very much like the universal scholarship/voucher plan promoted by liberals like Jack Coons and me in the 1970s. ⁷⁶ The “choice” schools are overwhelmingly created through the initiative of private actors as we envisioned. The per-child funding of these schools is substantial (often approaching the 85 percent of public school spending we initially proposed). Access to participating schools is universal (by lottery if there are too many applicants, which was one of the options we endorsed). There is substantial public regulation of the sort we favored (important aspects of which are meant to help assure fair access by children from low-income families), and yet these privately managed schools have a great deal of autonomy. And as with our proposal, regular public schools are left in place.

Yet, there is one very large difference: no state currently allows a charter school to be a religious school (whereas under our universal voucher proposal, faith-based schools willing to accept our regulations could accept voucher-carrying students).

Private Faith-Based Schools Today

In 1960 Catholic schools overwhelmingly dominated private schools in the United States and enrolled more than five million students. ⁷⁷ Neither is true today.

In 2009–10 there were more than 20,000 faith-based schools in the United States, serving about 4.3 million children. ⁷⁸ About a third of those schools were Catholic schools and they together served just over half of all children enrolled in faith-based schools (that is, about 2.2 million). In short, Catholic school enrollment has declined by more than 50 percent in fifty years. Since 1990 alone, Catholic schools dropped in number from 8,700 to 7,400. ⁷⁹

In 2010, there were 4,300 Evangelical Protestant schools and 2,100 Lutheran (Missouri Synod) schools. ⁸⁰ In terms of enrollment, about 700,000 students were enrolled in Evangelical-Protestant schools, about 300,000 students were enrolled in Baptist schools, and about 220,000 in Jewish schools. ⁸¹ Most of the remaining students in faith-based schools attended a miscellany of Protestant schools, such as Lutheran, Episcopal, Seventh Day Adventist, Assembly of God, Presbyterian, and Methodist schools. ⁸² Just over 30,000 students were enrolled in Muslim/Islamic schools. ⁸³

Private nonreligious schools are a small share of the private school market, enrolling about a quarter of private school pupils, or 1.3 million of America's school children (that is, almost 2.5 percent of all school-age children). ⁸⁴ These schools tend to be fairly high priced, and therefore often cater primarily to financially well-off families who prefer to exercise their choice of school for their children this way rather than via their choice of residence.

In some states, these private nonreligious schools could become charter schools now if they wanted to although many would find that inconsistent with their financial model which requires higher tuition than the funding they would receive per pupil as charter schools, to say nothing of the loss of their control over admissions that would come from charter school status.

Some people predicted that the voucher programs in places like Milwaukee and Cleveland would stimulate the creation of many new private nonreligious schools in those jurisdictions—that is, schools that would essentially be funded by the vouchers. But this appears not to have happened, perhaps in large part because those wanting to start new nonreligious schools usually find it financially more attractive to become charter schools than voucher-accepting schools.

Some people claim that those attending faith-based schools are taught to be intolerant, especially of people of other faiths. This appears to be a misconception. Notwithstanding (or perhaps because of) the faith-based content of their educational programs, studies suggest that students in and graduates of faith-based schools are more tolerant (and more likely to engage in civic activities) than their public school counterparts. ⁸⁵

Nonetheless, for many liberals the expanding place in the American education system of private fundamentalist religious schools is distressing because of what they see as the broader “culture war” in the country. Families using these schools are often understood to be socially very conservative, often against abortion, homosexuality, and the like; they are seen as the same sort of people who are trying to reintroduce prayer into the public schools and to ban the teaching of evolution from public schools or to force the teaching of “intelligent design” at the same time. ⁸⁶ As a result, it is difficult for at least some liberals to favor empowering low-income families to take more control over their and their children's lives. Many appear to solve this tension by retreating to the idea of a “wall of separation” between church and state. Despite this seeming automatic aversion to public money for education going anywhere near faith-based groups, as the next section demonstrates, faith-affiliated groups are in fact already taking advantage of the charter school system.

Faith-Based Schools that Become Charter Schools by Shedding Their Religious Nature

By 2007, investigators had already identified a number of charter schools that had strong religious roots, and a more recent article by Janet Decker and Kari Carr claims that there are now hundreds of charter schools that have replaced closed Catholic schools and that in a large share of those schools, the new charter school is in many respects a continuation of the old school, in terms of staff, students, and values, except that religion has been formally removed from the school. ⁸⁷

As described in a book by Lawrence Weinberg devoted to the topic, to survive in the face of existing state law, “parents can create charter schools that accommodate their religious belief, but not such schools that endorse their religion.” ⁸⁸ As Weinberg sees it, charter schools can adopt a cultural mission and curriculum that nicely fits with the school sponsors’ religious values. The school can accommodate the faith(s) of its students, including scheduling school holidays to fit its students’ religious holidays in the same way that public schools do; having released-time programs for children to leave the grounds during the day to attend worship service, as some public schools do around the nation; and perhaps making time during the school day for students voluntarily to pray and carry out other religious acts on their own. Such a school may also teach languages tied to the interest of religious faiths (such as Hebrew for Jews and Arabic for Muslims). But it may not be a faith-based school. Moreover, such schools may not select students on the basis of their family's religious faith, although, of course, there is likely to be substantial faith-based self-selection into such schools.

Not being a religious school means there may be no required or school-led prayer, no teaching of religion as a creed for students to follow, no religious symbols all around the school, and, more generally, religion may not permeate the curriculum in ways that private religious schools often argue is a central feature of what they offer.

According to Weinberg, these faith-inspired schools should probably be (and generally are) formed and managed by independent nonprofit organizations and not by religious organizations. But their initial sponsors and the board members who run the nonprofit organization can be a preexisting group of religious people of one faith, even religious pastors. ⁸⁹ If they are scrupulous about their independence, these charter schools could probably rent space from churches as the venue for their school. ⁹⁰

For a number of families and faith leaders, this development of the religiously affiliated charter school is good enough. They are very happy to have this sort of charter school that is not formally a religious school but which, as a practical matter, largely brings together children from the same faith and structures itself so that the teaching and practice of the family's faith is readily accommodated.

