i : The Efficacy of Educational Vouchers* R Zimmer, Michigan State University, East Lansing, Ml, USA E Bettinger, Stanford University, Stanford, CA, USA, © 2010 Elsevier Lt, lights reserved Glossary Children Scholarship Fund - Scholarship program for students to attend private schools funded by the Walton Family Foundation Indirect effects ~ Erects on students not ulizing ‘vouchers to attend private schools. ‘Means-tested— A ctterion that uses the ability to ;pay as means of being eligible for a voucher. PACES program — A Columbian program otfered ‘educational vouchers to over 144 000 low-income ‘students throughout Colombia from 1992 to 1997. ‘Scholarships - While both vouchers and ‘scholarships are entitlements to attend a private ‘school, vouchers are publicly funded while ‘scholarships are privately funded ‘Seetarian schools School affliated with particular faith oF rligion ‘Selection bias — Estimated etfect of programs could be biased because participants have chosen to participate, whichis not captured by the econometric. ‘model. ‘Selt-selecting — Students choose to attend a private ‘school. Vouchers —A governmentfinanced enttlement that students and their families could use to attend any ‘school, including private schools. Introduction Advocates for vouchers argue that vouchers create incen- tives for schools to compete for students and that such ‘competition can make all schools efficient. Milton Friedman originally proposed vouchers as a government-financed entitlement that students and their families could use to attend any school, including private schools (Friedman, 1955, 1962). Over time, sup- post for voucher programs has come from diverse groups such as conservative pro-marker advocates (including Friedman), Roman Catholic bishops hoping to reinvigo~ rate parochial schools in urban cities, and minority advo cacy groups. we effective and “The review an abbreviated vention of or reson review (Zimmer and Betinger, 2007. Voucher opponents argue voucher programs cream skim the best students and inequities if selection into voucher programs falls along racial or economic divides, They claim that by public school enrolments, vouchers intensify the fiscal rains already felt by public schools. Debates over vouchers have increased in the last decade of s0 with the advent of publicly funded voucher programs in Milwaukee, Cleveland, and Florida, Addition ally, he inerease in the number of foundations and phila st vouchers can increase reducing thropists providing private money for students to attend private schools has also increased the public awareness of voucher programs. Outside of the United States, voucher like programs have also increased in popularity; and while the motivation for these voucher programs differs from those in the US, debates over these programs rage world- wide, In this article, we examine many of the claims and counterclaims of the debate by surveying the research literature on both domestic and incesnationsl programs. In reviewing the existing voucher literature, we focus exclusively on empirical evidence on vouchers and do not include a review of other forms of school choice includ ing charter schools, home schooling, or the effectiveness of private schools generally. While some research has ‘extended beyond actdemic and behavioral outcomes or the distributional effects of Voucher programs and exam- ined the effects of voucher programs on residential hous ing markets (Nechyba, 2000), cos effectiveness of voucher prograns (Levin, 2002), on subsequent public support for ‘vouchers, or lack thereof Brunner and Imazeki, 2006) and ‘on who wins and loses under voucher programs (Epple and Romano, 1998, 2003), we only discuss these broader effects to the extent that the studies shed light on the specific ‘voucher programs that we review ‘This is not the first review of vouchers programs (eg, “itomer and Bettinger, 2007, Gill eal, 2001; Levin, 2002 McEwan, 2000, 2008). These reviews include the concep= ‘alization of, and rationale fos, voucher programs as well as evidence onthe efficacy of vouchers. This article builds ‘on these reviews, but focuses on the achievement and behavioral effects of vouchers on students and the effece these programs have on access and integration Structure of Voucher Plans Voucher programs differ greatly in their specific design Much of this variation is related to the differences in che 343i : 344 _The Efficacy of Educational Vouchers motivation for vouchers among the voucher advocates (Levin, 2007 Handbook) For example, many conservatives believe that flat-rate, modest vouchers can increase effi- shrough increased competition (eg, Friedman, Other proponents argue that means-tested vouchers can promote equity by providing berter educa~ tional opporcunities to low-income families in inner-city schools, Programs also differ in