Nonetheless, this is by no means ideal for families who want to choose truly religious schools. They can, of course, do that now via the fully private sector. But such families are unhappy to have to pay local property taxes and relevant state taxes to support a public service they would not use, especially if they have to pay again for that service privately. Moreover, a large number of families simply cannot afford the private option. They would probably most prefer a reform that provides them with school vouchers they can use to pay for all (or most) of the cost of their children's education in the conventional private sector. But the political prospects for the rapid expansion of school voucher programs currently look slim.

Another possible strategy for those favoring government-funded faith-based schools would be for them to create (or become) charter schools assuming that the regulatory burden was acceptable. But, of course, that is currently legally off the table in all charter school states.

This then brings us to the legal question to be explored here: is the current discrimination against would-be religious charter schools and their religious family users unconstitutional?

Religious Schools as Charter Schools?

Until now, most people have assumed that, as just discussed, the most that faith-based groups could hope for is to obtain charters for wholly secular schools that are in various ways affiliated with religious groups and that are accommodating of the religious practices of the families whose children are in the school. ⁹¹

At least a few legal scholars have examined aspects of the legal question addressed here although only Professor Aaron Saiger has directly considered the precise charter schools question. ⁹² From these writings emerges a distinction that will be elaborated below: while states might be able constitutionally to elect to allow religious schools to obtain charters, they might also be able to constitutionally exclude those schools from the charter school system (as, of course, all states so far have done). I try to cast doubt here on that latter conclusion.

Facial Discrimination on the Basis of Religion

Every state charter schools program discriminates on its face against religious schools. People who want to start every imaginable sort of school are entitled to ask for a charter except those proposing religious charter schools.

Does this discrimination against religion presumptively violate the equal protection clause of the Fourteenth Amendment or the free exercise clause of the First Amendment?

There is no obviously close analogy in the case law here. But consider these hypothetical examples. Suppose a state government decided that it will provide financial assistance to poor single mothers but only those who are atheists (arguing that those who are religious should be taken care of by others of their faith). Perhaps such a policy might be implemented by denying aid to otherwise eligible single mothers who attend church. Would not that be unconstitutional?

Or suppose a state decided that its Medicaid benefits (payment for health care) are available at any hospital except for medical care provided at religiously affiliated hospitals (for example, Saint Francis or Mount Sinai). Perhaps such a policy might be implemented by refusing to pay for care at hospitals where clergy of one faith are on the payroll and available to patients in the hospital, one faith's religious services are held at a chapel in the hospital, the hospital board is dominated by clergy, or the hospital has adopted policies concerning what services it will provide that are faith-based (for example, no abortions). Of course, some hospitals with faith-based histories have today become wholly secular institutions, apart from their names, and a policy of excluding religious hospitals from participation in Medicaid might be designed to prompt the remaining religious hospitals to become secular. But would not this exclusion from the program of religious hospitals and Medicaid users who seek care in religious hospitals be unconstitutional?

Or suppose a state government announced that any bakery can compete for a contract to provide bread for public school lunchrooms except bakeries run by religious orders such as the Sisters of the Poor or the Christian Brothers. While bread is bread, some might find it symbolically tainted if supplied by religious groups. Yet would not this open discrimination against religiously based bakers be unconstitutional?

Or suppose the federal government decided that food stamps it provides to low income people are not valid for kosher or halal food (arguing, say, that people who buy those typically more expensive products should not be depending on food stamps). Could the government constitutionally exclude from the program the sort of food whose consumption is required by the faith of poor people receiving food stamps? I doubt it.

While these hypothetical examples all suggest nearly unimaginable scenarios today in the United States, they also seem, at least on their face, to be highly suspect as a matter of constitutional law. In each instance, the individual or organizational participant involved is being singled out for worse treatment on the basis of religion. The hospital, grocer, and baker are offering secular products (health care, food, and bread) but may not participate in the program because of a religious attachment to what they provide. So, too, the individual participants are unable to satisfy a secular need they have (to cash because they are poor, to medical care, to the food they want to eat) because either they or those with whom they wish to deal are engaged in a faith-based activity.

To me, these hypothetical scenarios would appear to violate both the Free Exercise clause of the First Amendment and the Equal Protection clause of the Fourteenth Amendment. In Fourteenth Amendment lingo, they involve a “suspect classification” that would have to be justified on the basis of a “compelling state interest.” ⁹³ And as the First Amendment makes clear, it is very difficult to see how the state can have a compelling state interest in penalizing the free exercise of religion.

If this way of looking at the issue is correct, then it would provisionally seem to follow as well that it is unconstitutional to single out religious schools and deny them the right to be charter schools. After all, they are offering to provide the secular service of educating children, but they are precluded from participating in the government's charter schools program because they are simultaneously engaging in religious activities. So, too, the families who wish to have their children educated in a school that reinforces the parents’ faith are denied that choice. Can this open discrimination against families wanting to make a faith-based choice and providers wanting to offer a faith-based public service be justified?

If the state were only to operate and fund traditional public schools, the same exclusion of funding for those seeking faith-based schooling for their children would follow. But in that scenario, one might argue, the state is speaking through its regular public schools, seeking to educate children in a publicly determined way, promoting publicly reached values, and so on. This is how public education has been traditionally understood. And when the government uses the purse to fund the way it speaks, it may well have no obligation to fund other speakers. ⁹⁴ In this respect I agree with the conclusion of Laycock and others that just because we have a compulsory education system that offers families government-funded public schools, this does not constitutionally require government also to provide scholarships or vouchers for families that prefer religious schools (although not everyone agrees). ⁹⁵

But once the institution of charter schools has been established, things look and feel very different. Now, any private party who can make a showing that it will pursue the basic secular goals of education may seek a charter (and with that public funding) regardless of the rest of the trappings of the school. To be sure, the school must teach reading, writing, and arithmetic to elementary school age children, and provide college prep education and/or vocational training for secondary school children. But the school can be funded and willing parents may choose it for their children regardless of other values the school seeks to impart to its pupils, the teaching style it adopts, the rest of the curriculum it offers, the nature of its teaching force, and so on. Except the trappings of the school may not be such that it is a faith-based education that is being offered along with the secular education of those enrolled.