their advacacy of subsidies for transportation, degree of information availability, reg- ulation of admission policies, and accountability mecha isms (Jencks, 1970, Levin, 2002). In practice, voucher programe differ substantially in their level of support, the target population, and admission of sectarian schools, “The various programmatic details change wit to who participates inthe voucher program and are likely to influence the generalizability of individual programs and studies. For instance, programs that restrict participation 1 low-income students are likely co have different disei- Dutional effects than 2 universal voucher program, Pro- grams that allow scudents to attend sectarian schools are also likely to attract a particular set of students ~ specifically students seeking religious instruction opportunities. Finally, the scale of the programs may affect whether these pro= ‘grams will generate general equilibrium effects that may be ‘even larger than the localized effects of voucher programs (eg, Nechyba, 2000; Epple and Romano, 1998, 2003). AAs the programmatic design can affect outcomes this icle, as in our previous review, highlights the details of ‘each voucher program and the associated outcomes indi- vidually, We synthesize che results in the final section and provide a summary of the programs and outcomes in Table 1 reference Domestic Voucher Programs Domestic voucher programs have been funded either by public taxes or private donations, primarily through phi- lanthropists or foundations. While the scale and regula tion of public and privately programs may differ, they are similar in that they both provide a subsidy for students to axtend private schools Publicly Funded Programs The Milwaukee voucher program The most visible domestic voucher program began in Milwaukee in 1990, The program focused on low-income students, The program started small being capped at 1% of the school distrie’s population, In examining the pro- gram, Witte (2000) compared the perf students to that of all other students in the Milwaukee school system finding no differences incest scores. Greene ef al (1997, 1998), found increases in reading and math test scores when they compared outcomes for ance of voucher voucher recipients who had stayed in voucher schools for multiple years to outcomes for a comparison group of students who had wanted to use the voucher but could not because of a lack of space. These divergent results led to an often heated debate over the results and methodol- ogies (eg, Mitgang and Connell, 2003; McEwan, 2000) Rouse (1998) attempted to resolve the controversy: Rouse took advantage of the face that oversubscribed schools used randomization to assign vouchers. (McEwan (2007) discusses how randomization is the gold standard in eval- uuative research). Rouse found slight positive, significa effects of the voucher program in math test scores, but none in reading. Milwaukee's vou et experience also sheds light on the impact voucher programs can have on access and racial integration, African-Americans made up over 62% of the voucher population, Hispanics made up about 13% (WLAB, 2000}, The program successfully targeted low income families with the average family income for pro- gram participants being around US$11 600 (Witce, 2000 Other snudies showed that the voucher parents had slightly higher educational levels (Rouse, 1998; Wite, 1996, Wiete and Thor, 1996) than other parents in Milwaukee. While the early Milwaukee program offered insights to policymakers, the small scale of the program limited the policy implications. In 1995, the state of Wisconsin greatly expanded the scholutship cap and the range of private schools that students could attend. These changes dramatically redefined the scope of the program. While only 341 students attended seven private schools in 1991, over 15000 students attended dozens of sectarian and non- sectarian schools by the 2004-05 school year (Gill etal, 2001). This expansion led some researchers co investigate ‘whether Milwaukee's program resulted in changes in nearby public schools (eg, Hoxby, 2005). Unfortunately, the ex panded program did not require testing within the priv axe schools, which resulted in no new studies of the program. ‘commissioned study (funded by the Bradley and. Joyce Foundations among others) seeks to reevaluate the Milwaukee voucher program by having private schools voluntarily administer a standardized test The Cleveland voucher program In 1995, the Cleveland Monicipal School District began The program was initially intended to allow vouchers to bbe used at sectarian schools; however, courr decisions immediately blocked this provision. Not ur years later were students allowed to choose among both sectarian and nonsectarian schools In 2007, Cleveland’s program served over 6000 stud Cleveland's voucher progr