Of course, as discussed above, a charter school is not completely free from public control and must in some sense have its overall program blessed by the public body that gives it the charter. Nevertheless, at the level of principle this sort of oversight does not seem different from the regulation of clearly private schools that we see in many states and whose regulation is clearly constitutionally valid even in a legal regime such as ours that guarantees parents the right to choose private schools in lieu of public schools for their children. ⁹⁶ After all, aspects of the curriculum, the school year length, aspects of teacher qualifications, standardized test-taking requirements, and the like are matters that states in some cases do, and certainly could, validly require of the private school sector.

To be sure, not everything that calls itself a school necessarily is the sort of institution that should count as a school for purposes of the charter school law. So, for example, if twelve-year-olds at a “charter school” are going to do nothing but read and recite liturgy all day, the “school” probably could not get away with claiming that through this “reading and speaking” training it would be providing minimally adequate secular education, to say nothing of its failure to offer the educational curricular variety generally appropriate for children of that age, like math and science. Such a pretend “school” should not be granted a charter. But parents sending their children to a conventionally private “school” (or “home school”) of this very same sort could (and should) also be prosecuted for violating their obligations under the compulsory attendance laws. ⁹⁷

In sum, if the exclusion of faith-based schools from the charter school system seems presumptively impermissible under the principle that government may not discriminate on the basis of religion, are there other principles that can trump that presumption?

Do States Have a Compelling Interest in Not Awarding a Charter to a Religious School Because to Do So Would Violate the Establishment Clause?

Suppose a state voluntarily agreed that charters could be granted to religious schools that meet all of the normal requirements for charter schools. Would the funding of such schools violate the Establishment Clause of the First Amendment? If so, then it would seemingly follow that the Free Exercise or Equal Protection claim raised in the prior section would fail. That is, under this line of analysis, states that refused to fund faith-based charter schools could successfully assert a compelling state interest in support of their decision: not to violate the First Amendment.

But I do not believe that the U.S. Supreme Court should or would reach such a result. On this question, I agree with Saiger. ⁹⁸

The Pierce case makes clear that in the U.S. government may not require all parents to send their children to conventional public schools. ⁹⁹ Attendance at private schools (including religious private schools) must also be recognized. It is also legally permissible for state government, if it chooses to do so, to provide financial support to these conventional private schools and the users of these schools at least in certain ways. ¹⁰⁰ But government may not provide financial aid in ways that violate the Establishment Clause of the First Amendment.

Would funding religious schools through the state charter school program violate the Establishment Clause? Many would say that this clearly is the case and for that reason it cannot be unconstitutional to exclude faith-based schools from the charter school program. Indeed, they would say this is why states wisely chose from the outset to exclude religious schools from the program. But is this correct?

For some the answer is simple, and the legal analysis would go like this: Charter schools are public schools. Public schools may not be religious schools. Therefore, funding public religious schools is forbidden. But this is too simple because it simply assumes that charter schools are public schools for Establishment Clause purposes. And I do not believe that is correct.

In the 1982 case Rendell-Baker v. Kohn, ¹⁰¹ the U.S. Supreme Court made clear that even though a school is 100 percent funded by the government, it is not necessarily a public actor for constitutional law purposes. The school in that case was a private school that had a contract with the Boston public schools to provide education to certain difficult students. The school district paid the private school for this service. Some teachers at the school brought an action against the school alleging constitutional violations of their rights. The U.S. Supreme Court held that this school did not engage in “state action” and hence the claims were dismissed (as the constitutional rights alleged did not apply against a private actor.) Although the contract with the school in Rendell-Baker looked very much like the charter that charter schools receive these days, this decision should not be taken to fully resolve the issue before us, since that case did not involve the Establishment Clause.

Professor Robert O'Neil, a leading First Amendment scholar, indirectly explored this question in 1999. In a chapter centrally about whether acts by voucher-funded schools constitute “state action” and thereby trigger various constitutional rights, he notes in passing that charter schools (in contrast, at the other extreme, to home-schooling parents and in contrast with Rendell-Baker) do engage in state action. ¹⁰² But for purposes of his analysis O'Neill made certain explicit assumptions about public control over charter schools, and he acknowledged that a charter school system of a more decentralized and autonomous sort might be different. In any event, he did not address the religious discrimination issue before us here.

At about that same time, a student note that focused on Texas’ charter school system largely came to the opposite conclusion, finding that courts would probably not consider charter school operators as engaging in “state action” especially when the issues before the court concerned matters such as teacher and student rights. ¹⁰³ Like O'Neil, the student note did not examine the issue of the nature of charter schools in the context of the Establishment Clause.

In May 2014 the Office of Civil Rights of the U.S. Department of Education distributed a “Dear Colleague” letter that simply announced (without any legal analysis) that charter schools are public schools and therefore subject to federal civil rights statutes just as are regular public schools. ¹⁰⁴ The letter says nothing about whether charter schools engage in state action so that teachers and students have constitutional rights against charter schools, although that would seemingly follow from the letter's assumption. In any event, the letter understandably pays no attention to the Establishment Clause.

I agree that under the U.S. Constitution a school district could not choose to create and operate a Catholic school (even if it were willing also to create and operate schools teaching or following other faiths if there were family demand for them).

But that is not necessarily the right way to look at charter schools. States and school districts do not advertise, “We are looking to award a charter to a school that will emphasize training in science, or a school that will be bilingual, or a school that will have strict discipline, or a school that will evaluate its pupils on the basis of a portfolio assessment rather than standardized tests.” That may be how districts go about creating alternative or magnet schools. Rather, the chartering bodies wait for those seeking charters to come forward and propose a school. And these proposing institutions are private organizations, not public institutions. Viewed in this way, the funding of charter schools looks and feels very much like the funding of voucher schools in the sense that in both instances the government is putting up money so as to facilitate the choice by families of privately run schools they prefer for their children.

Put differently, although awarding a school a “charter” and calling it a “public charter school” may suffice to make it a public school for state law purposes where state constitutions restrict financial support to “public” schools and prohibit aid to non-public schools, this does not resolve the question of how to treat the school for Establishment Clause purposes. For that, the fact that the initiation of the school comes from private parties and the fact that families are never assigned to the school but only attend by their own private choice arguably makes a great deal of difference. After all, the core point of the Establishment Clause is to prevent government from “establishing” a religious institution.