answers about the program's impact on academic achieve ment, As in the Milwaukee progeam, the problem with the evaluations was whether the researchers used an ade~ {quate comparison group. The most often cited evaluation n has not yielded clear305, Tho Efficacy of Educational Vouchers punoiByeea ex 800 uepns ue Bupuedep Lo pusdap syoaya s198n uo 2249 ON 201008 ju oBBas0U4 opiartuayers ON spooye anton si98n Uo 819949 ON 501098 180) aseo10u) spi-worsig onenpesé pouse YB ue sao onnsod vonenpesé tooyes yBiy ue soy annsog ‘i601 pue porsjdisoo ‘1604 jooyos Uo 532949 349504 “suaiines 1o42non Bue “suepms 1 95 fue 0) 2842 ON swuapris yx so0ye aansog oeig Buowe A syonues yum suotssouBoy sjonueo im suojssouboy sjonueo yim suolseeuboy, sjonuco yum suotssouboy uonedeSBe:sos1ueo yim SueIsaAoy Jepou uonostes veuro0y) sjonueo yim suojssaib0y ‘souaserd _euan0n 0} ofgeyes eWaLNNU suayonon jo vane zope ‘vayonon jo uojeziopuey, siayanon jo uoqeziuopuey siasn teyonon siosn toyonon siasn seyanen siasn 1oyonon worse e080) issn soysnon si9sn anon, worsAs 89% ‘sjo0y9s jeuonea0n 1 siosn soyenon issn soyonen siasn eyanen siesn soysren siasn soyenon siasn eyenon (002) wewyo. (002) owes pue ue aoWy tooz) vemaon (cop2) ezanyues pu ‘seiaauon ‘onese (002) einbin pue vost (9661 (oz) Pa (661) tion ewoy pue e204 (6002) e631 eno, (e002) eipenees pur sowery e6uneR (e002) ow pue vebunieg yeabuy (2o02) 8p ue vouery woojg vebunieg suBuy quire (©002) ums pue (02) naz pur 0694 ‘@o02) uostajeg puE namon (z00%) 18 4EHY eiBoxdsayonon aun wou yoaye apn siasn eyenon (2002) 150H2 oN, si9sn soyren (00% Ho pucionaig 24 010 mai 129} ‘sjonueo oyeuenes og ue Kuo wes Us eiaaye 1uEDHIS yu ubtsap reyuousuodso-isenD siasn soyanon (2661) esnou wew ue Supe: won u sioay weOMIUoIS ubtsap jeqususuedee se siasn oyonon (96611661) o€sBLEIP Losiaag ‘eu8910 oy fenueisans oN, ‘sjonuee yim suojssaiboy si9sn ayanen (os61) 0m aaxnennn ‘sas hoy, ‘ABayens woneaunuepl uo si20ye seinseony siouiny ey 842R0n ‘iayonen uo Warees uo ARUURS Fe 7 poses eu akan oc neni : 346__The Efficacy of Educational Vouchers of the program (Metcalf et al, 2003) compared the aca- demic achievement of voucher users to two groups of students: (1) students offered vouchers who did not use the vouchers, and (2) a matched set of students who were not offered vouchers. Other researchers, including Gill al. (2001), argued that the study failed to create an adequate control for students selfselecting into the pro- ‘gram, Belfield (2006) used a different comparison group focusing on applicants who were rejected in their applica~ tion to privaze schools, Overall, he finds little effect on academic achievement ‘Another key motivation for establishing the voucher programs was ta improve students’ access to better schools Kim Metcalf (1999) found that the program success- fully caxgeted minorities and students from low-income families, The mean income level of students uclizing the vouchers was US$18 750 and these seudents were more likely to be Aftican-American students than a random sample of Cleveland students. As in Milwaukee, parents ‘of voucher applicants had slightly higher levels of educa- tion (Metcalf, 1999; Peterson et al, 1999), Other publicly funded voucher programs Voucher programs are also in place in Vermont, Maine, Florida, and Washington, DC. Of these programs, only DC's Opportunity Scholarship program has any publicly released results, which are only the first-year impacts. ‘The Institute of Education Sciences-funded evaluation ‘of Washington's program is using a randomized design to evaluate the effectiveness of the program, ‘To date, the results suggests chat parents are mote satisfied with the program, but students utilizing the vouchers do not report higher levels of satisfaction of feeling safer In addition, text scores for these students are on par with nonvoucher seudents (Wolfe ul, 2007). However, caution is warranted in interpreting these results as they are noted only after the first year of implementing the voucher program. Domestic, Privately Funded Programs Roman Catholic dioceses have provided scholarships to students to attend their schools for years. The use of philanthropists and foundation money for private school scholarships was first popularized by J. Patrick Rooney, sho formed the Educational Choice 1991 This trust allocated private scholarships to low- income students to use at private schools in Indianapolis. Other progeams in Milwaukee, Atlanta, Denver, Detroit, Oklahoma City, and Washington, DC, soon followed. Some of the programs were established and financed by conservatives who wanted to either open up sectarian school options to more students or to create competi= tion berween private and public schools. Other programs (eg, Oklahoma City) were funded by liberal voucher Sharitable Trust in nts who wanted to provide additional schooling ities for inner-city In 1994, the Walton Family Foundation helped create the Children's Educational Opportunity Foundation (CEO America) to support and create private scholarship pro- grams, Shortly thereaites, the number of privately funded programs quickly expanded across the country. Of these programs, one of the most ambitious was a CEO America program in San Anconio funded at US$S million a year for at Jeast 10 years. The program provided fall scholarships to over 14000 ater students making it the largest pri= vately funded voucher program in existence at the time. In 1998, the Walton Family Foundation helped create the Children’s Scholarship Fund (CSF), CSP partnered with local funders ¢o provide scholatships in cities across the nation, In April 1999, 1.25 million low-income students applied for 40.000 partial scholarships, Over the las decade, more and more of the programs developed, often receiving very little national artention, making it difficule to know ‘exacily how many of the programs currently exist. Of these programs, CSF has received the most national astention partially because of ts relatively large scale and partially because CSF used randomization to asign scho- larships. With randomization, researchers cen compare scudents who participated in the lotteries to determine the effectiveness of these programs, Zimmer and Bectin- ger (2007) discuss some of the challenges in these studies.) Other studies have been cartied out in Charloxte, Dayton, New York, and Washington, DC among other locations While Greene (2000) found a positive overall effect on test scores in Charlosee after 1 year, these studies have generally shown no effect. The one exception is among ‘Affican-American students, Howell and Peterson (2002) and Myer ef al (2002) found that African-American stu dents tet scores improved in New York's program. How- fever, additional research by Krueger and Zhu (2002) showed that this research was sensitive to the definition of an African-American student. While the original anal- ysis used the student's mother’s identified race for the scudent’s race, Krueger and Zhu define African-American seudents as those whose mother or father are identified as African-American, This more inclusive definition results ina scholarship effect size that is smaller than reported in the original study and is not statistically significant In 2 more recent study, Bettinger and Slonim (2006) focus on the effects of vouchers on nonacademic ou comes. Using surveys and new methods from experimen- tal economics, they attempted to measure the effects of the vouchers on traditional outcomes such as test scores and on altruism, ‘To test the effects of the voucher pro- gram on altruism, Bettinger and Slonim gave students US$10 each and invited them to share some of their money with charities. In this context, they show that voucher recipients gave more to charities as a result of the voucher program, dentsi : Tho Efficacy of Educational Vouchers 347 Given that these programs were designed co target impoverished students, it is not surprising to find that participants have selatively low income, Peterson (1998), for example, found tha the average income level of families of scudents participating in the New York schal- acship program was only USS10 000. The parenss' educa ton level ofthe participants, however, was slightly higher than the average level ofthe eligible population, Similarly, Howell and Peterson (2000) and Wolf etal (2000) found that the families of participating students were low income. In terms of race, 100% of students participating in the Washington, DC scholarship program were mino- tities, while 95% of the students in New York were minority, mostly Hispanic students. In Dayton, the per- centage of minorities was lower bu was still 66% Finally, these privace voucher programs cypically serve a low percentage of special education students. For ex ample, in New York, 9% of participating students have disabilities compared to the district-wide average of 14% fof special education students (Myers ef al, 2000), In Charlotte, 4% of participating students had disabilities, while the distict-wide average of special education stu- dents was 11% (Greene, 2000), International Voucher Programs ‘There are a number of educational voucher of voucher- like programs across the world, including programs in Chile, Colombia, Sweden, Netherlands Belize, Japan, Canada, and Poland, These voucher programs often differ significantly from the US programs either in terms of the motivating factors of the policy context For example, many of the non-US programs are motivated by the goal of increasing school attendance among giels of low-income students. The relationship beeween church-sponsored pri- vate schools and public schools is also less defined in other countries where, in many cases, religious