We know from Zelman, the Cleveland school-voucher case, that merely providing funding that benefits a religious institution cannot by itself be enough to violate that principle. There is every reason to believe that religious organizations benefit from a voucher program that provides publicly funded scholarships that families sign over to religious schools to pay for their children's tuition. The same would be true if the Sisters of the Poor Bakery were allowed to bid for and win a government contract to provide bread to the schools, or when Catholic hospitals are allowed to participate in Medicaid. It is also clear that the Cleveland voucher plan enabled families to pursue their religious faith just as current rules allow food stamp recipients to spend them on kosher or halal food. These benefits to the exercise of religion are collateral side effects to the use of public funding to satisfy other secular objectives such as to keep poor people healthy and fed, to provide nourishing food in public cafeterias at low cost, and, of course, to help children become educated.

Still, this does not necessarily mean that the formal arrangements by which the money goes to the religious organization are necessarily irrelevant for constitutional law purposes. Justice O'Connor's concurring view in Zelman, ¹⁰⁵ which provided the crucial fifth vote to uphold the plan, was that it made a great deal of difference that the voucher went to the parents. For her, the payment of a lump sum directly to a religious charter school based on the number of students it enrolls would probably be unconstitutional on the ground that the connection between government and religion would be thought too tight by the average informed citizen a matter that she felt was central to deciding the First Amendment question.

But, of course, Justice O'Connor is no longer on the Court. And, in my view, the present Court membership is likely to see things differently (O'Connor gave too much weight to form over substance). Two justices from the five-member majority in Zelman, Justices Thomas and Kennedy, are still on the Court. I believe that at least Justice Alito, Justice Gorsuch, and Chief Justice Roberts would join with them in looking at the First Amendment differently from the way Justice O'Connor did. This group of five Justices is likely to conclude that there is no economic difference between the direct funding of a charter school, which only gets the money if parents decide to send their children there, and the indirect funding of a voucher school, which only gets the family's voucher signed over to it if parents decide to send their children there. Hence, I believe these justices would conclude that the “primary effect” of both the programs is the same, that is, to help educate the nation's young. Moreover, the “primary purpose” of both plans is also the same, that is, to expand family choice in education by allowing for the government funding of privately created and managed schools that parents select for their children (“primary effects” and “primary purpose” being two tests advanced in the Lemon case that the Court has deployed on many occasions including Zelman in its application of the Establishment Clause). ¹⁰⁶

Or put differently, I believe this group of five would adopt the broad approach of Chief Justice Rehnquist speaking for the majority in Zelman: (1) would parents be making a “genuine and independent private choice” by selecting religious charter schools for their children; and (2) would a charter school program that includes religious charter schools be “neutral”? ¹⁰⁷ Surely they would answer “yes” to both of those queries.

Furthermore, in contrast to the school voucher plans that have been created to date, there is every reason to believe that even if faith-based schools could become charter schools, most of the schools participating in the plan will be nonreligious schools. Certainly if one faith-based school were to sue to become a charter school, then at that point it would be seeking to become a single parochial school in a sea of secular schools. Hence, the symbolic connection of government with what were overwhelmingly religious private schools in the Cleveland voucher plan would be absent. Moreover, even over the longer run, one would expect that nonreligious schools would remain not only a substantial core but probably the majority of the participants in the charter school scheme. Hence a feature of both the Cleveland plan and all of the prior aid-to-private school programs that came before it that bothered at least some of the justices would be absent. ¹⁰⁸

There might arguably be more “entanglement” between the public and charter schools than between the public and voucher schools (“entanglement” being another test put forward in Lemon that the Court has frequently pointed to in the past in Establishment Clause cases, although it appears to have put that aside more recently). ¹⁰⁹ Entanglement has always been a complicated and slippery concept. The symbolic entanglement between the state and religious charter schools would primarily be a matter of how one “sees” the two systems. Voucher plans clearly enable families to choose private schools. I have been arguing that, on close examination, charter school plans do the same thing.

To argue that there is a difference because charter schools are “labeled” for some purposes as public schools seems the wrong way to look at things. For just as charter schools are seen today by some observers as obviously public schools, if you listen to core supporters of conventional public schools you will hear the opposite. They regularly loudly complain that charter schools drain funds and desirable families from the public schools and that charter schools are run by private entrepreneurs and hence do not have the central characteristics that they ascribe to being a truly public school.

In Widmar (1981), ¹¹⁰ Lamb's Chapel (1993), ¹¹¹ and Good News Club (2001) ¹¹² public authorities defended their decision to exclude religious groups from using public facilities on equal terms with other groups on the ground that to do so would violate the Establishment Clause, in part because this would entangle the state and religion. The Supreme Court rejected this argument in all three cases, finding that including such groups among those who are allowed to use the facilities would not be unconstitutional.

In the Rosenberger case (1995), ¹¹³ the University of Virginia had denied student religious groups funding which was otherwise available to other student groups. The university again argued that to fund religious groups would violate the Establishment Clause. The Supreme Court rejected this argument once more. Viewed this way, Rosenberger is a clear precursor to Zelman. The Court's view was that the relevant student activities fund was used to pay for communications initiated by various student groups. This was not government speaking, but rather the government promoting speech within its student body. In such a setting, it would not amount to an establishment of religion to include student religious groups in the program. If the charter school program is seen in the same way, then Rosenberger provides a strong analogy for the issue under discussion here.

To distinguish Rosenberger it would seem that the charter school system would have to be viewed as a mechanism by which the state seeks to offer exclusively nonsectarian schooling in innovative settings. But, Establishment Clause concerns (or hostility to religion) aside, it seems inconsistent with the state's central objective of promoting family choice in education to then automatically reject educational programs that some parents want for their children because the program is proposed to be offered through a religious school. Moreover, to the extent that charter schools are viewed as vehicles for improving the educational attainments of America's children, it would seem bizarre to exclude religious schools when the educational success of faith-based schools has been quite strong as compared with public schools. ¹¹⁴

To be sure, the on-the-ground politics behind the charter school movement may have required that charter schools be nonreligious schools for advocates of the plan initially to achieve the needed majority vote. But, it seems to me that if a similar argument had been made in Rosenberger, that the student activities fee could only have been adopted if student religious groups could not receive any of the proceeds, that argument would surely not have saved the restrictive nature of the University of Virginia program.