groups operate public schools We focus our attention on two programs that have figured most prominently in debates on the efficacy of vouchers — Colombia's Programa de Ampiacion de Coberturs de a Educacisn Secundaria (PACES) voucher program and Chiles national voucher program, ‘Colombia PACES Program ‘The PACES offered educational vouchers to over 144000, low-income students throughout Colombia from 1992 to 1997. The voucher program awarded full private school tuition to secondary school students who wished co trans- fer fom public to secondary school. Over time, the voucher’ value did not keep up with the cost of inflation. Tn contrast to the US programs, the motivation for Colombia's voucher program was to expand the capacity vate school at the start of their of the public school system. In Colombia, most high school buildings host multiple high school testions per day. Since most private schools were not overcrowded, Colombia established the voucher program co take advan- tage of excess capacity in the private system, Tn cities where PACES was oversubscribed, cities held loweries to award the vouchers, Angrise «tal (2002) use this randomization to measure the effects of the program. ‘They find that within 3 years, voucher scudents had com- pleted about 0.1 years of schools more than their peers and had test scores about 0. standard deviations higher than those who did not receive vouchers. They also find that the incidence of child labor and teenage marriage was lower as a result of the educational voucher. Based on follow-up research on students’ high school careers, Angrist et al (2006) find that voucher lottery winners were 20% more likely to graduate from high school than voucher lorery losers. In Colombia, the evidence on the impacts of statifica- tion is less developed and conclusive chan the effects on achievement. Colombia's educational vouchers targeted low-income families living in the poorest neighborhoods, and Ribero and Tenjo (1997) report that the vouchers were largely effective in reaching this population. Voucher applicants generally came from families with higher edu- ‘cational levels than other families in the same neighbor- hoods (Angrist of al, 2002), Chile In 1980, Chile embarked on an ambitious series of educa tional reforms designed co decentralize and privatize edu- cation. Ar the urging of Milton Friedman, who along with other prominent economists advised the Pinochet govern ment at the time (Rounds, 1996), Chile established per haps the world’s largest ongoing voucher program, The program offered tuition subsidies to private schools. Before the reform, the budgets of public schools were insensitive to enrolment, however, ater the reform, local public schools last money when students tansferted to voucher schools Unlike Colombia, where vouchers targeted poor seu- dents, the Chilean syscem was available to all sudents Additionally, unlike other programs that donot allow selective admission, voucher schools in Chile could admit the students they most preferred. As a result of these policies, Rounds (1996) found that the poorest families were less likely to attend voucher schools Research on the efficacy of Chile's voucher program has been much more difficule co interpret than Colombian research because of the nacure of this admis sion policy. Some of the early evidence suggested that voucher schools modestly outperformed public schools. ‘This finding was common in many papers (eg, Bravo eral, 1999; McEwan and Carnoy, 2000) bu was sensitivei : 348 The Efficacy of Educational Vouchers to the types of controls included in the empirical model, the specific municipalities included in the sample, and the statistical methods used. McEwan (2001), for example, found that Catholic voucher schools tended to be more effective and productive than other schools, but only for certain specifications of the model. McEwan and Caznoy (2000) show that academic achievement is slightly lower in nonreligious voucher schools, particularly when located outside the capital. Given that thar these voucher schools have less funding, however, McEwan and Carnoy suggest that they could still be more cost effective th public schools As research on the Chilean system continued, many. researchers took note of the fact that the voucher program akered the composition of both public and private schools. For example, Hsieh and Urquiola (2003) showed that more affluent families took advantage of the vou- chets, Similar to McEwan and Carnoy (2000) and Carnoy (2000), Hsich and Urquiola suggest that this shift in stu lent populations could account for the finding that pri- vate schools appear to be more effective than public schools. In the early 1990s, many voucher schools began charging twition in addition to the voucher, and the difference in parents’ incomes and education levels between these tuition-charging voucher schools and the cther voucher schools was significant (Anand etal, 2006). How this increased sorting across voucher and non- voucher schools affects student achievement depends on the nature of peer effects. If improvements in peer quality Tead to better educational outcomes for voucher users, then the sorting associated with the voucher program could improve their outcomes. At the same time, given the exit of high-quality students from the public schools, the students left in those schools may have systematically worse outcomes because of the peers, The aggregate effect of the voucher depends on the strength of these two effects Hsieh and Urquiola (2003) argue that the only way to idencfy the overall effects of the voucher program is (0 focus on aggregate outcomes because it is dfficuls, if noc impossible, to remove the selection bias inherent in com- parisons of different schools. The change in aggregate test scores reflects both the direct achievement effects for voucher recipients and the indirect effects for students sho remain in public schools. When the authors look at cchanges in aggregate test scores throughout Chile, hey find no evidence that the voucher program increased ‘overall test scores, Gallego (2005) provides other evidence on the Chilean voucher program, Gallego uses an instrumental variable approach to estimate the effects of the program. His findings suggest positive effects of the voucher program fon the academic outcomes of students throughout muni- cipalities where the voucher program had more penetra- tion, While the result may be indicative of competitive leterioration of their effects, it is driven in pare by the effects of the program on voucher recipients It echoes earlier research (McEwan, 2001; McEwan and Camoy, 2000) which suggested th voucher schools alfiliated with Catholicism had better outcomes than other voucher schools, public of private As McEwan and Carnoy (2000) show, Catholic schools produced berter students ata lower cost chan other, public or private, voucher schools, Summary and Conclusion Table 1 summarizes the achievement findings from both sic and international voucher programs. The table highlights the location, program, methods used in the analysis, and theit key findings Research on domestic voucher programs has produced inconsi sults, While advocates have often pointed ta positive effects for Aftican-American scudents in the CSF program as proof that vouchers can work, che serengeh of these findings has been questioned ~ leaving many skeptical about the purported benefits of vouchers International studies have also found mixed results, The dome ‘evidence from Colombia indicates that voucher users have better academic and nonacademic outcomes than they would have in the absence of the voucher. In Chile, the estimated effects differ across studies, and a number of confounding factors make it difficult to ascertain the true ‘effects. The research generally shows chat students who take advantage of vouchers ate disadvantaged students, ‘especially in voucher programs thar are means tesced and specifically target such students “To date, most of the voucher sndies focus on programs that are typically small in scale, Hence, it is not clear whether the results generalize to large-scale programs. For a program with massive movements of students, an important policy consideration is not only who is partici- pating in these programs, but what happens to the racial/ ‘ethnic and ability distribution of students in public schools In sum, researchers have failed to come to consensus on the efficacy of vouchers as a reform effort. This murki~ ness may be clarified as new studies emerge, including seudies of che current Milwaukee program and the feder- ally funded Washington DC vouchers, Future research needs to peer inside the black box and to examine the mechanism in which outcomes dif- fer among schools. Voucher researchers should take the opportunity to look at private schools that may be doing things differently and how the variation in the operation affects outcomes. Additionally, future research can shed additional light on the effecrs other critical outcomes such as individual behaviors, edu cational attainment as measured by college attendance, and wages, of vouchers oni i i ! ‘The Efficacy of Educational Vouchers 349 Ic is also interesting to consider the value of vouchers relative co other forms of choice. When the idea s were first introduced, the only choices a family could make was sending their students to private schools without subsidies or choosing among various pub- lic schools based on residential location. However, over the last 50 years, a number of other alternatives have evolved including charter schools, magnet schools, and other inter- and intra-district choice programs. A relevant policy