In Maine, Vermont, and New Hampshire some small school districts traditionally have not operated their own schools. ¹¹⁵ Instead their practice has been to pay for their students to attend elsewhere, sometimes at private schools. This program, which goes back a very long time, is functionally equivalent to a voucher plan in those communities where families are given a number of private choices as to where their children may attend at public expense.

In 1981 the Maine legislature voted to exclude private religious schools from the program on the ground that including them would violate the Establishment Clause. Some years later the town of Raymond, which did not operate a high school and paid for local students’ education at a variety of private schools but not faith-based schools, was sued by a group of families seeking to have their children's education at a Catholic high school paid for by the town. The attack on the exclusion of religious schools from the family options was based (for purposes relevant here) on the claim that this violated the Equal Protection Clause. This exclusion was defended on the ground that to include those schools would violate the Establishment Clause. And in a 1999 decision the Maine Supreme Court agreed with the latter claim. ¹¹⁶ This decision came before Zelman, however, which I believe Zelman overrides. ¹¹⁷

Justices Breyer and Ginsburg dissented in the Cleveland voucher case, and so there is reason to believe that they would find it unconstitutional to allow religious schools to be charter schools. ¹¹⁸ Yet, even as to them, recall that whereas the vouchers in Cleveland were overwhelmingly used in religious schools, the issue addressed here is whether religious schools could seek to become charter schools and thereby obtain a foothold in a program dominated by nonreligious schools.

As for Justices Kagan and Sotomayor, although they are typically lumped in as part of the “liberal” wing of the Court, we do not have a definitive view of how they would view the matter. Justice Kagan dissented in the Arizona tuition tax credit case in an opinion that technically only argued that the plaintiffs there should have standing to challenge the program; but the text suggested that she might well have concluded on the merits that the program was unconstitutional. ¹¹⁹

While Justice Sotomayor joined the Kagan dissent in that case, she did not write on the matter. Her decisions on religious cases as a lower court judge are not terribly revealing. ¹²⁰ She has not issued an opinion on a school funding case as a Supreme Court justice although she has been accused by some for decisions in other areas, as inappropriately pro-Catholic. ¹²¹ I think little can be made of this record, although it is plausibly relevant that she was raised a Catholic and attended Catholic schools and at some level must feel that this experience helped shaped who she is and her subsequent career. ¹²²

The upshot is that it could well come down to how our newest Supreme Court justice would vote on the matter. There is not a lot of strong evidence in Justice Gorsuch's record as a lower court judge on which to base a confident prediction. But given his generally conservative background and his own Catholic education, I have grouped him with Justices Alito and Roberts, and so I feel reasonably comfortable in concluding that the U.S. Supreme Court would probably not find it to be a violation of the Establishment Clause for a state to choose to award a charter to a religious school. But, of course, none have done that so far.

Even If States May Award Charters to Religious Schools, Must They?

However, on the question of whether, even if states may award charters to religious schools, they must do so, I offer a position different from that of Saiger and Laycock. In the Witters case, ¹²³ decided by the U.S. Supreme Court in 1986, a blind college student in the state of Washington applied to the Washington Commission for the Blind for financial aid. He was attending a Christian college, studying to become a pastor, missionary, or youth director (of a religious organization). His application was denied and the Washington Supreme Court upheld the decision on the ground that to award him funding for this sort of education would violate the Establishment Clause. The U.S. Supreme Court unanimously reversed. Although the facts there involved higher education, this decision clearly helped pave the way for the Zelman decision. Based on Witters, so far as the federal constitution is concerned, Washington could have granted financial aid to that student.

But just because Washington could elect to support such a student without violating the federal constitution, must it? This issue returned to the U.S. Supreme Court in Locke v. Davey. ¹²⁴ Like many other states (the count is usually put at thirty-seven), ¹²⁵ Washington has a provision in its constitution that bars the funding of religious schools. Many of these provisions were adopted in the later part of the nineteenth century as part of an anti-Catholic school movement that tried (but failed) to achieve a similar amendment to the federal constitution. ¹²⁶ These so-called Blaine Amendments (after the Congressman who headed the national effort) are phrased differently from state to state and their words have been interpreted differently by state supreme courts. ¹²⁷ Sometimes the provision prohibiting aid to religious schools is joined with a provision prohibiting aid to any school not under the control of the state. ¹²⁸ As mentioned earlier in this article, the existence of Blaine Amendments is one reason why charter school advocates initially were keen to term charter schools public schools (and why they argued that the “charter” feature satisfied the “control” requirement of some state constitutions).

Given its interpretation of the Washington constitution, the Washington Supreme Court determined ¹²⁹ that it would be unconstitutional under its law for the state's generally available Promise Scholarship program to provide financial aid to support Joshua Davey because he was studying devotional theology at a religious college, and so the administrators of the program denied him the financial aid he sought. Since the U.S. Supreme Court in the Witters case had made clear that Washington could aid Davey without violating the Establishment Clause, Davey believed he had a strong case that the state was violating his Free Exercise rights by excluding him from the program (that applying the Washington Blaine Amendment in this way was unconstitutional under the federal constitution). But he lost.

Writing for the U.S. Supreme Court, Chief Justice Rehnquist concluded that there is some “play in the joints” ¹³⁰ between the Establishment Clause and the Free Exercise Clause. Rehnquist cited the Walz case ¹³¹ which long earlier upheld as not violating the Establishment Clause a “neutral” law that allowed churches along with other nonprofit organizations to be exempt from property taxes. The idea the Chief Justice sought to convey in his opinion was that, while the opponents of the churches’ tax exemption did not have a valid Establishment Clause claim against the exemption, the churches in Walz also had no Free Exercise claim that constitutionally entitled them to a property tax exemption if it had been extended only to other nonprofit organizations. Put differently, according to the Chief Justice, not every matter involving religion in some way has to be resolved by the Court by applying one or another of the clauses of the First Amendment.

But notice that the Walz-based argument advanced by the Chief Justice in Locke was not based on the holding in Walz. There, the Court was not actually called upon to decide whether excluding religious groups from among all other nonprofit groups from the tax exemption would or would not violate the Free Exercise Clause. ¹³²

Putting aside this hypothetical analogy, what Rehnquist specifically concluded was that while Washington could have awarded aid to Davey if it wished, it was not constitutionally obligated to do so. In its details, Locke may look to some like a very narrow and special case. The applicant wanted to obtain a “devotional theology degree” so as to be trained to be a minister. So, it is perhaps understandable that a program of purely religious instruction for a religious career is something that the justices felt states should not be required to support even if they could. But is Locke really only narrowly restricted to this sort of case, or does Locke suggest that the Court believes there should be this “play in the joints” as a general matter, especially when it comes to state funding decisions?