question has to do with the advantages and disad- vantages of voucher programs relative to these other choice options, expecially when many of these choice options are politically more feasible. ‘See a/so: Competition and Student Performance; Educa- tional Privatization; The Economic Role of the State in Education; The Economics of Catholic Schools; The Economics of Parental Choice, Bibliography ngs J, Betinger. Bloom, €. Kemet, Mane “ing 7002). The eect a choot vouchers on stuns Exisenc rom Colombia, American Econom Reiow 925, 1595-7558 Anes J. Benger, ana Kremer, M2008), Long-term easton Consequences of secondary school vouchers: evance fom ‘uinisratue recorasin Celemba, Amercan Econemie Review 9663, 847-62. Batol C. 2. 2008, The evidonce en education voueners: AN ‘apaleation to the Cleelans senolarenp an! worng program, epi ncspo.orgfpubleatons sles(OP112 ga accessed shine 2003, Botte. E. ac Slonim, R206). Using expormonta economics to ‘measre tn eee fanaa educational experiment on ara, “Touma of Puc Economies 808-8), 1625-1648, Bravo, D, Convers, D. anc Sanhvezs,C. (1898, Renamento ‘scala, ceuguatsd ybecha de desemnpara privsda/pubiee: Chie "982-1997. Docunento de Taba, Desartamento de Econom’, sso Cie, 1988, Bearer fan maze J (2008 Tit choice andthe voucher. Wiking Pape. Unversy of Connecticut Econemics Devarer: (Ohute, J. and Meco. T. M1960. Potis, Market, and America's ‘Schools. Washington, DO: Brookings Istution ‘eons 5 Ean Sugrman, SD. 1978). Eoueation by Chace: The {Case fr Famiy orto kel, CA: Univer of Calfria Press, Epp, D. and Romano, 2. 1888). Carootionbetioon pate and Dube senooks, vouchers and poe group eves, American Ezaname Rene 88, 28-63 Epp ©. and Romano, A. 2009). Neghacrneed sero, chace, and ‘ne astrenuton of ecicatonal oe, n Hob, CW) The Ezonames of Schoo! Ghoee. pp 227-286. Chicago, L: The Unies of Cicage Poss Fedman, Wt, 958). The cle? govemmont education. In Slo. A ied) Beers andthe Publ terest p 122-144, Pecan, Nu Rugors Unersy Press. Finan, M1962), Caplaien and Freedom, Cncge, I Univer of Chicago Press CGallago, FA 12005), Vouchar-senal compton, canines, and ‘uteomes,Eucence tem Chl RY mimeo GiB, Tinpane, Pt, Ross, KE, and Brower. J 700) ‘Rhetoric versus Resiy: hat Hf Krow ad What He Neos to ‘row about Youchors and Charter Schoo's. Santa Meriea, CA: END Edueaten ‘Greene, J. (2000), The Elect of Shoe! Ghoee/n Exauaion of the ‘Gharote Chen's Scholarship Fura, New York: Cerio or Gc Ihnovaten atthe Manhatan rai, Creane,J.>_ Peterson, P. =, and Du, J 1007). Beciveness of Schoo! ‘Choe. The Mivaukee Experiment. Cambrcge, MA. Haars Unoraty rogram in Eaueatom Ply and Governance Green, JP, Person PC, an Dx, J (1988, School choice 'r Mluaukoo: A randomized experiment. Peterson, >. Ean Hasso B.C. feds Leamng tom Schoo! Chace, op S85-358, ‘Washington. 00: Brockngs Istain, Howe WG. and Peterson, £2000) Schoo! Choe n Dayton, ‘Ohio valiabon alr Ore Yar, Cambrage, Wie Program o9 Es.caton Pokey and Govemanco, Harvard Unversty. Howe WG. and Patrsan >. E2002). he Education Gap: Vouchors {ara Liban Schools. Wlashingon, DC: Boakngs ste Helen Cana Uru, M, 2003) When soncole cornet, how co thay ‘compete? fn assossmnto! Che's natonniceschoal voucher rogram. NBER Worn Papors 10008. nets, C1870) Esucatoral voushars: A report on financing ‘mentary educaton ay grants ‘c parents Hep! Prepared under (Grane CO 8549 er the U.S. Orice ef Eearame pear, ‘ashing, 20: Cone ‘r Pokey Stes. ger fe 8. an 2h (2002). Arother Leek a he New Yor ity ‘Shoo! Yeuchor Exoarmont Wek Paper 8478, Cambie, Mi [atonal Buren of Eeoncmic Reseach, Le. H.2002) A comaranensie Famewk or ealusting ecuationl ‘ouster EducatonalEvatuston and Pay Arai 24), 158-174 ove (2007 seus educational pivatzaton. ad Hane ake.) Handbook o Researen n Edveaton Pearce arg Potcy pp 447-466, Now York Amecan Edvcation France ‘esciaton/teaum, McEwan. Ps (2000. The potental impact of large-scale voucher ‘programs. view of Fascatona Resoareh 70, 103-128. MeCian PJ, (2001, The etecthenese of publ, Cahole ard ‘ponolgiis ple sxnoolsn One's voucher ayer. Easton Econamis 92) 102-128 ‘McEwan Ps. [200% The potental impact vouchers. Peabody “urna of Eucaton 798, 87-89 McEwan, and Camoy, Me (2000). The etlecveness and eficeney ‘of pate schools Gale's woucher sya, Edueatonal Evan ana Potey Aralys 2, 213-288 Meta, KK 7909, Evasion of he Cleveland Scholae and Tutoring Grart Program: 1995-1999, Bleringzn, IN Tre Ilana Genter or Evaluation Moica. K. Wes, SB Logan, Ns Paul Ke Ma aed Bo0no, Wd "2003 Evalion of ne Clevelind seholarehip ans rng pega, ‘Sanmary Report 1998-2002, Shoringin, Nha Unveresy School ot Fasten, ‘tgang |. D. and Cannel ©. V. 2003). Eeluctonl vouchers anc the ‘ec. Lev Mfc) Prvating Educator: Can he Marketplace Delvor Ghote, Ener, Equty, ard Social Cohesion? bp 20-38. Sauer, CO: Wessew. Myers, Peterson. Mayer, D., C204 J, and Howell