In his very thoughtful 2004 article Laycock argues that the Court is likely to apply Locke broadly in cases involving public funding, ¹³³ and Saiger agrees with this. ¹³⁴ I am not so sure, however.

Some might initially think that Locke is simply incompatible with Rosenberger. But Rosenberger was decided not as a violation of Equal Protection or the Free Exercise clause but of the Free Speech Clause on the grounds that refusing funding for the promotion and publication of papers by a religious student group amounted to “viewpoint discrimination” in violation of this part of the First Amendment. And while those who seek to gain approval for a religious charter school might argue that they too are attempting to exercise their free speech rights, it is not at all clear that they can make out a “viewpoint discrimination” claim. Funding schools and funding pamphlets may not be understood to be the same thing, a point that seems to underlie Locke, although Laycock is not convinced of the distinction. ¹³⁵

In any event, by now there are several new members of the Court. The vote in Locke was 7–2 with Justices Scalia and Thomas dissenting. Even if Chief Justice Roberts, Justice Alito, and Justice Gorsuch were to join with Justice Thomas, Locke might well remain good law because Justice Kennedy voted with the majority in that case. Yet, the question is how broadly or narrowly Kennedy and others would interpret Locke. That is not self-evident.

Lower courts have grappled with Locke. Then judge Michael McConnell, now a Stanford Law professor, writing in 2008 for the Tenth Circuit in the Colorado Christian University case ¹³⁶ distinguished (or avoided) Locke by finding that Colorado's college scholarship program (not initially self-evidently different from Washington's) actually discriminated among religious groups and therefore was unconstitutional regardless of what Locke might otherwise imply. ¹³⁷

But in 2004 in Eulitt v. Maine Department of Education ¹³⁸ fast on the heels of Locke, the First Circuit read Locke far more sweepingly and upheld the Maine law, noted earlier, that allows nonoperating school districts to pay for the education of their students in private schools but not in religious private schools. Regardless of whether it might be permissible to make such payments, the panel concluded that Locke clearly gave Maine the right to decide not to fund students attending religious schools.

Following language in Locke, the opinion says that there is no religious animus behind the Maine program. Rather, the reasons for limiting the schools where students may attend with public support “include Maine's interests in concentrating limited state funds on its goal of providing secular education, avoiding entanglement, and allaying concerns about accountability that undoubtedly would accompany state oversight of parochial schools’ curricula and policies “(especially those pertaining to admission, religious tolerance, and participation in religious activities).” ¹³⁹ There is a certain irony here because, at least as a nationwide matter, Blaine amendment limits on funding children attending private schools are undoubtedly rooted in animus towards Catholics and Catholic schools. But, as Laycock points out, the way the U.S. Supreme Court dealt with the Blaine amendment history in Locke makes it highly doubtful that actions taken in the nineteenth century will come back to strike down state political choices made in the late twentieth or early twenty-first centuries. ¹⁴⁰

Moreover, the First Circuit panel correctly pointed out that when the U.S. Supreme Court in Locke talked about animus it asked whether “the state action in question imposes any civil or criminal sanction on religious practice, denies participation in the political affairs of the community, or requires individuals to choose between religious beliefs and government benefits.” ¹⁴¹ Of course, neither the Maine plan nor the Washington scholarship plan in Locke criminalized attending private religious schools. This serves to distinguish the Babalu case where a specific religious rite was criminally outlawed. ¹⁴² Nor, of course, do the Washington or Maine regimes prevent those who attend religious schools from, say, voting.

Whether these plans require individuals to choose between religious beliefs and receiving benefits is a harder question, however. To be sure, the students are free to accept the government benefit by attending nonreligious schools, and that seemed enough for the First Circuit, which saw itself as just following the Locke opinion in this respect. Is this convincing in the setting before us involving religious charter schools?

In some faiths there may well be a duty to send one's children to religious schools, and under current charter school rules those families surely are directly forced to choose between violating their faith and giving up the benefit of subsidized education for their children. Even those who are not compelled by their faith but wish, in pursuance of their faith, to give their children a religious education are in a genuine sense financially penalized. These parents seem to me more directly burdened than those in the college scholarship plans who are seeking training in religion. Being trained so that you can later be hired in a religious role may be the educational goal of those college students, but it is not quite the same as the practice of their faith. By contrast, providing one's children with religious education is. Nonetheless, the Eulitt panel clearly read Locke to permit states to decide not to expand their programs to include religious schools. And if this is correct, then it would also seem to follow that states could as well choose to exclude religious schools from their charter school plan, even if it would be constitutional to include them. And this is exactly how Laycock reads Locke: when it comes to state funding, very little is required by the Free Exercise Clause (or the Equal Protection Clause) even though he agrees that “[r]efusing state funding for math and reading, because the school also teaches religion, is clearly a penalty on teaching religion and on attending a school that does so.” ¹⁴³

So, how should we think about Locke today? Let's turn back to the hypothetical examples of religious-based restrictions I gave at the beginning of this article. It still seems to me that in the great unlikelihood they would be enacted, at least some of them (and probably all of them) would be stuck down by the current U.S. Supreme Court, notwithstanding Locke. Giving welfare only to atheists? Surely that makes poor single mothers choose between money and going to church and financially penalizes those who make the latter choice. That this is a state spending choice designed only to help certain mothers seems to me unlikely to save the provision. As I suggested earlier it is possible that legislators might honestly think that religious groups will take care of their own faith members, which means they are perhaps not acting with the sort of animus towards religion that the Court had in mind in Locke. But the plaintiff will be someone who is poor despite her faith, and I cannot see such a blatant discrimination being upheld.

What about preventing food stamp recipients from using those food stamps to buy the food their religious beliefs require them to eat? I cannot see how or why there would be “play in the joints” for this restriction, either. Congress can decide that food stamps may not be used to buy sugar-sweetened beverages. But to attach a condition (on improvidence grounds) that prevents poor Muslims and Orthodox Jews whose faiths require them to eat only halal and kosher food from buying that food seems indefensible even under Locke (especially because food stamp users are otherwise now entitled to use their benefits for other high-cost foods if they wish).