WG. 2000 ‘School hoicein New Yar iy ater Two Yeas: An Evaluation of he ‘Scheo! Groce Schaar Program. Washegton, OC: Mathematics Potey Research ‘Myers0.P, Person... Miers, D.E. Ttle,C.C. ancHowel. WG. 1200). Sihecl choba intew Vor Cry ater tee yours. Anevaaion ‘ofthe schol choos scholarps progam. Rep. No. 8404-045, Pancoton, Nu thera Phy Reserch. [Nechyba, T2000. Moby, ageing a rate school vouchers. “imorcan Econemic Rie 80, 730-146 Peterson, P.E (198). An Evaluation ofthe New York ty School ‘Shoe Scholwshiog Program: The Fst Year. Cambriage. MA Progam on Eascaon Ploy and Governance, Harare Unser Peterson. P., Howe W., and Groene JP (1999).An Featation ofthe Clovoland Voucher Progra afte Tv Yeas. Catrige. MA Program on Eoveaton Poly and Governance Hanara Un, ‘ibe, Rand Tena, (1997), Unversty de los Andes, Daparment ot Econemios Wering Paper. Pounds, P."-(1996) Wi pursut of Nigher quslysacrite equal ‘opperuy mn eduation? An anak fhe educatianal voucher system fy Santago, Sci Slonee Guaron) 77s, 825-845,ipo, Pun, Ali sed 350__ The Efficacy of Educational Vouchers rouse, Cx, 1988), Pras school vouchare and student achievement "revaluation of he Mila.koo pavovtlchace progr. Qoarteny ‘loumal of Economics 143, 952-802, \Wesconan Legiabve Aut Sureau WNLAB) (200). Miwaukeo Parental ‘Shove Progam: An Evauston. Madison, Wi Legilatve Aut Buea. ite, JF. {996). Who bene from he Miva choice program? Flo, @., mcr, AF, aed Ore. ous) Who Chooses, Whe Loosee? Cute, nstitene, andthe Unequal Etect of Schoo! {Ghoice, po 118-187. New Vere Toacrors College Press. ite F. 2000). The Maret Aperoach fo Eeucabon Pnceton, Nb Princeton Universi Press ite JF ana Thom, CA 1886. Who chooses? Vouchor and intr chole programs in Wiiaukes, American Joatal of Easton 104, 187-217. Will P, Gutman, B., Pura. Mf. lzo, Land ssa N (2007, valistonof.0. cppertinty sohatrshi progres pacts ste ‘ono yar ratte ol Eaueaton Seance, US Deparment o! Eaueaton NEE 2007-4009, Wo P, Howl, WG, and Patrson, .E, 2000, Schoo! ‘Ghoie in Washington D.C. An Essuatan afer Che Year. ‘Gamorage, MA. Program on Educaton Pay and Govemance, Harare Univer Zeer, Hand Bette, E (2007. Bayer hereto: Surveying he evence cf vouchers and tax eects. Lads, H. ard see ds Handbook of Research Faveaton Finance and ley. pp 447-266, New Yorke Amercan Edacation France ‘Aesociatonvetoaum, Further Reading ‘ele, 0. A. and Westen .L (2008, Eoatinal tax ced for ‘rmoia's school, NOSPE Paley Brot 3. phwanesce.crgl ublctons les/P808.fccessed Jane 2008, Bertnger, F. Kremer, W. and Saavedra, 2008) How eduestenal ouner aloe! sliders the ree ol vocalona ign senooks Colombia, Case Vester Resere ries, ‘cornay, M. (1808, National yousna Glan n Chil and Sweden ie ‘vata sare make fr beter ecueston? Carpets Edveaten Review 8219, 309-537 CCoeran, J Hr Ta Kine, 8. (1882 Cognitive ‘uleeres in pub and pate schocls, Sociology of Education 55.65.76, vans, WN. ane Sehwag, P1895) Finishing highschool ane sath ‘lege: Do Cathal shcoke make a dterence? Quarry Joumal of cones 1208), 881-974, Fao, ©, and Lucia, J (2000) Sex, drugs ant Oatrote chooks: Privat schol rd non-maraladolescon| behaves, NBER Working Paper No. 7990, Noverbor 2000. Ca Band Bosker, K. 2007, Sencl competion and studs fouleomas, In Lad H.F an Pace 8 ods) Handbook of Research Eduction Finance and Poy, po 185-702. Now York ‘rraran Caveaton Finance Mesecalen/=Hoaum, ‘Groone, JP. and Hal, 12007). The GEO Herzen Scholashio Progtam: A Case Siua’of School Vouchers inthe Edged Inagpendect Schoo Distt, San Anton, Washingon, D> Natheraea Peley Resear, Howe WS, (2008, Dynamic selon eects n mear-tested, ban ‘Schoo veer progres, vounal of Pley Anas and Management 282}, 225-250. king, Ratlngs,L. Guar, M. Pars, ©. and Tones, © (1897, "Colombia's targeted educaten voucher rogram: Features ‘covrage and parson. oka Pape No.3, Sais an Ire Essien o Eaveaton Anvorms, Ooveloomon! Esanomice Fesoaren Group, The Wits Bank Lad HLF. anc Hansen, J 8. ed 1098. Making Maney ater: rancing Amenca's Schoo. Washes, DO: Naonal Acacery Press. oun Hi 1988). Cluctonl voucher: lectveness, choles, and ‘cost, veumal of Polcy Arabi and Management 17}, 372-392 Lenn H (1809) The publinpevate nexus fh ecucaton. American ehavcal Scents 48), 124-137 Peterson. PE. Camobel,0.£, and West, M.R. 2001). Ae Evaluation ‘ofthe Basie Fund Seholrship rogram the San Farsco Bay ‘rea, Calor, Cambri, MA: Program on Education Poly and overnanen Torr, .F (1096, Publ funding and orate schootng acoss ‘Saintes lou ofa and Fecramice 88, "21-148, Relevant Websites hepsi tvookngs edu —Brockngs Ist, heb /tece‘ern sem ~ Carr fr Ecucaton Reform. psi ecweaiorg ~ Eduesion Week ‘pnt racaesoundton org ~Fredran Fourdston fer cational rae, tp 2wen8a.r9— National Education Associaton ‘rept ncape.o9 ~ National entero the Sty of ration in Fovsaton, ‘sep vancerlt.ed— Varderis Nsonl Centar on Scho ‘Shove,