Both the welfare and food stamps examples seem to me to involve explicit penalties arising from the exercise of religious beliefs. And it seems to me that, notwithstanding either Locke or the First Circuit in Eulitt, this is true as well if a family signed its child up for a religious school (in furtherance of or perhaps even as compelled by its faith) but the school then was denied a charter for the sole reason that it was a religious school. Now the family would have to pay tuition for its child.

It would seem therefore to come down to whether, in retrospect, Locke will be understood as being about a historic avoidance of public funding of ministry training. I think it should be.

Notice again that the argument for allowing religious schools to become charter schools is not the same as arguing that private school users have a constitutional right to vouchers. As I discussed earlier, that would be a much larger reach. Remember that charter schools are subject to restrictions, and religious schools would also have to comply with these were they to become charter schools. Indeed, some religious schools, just like some private nonreligious schools, would not want to become charter schools even if they could. Moreover, these private religious schools might oppose allowing any religious schools to become charter schools because that would create competition that they might well not appreciate.

But for those religious schools willing to accept the charter school program on the same terms as other charter schools with their own distinctive curricula, it is difficult to come up with a justification as to why they should be automatically excluded. Saiger observes that states have political and financial reasons not to include religious schools as charter schools. ¹⁴⁴ This perhaps helps us understand why the current laws are drawn to exclude faith-based schools. ¹⁴⁵

But, and Saiger agrees, political and economic reasons cannot automatically constitutionally justify the discrimination. After all, political or financial reasons for, say, keeping churchgoers off welfare, or religious hospitals out of Medicaid, or religious bakeries out of the government contracting business would, to me, seem insufficient arguments to justify such exclusions. If charter schools are really about giving families the ability to choose among privately created schools that are willing to meet certain basic educational criteria, then on what constitutionally acceptable basis are schools that educate children in a context permeated by religion to be excluded?

Many ordinary citizens will respond by saying that they do not want their tax dollars going to schools that conduct prayers, teach children that sex outside of a heterosexual marriage is a sin, teach intelligent design as an alternative to evolution, and have children learn that the Bible the school uses is the word of God. Many of these people who point to an objectionable use of their tax dollars do not really like wholly private schools teaching these things either, but they have come to accept Pierce. To have public money and public sponsorship associated with such schools, however, would be very distasteful to them even if the children attending these schools learn the basic secular education skills that all charter schools undertake to teach. Ultimately, these people are troubled by the whole notion of “family choice,” fearing that this could turn America into a religious battleground of the sort they have seen in Northern Ireland, the former Yugoslavia, and Lebanon and elsewhere in the Middle East. ¹⁴⁶

Others, of course, see family choice in education quite differently and would point to countries throughout Europe and to Canada, Australia, and the like, where public funding of religious schools has long been in place, or to the former Soviet-dominated nations that, once having obtained their freedom, promptly put in place schemes to fund family choice of religious schools. ¹⁴⁷ These supporters of family choice argue that empowering parents to choose what they think is best for their children is not only good for families but also a way of making more families appreciate the tolerance America shows for differing points of view. For many parents, teaching values to their children is their very most important exercise of their Free Speech rights. Family choice proponents believe that when people realize that they are supported in expressing their rights, they will support that in others as well.

Another point made by family choice advocates is that if faith-based schools could become charter schools, some of the most contentious religion-based battles in our traditional public schools might fade away. Those who are so keen on prayer, bible study and the like and so opposed to certain matters they now find present in our public schools (like sex-education classes) could move their children to charter schools that are more in line with their values.

This brings us to a final distinction to be made between the problem before us and the one presented in Locke. In that case, the plaintiff chose a job-related course of higher education to pursue. When it comes to elementary and secondary education, however, parents are compelled to force their children to participate (on pain of criminal prosecution and/or child neglect proceedings). For many low income parents who cannot afford private religious schools this means they must forego the exercise of their faith and send their children to a traditional public school whose value teachings the parents may well find in conflict with their faith.

So it might come down to this: If charter schools are really about empowering families to choose from a wide range of privately created options as a way of finding the sort of school that parents believe is best for their child, then to allow states to exclude the preferences of a significant number of families because of what appears to many to be a hostility towards religion seems unacceptable. But if instead charter schools are seen as a way of delivering variations on what conventional public schools try to do and all too often fail to do, then in light of many people's acceptance of the idea of a high wall of separation between church and state, perhaps a choice to exclude religious schools from the game might be thought acceptable. Put differently, are charter schools about liberty or about bureaucracy?

For Justice Kennedy, who could well be the key vote on this issue, the “freedom” or “liberty” to direct your children's education in ways that are in harmony with your family's religious faith is likely to be seen as part of the individual freedom in intimate matters that Justice Kennedy has been supporting in a range of other important decisions. ¹⁴⁸

Conclusion

Under the U.S. Internal Revenue Code, taxpayers are allowed to take deductions for contributions made to qualifying charitable organizations. ¹⁴⁹ At present these include religious institutions. Could Congress constitutionally change the law and exclude religious institutions, providing tax deductions only for donations to other nonprofit groups that provide charitable or educational services? The dictum in Chief Justice Rehnquist's opinion in Locke asserts that Congress could do this. Although I am not convinced this is correct, let us assume for these purposes that Congress could deny tax deductions for contributions made directly to churches to support the salaries of their pastors and/or the upkeep of the buildings used for prayer and other religious worship.

But what if Congress also denied tax deduction status to donations made to religiously affiliated organizations which provided social services, medical services, or educational services as part of their religious mission? For example, could Congress legally distinguish in this way between donations made to the Red Cross and the March of Dimes on the one hand and Catholic Charities and the YMCA on the other? And closer to the problem examined here, could Congress legally distinguish between contributions to Stanford University and to Santa Clara University? I do not think so.

If I am right, this means that the “play in the joints” emphasized in Locke ought to be restricted to legislative decisions regarding public financial support of wholly or centrally religious activities. On this analysis, I believe that religious charter schools should be outside that category because, in addition to their religious mission, they are also squarely about educating children in core skills that schooling is fundamentally about. Put differently, a church and a church school are not the same thing if the school is really a school.

How the currently sitting Supreme Court justices think about Locke was revealed on June 26, 2017, in their decision in Trinity Lutheran Church v. Pauley. ¹⁵⁰ In that case, the U.S. Eighth Circuit Court of Appeals, in a 2–1 decision, rejected arguments that Missouri's Blaine Amendments violate the U.S. Constitution's First and Fourteenth Amendments. ¹⁵¹ At issue was the denial by Missouri's Department of Natural Resources on Blaine Amendment grounds of a grant application by Trinity Church for a Playground Scrap Tire Surface Material Grant that would have allowed it to resurface a playground at its day care and preschool facility on church premises. The petition for certiorari framed the Question Presented as such: “Whether the exclusion of churches from an otherwise neutral and secular aid program violates the Free Exercise and Equal Protection Clauses when the state has no valid Establishment Clause concern.”

The Court held that the Missouri program violated the Free Exercise Clause (and put off the Equal Protection Clause claim) because it precluded the church from participating solely because it was a church. The vote was 7–2, with Justices Sotomayor and Ginsburg in dissent. Justices Alito, Gorsuch, Kagan, Kennedy, and Thomas joined Chief Justice Roberts's opinion in favor of the church. While not overruling Locke, it considerably narrowed its force: Locke was about a future educational goal, not a current exercise of one's faith; and Locke was about becoming a minister, whereas the case before the Court was about paving a playground of a church's day-care program. Perhaps most importantly, the Chief Justice's opinion is much more in line with what I argue above as to how one should evaluate whether a program penalizes the claimant because of the exercise of religion. Here Rehnquist's argument in Locke is pushed aside. Missouri's application of its Blaine Amendment to block the church's ability to compete for a place in the state's playground paving program was easily framed as a penalty on the exercise of religion. Justice Breyer issued a narrow concurrence emphasizing that since this case was squarely about paving the playground of a preschool, his vote should not be taken to imply his opposition to Blaine Amendments generally. Justice Gorsuch wrote a concurring opinion, which Justice Thomas joined, urging a more sweeping holding in favor of the church.

To me, it is most encouraging (a) that Justice Kennedy backed off from his support of the “play in the joints” idea featured in Locke, and (b) that Justice Kagan also joined the Chief Justice's opinion. In her dissent Justice Sotomayor emphasized that the church clearly ran its preschool as a religious program. But this did not at all bother the justices on the other side.

I view this decision as a very promising first step in the direction of the argument I present in this article about faith-based schools seeking to become charter schools. But it would be wrong to assume that the Court has already decided the matter my way. Broad funding of day-care centers was not at issue in the case before the Court. Plus, the program the church sought to participate in was clearly about helping out qualified private day-care programs. I do not view those distinctions as insurmountable hurdles, but it will be some time before we see how this day-care decision plays out in the charter school context.

If I am correct that it is unconstitutional to adopt a pro-family-choice charter school program and then exclude religious charter schools from participation, then I predict that two important things will follow from such a legal decision. First, I assume this would not invalidate a state's entire charter school program under state law even if the state has a Blaine provision in its constitution. The easy way to get to this result is for state courts to conclude that, for purposes of the state constitution, faith-based schools that receive charters are “public” schools since they, like all other charter schools in the state, must take all comers, may not charge tuition, are subject to the terms of the charter, and are publicly funded. Second, I will assume that, as a political matter, states will not entirely repeal their charter school programs once they learn that faith-based schools must be allowed to apply. I base this on the growing political power of the existing nonreligious charter schools and their constituents who would fight hard to prevent their demise.

On these assumptions, would existing private religious schools then flock to become charter schools (at least in states that permit existing private schools to covert to charter schools)? I doubt that would occur at the outset. Many existing religious schools would want to retain features of their operation that would be disallowed by becoming a charter school, perhaps most importantly giving up the right to be selective in which applicants they admit to the school. Others would be nervous about trying to get into the charter school game, fearing that bureaucrats might be hostile to applicants whose right to a charter was won only through litigation.

Nonetheless, some religious private schools, I predict, would seek to convert to charter schools. I have in mind, for example, urban Catholic schools that today serve children from low-income families, many of whom are not Catholics. These schools may well be happy to comply with the charter school rules in order to gain public funding so long as they can remain Catholic schools. I also have in mind some underfunded Christian (or Jewish or Muslim) fundamentalist schools that currently enroll children of modest income families with strong religious beliefs. While some of these schools and the families affiliated with them would have nothing to do with government for fear of intrusion on their religious goals, others, I predict, would seek to become charter schools in order to gain the funding that would allow them to provide a much stronger education to their students than they are able to afford on their own. These schools would be happy to accept all applicants, realizing that, as a practical matter, few if any families without similar and strong religious beliefs are likely to apply; and, unlike private schools appealing to well-to-do families, such schools would have no felt need to charge tuition beyond the charter school funding level.

As yet another example, I predict that some of the existing faith-affiliated charter schools discussed earlier would convert to religious charter schools. Indeed, other groups thinking of starting or converting to a faith-affiliated charter school would likely be more inclined to take up charter school status if they could be openly religious schools.

After this modest—or moderate—rush to take advantage of religious charter school status, I imagine that the remaining existing private religious schools would wait and see how things turn out. Would religious charter schools be regulated in ways that religious groups find intolerable? Would religious charter schools thrive? That could well depend on politics, and the politics could well be changed if religious schools are allowed to be part of the system.

Of course, in states that neutrally preclude all existing private schools from becoming charter schools, a holding that it is unconstitutional to prevent religious schools from achieving charter school status would only enable new religious schools to apply for charters. ¹⁵² In those states it is difficult to predict just who would come forward with such proposals and from where they would draw their students. But it seems clear that, at the start at least, fewer religious schools would become charter schools than if the state rules permitted conversions from existing private schools to charter schools.

Still, I would expect some religious families now unhappy with the public schools their children attend but unable to afford to start private schools would create religious charter schools. I would also imagine that some urban Catholic schools would close and lease their facilities to new faith-based charter schools. These new schools might well appeal to many or most of the families whose children attended the former school. ¹⁵³

In closing I note that were faith-based schools permitted to be charter schools, then the structure of school choice in place in the states with charter school programs would be remarkably similar to the school choice plan that Jack Coons and I advanced more than forty years ago.