Sound research has demonstrated consistently that school choice policies improve public school performance.
Thirty-one empirical studies (including all methods) have examined private school choice’s impact on academic outcomes in public schools. Within that body of research, 29 studies find that choice improved the performance of nearby public schools. One study finds no significant effects. To date, no empirical study has found that school choice harms students in public schools.
Four recent research studies support this conclusion:
- A 2016 study by Anna Egalite of North Carolina State University looked at the impact of the Louisiana Scholarship Program (LSP) on Louisiana public schools. Egalite found, “The competitive threat of the LSP ranges from negligible to modestly positive in the public schools exposed to the threat of competition, with effect sizes growing in magnitude as the competitive threat looms larger.”
- A 2014 study by David Figlio and Cassandra Hart of Northwestern University examined the competitive effects of the Florida Tax Credit Scholarship Program on public schools. They learned that more access and variety of private schools increased the competitive pressure on public schools in the wake of the policy announcement. They state in their conclusion, “The fact that we observed generalized improvements in school performance in response to the competitive threats of school vouchers, even in a state with rapid population growth, suggests that voucher competition may have effects elsewhere.”
- A 2011 peer-reviewed study by Jay Greene of the University of Arkansas and Marcus Winters of the University of Colorado – Colorado Springs looked at the impact of Florida’s McKay special education voucher program on Florida public schools. Greene and Winters found there was approximately a “12 percent reduction in the probability that a fourth- through sixth-grade student” was diagnosed with a learning disability in a public school with average levels of competition. They also found that “being in a public school surrounded by the average number of McKay-accepting private schools was related to an increase in academic proficiency of about 0.01 standard deviations in both math and reading. The positive but very mild competitive effect is consistent with what has been found in previous research evaluating more conventional school choice policies.”
- A 2009 study by Jay Greene and Ryan Marsh of the University of Arkansas considered the systemic effects of expanding school choice in Milwaukee. Greene and Marsh found that public school students in Milwaukee fare better academically when they have more free private options through the voucher program. They concluded, “It appears that Milwaukee public schools are more attentive to the academic needs of students when those students have more opportunities to leave those schools. This finding is robust across several different specifications of the model.”
MYTH: Vouchers hurt public schools by taking only the best students.
Many people are concerned about the impact school vouchers will have on public schools. One concern is that voucher programs will “drain money” from public schools. Another is that they may result in “creaming,” a situation in which the brightest students use vouchers while the students who are hardest to teach stay in public schools.
In addition to fears that vouchers will harm public schools, there is also a related contention that vouchers will not have as much positive impact that has been claimed. Some have argued that vouchers cannot spur public schools to reform because public schools are too weighed down by bureaucracy, unions, or other barriers to change.
FACT: Vouchers improve public schools by providing choice and competition.
Although evidence showing that vouchers improve public schools is counter-intuitive to many people, it is not hard to explain. One reason vouchers improve public schools is that they enable parents to find the right particular school for each child’s unique educational needs. Children have different needs and preferences, and everyone’s schooling experience can improve if children are allowed greater freedom to find the right niche.
Vouchers also provide positive incentives for responsiveness and improvement that are lacking in the traditional public school system. When public schools know that students have a choice and can leave using vouchers, those schools have a much more powerful incentive to improve their performance and keep those students from walking out the door.
EVIDENCE: Data confirm vouchers serve disadvantaged students well and improve outcomes.
The available evidence suggests that voucher programs do not “cream-skim” the best students. To the contrary, the best analysis of this question found voucher applicants in three cities and a representative sample of the eligible population to be virtually identical on a variety of demographic and educational indicators.1
The acid test, however, is what actually happens to public school outcomes when vouchers are implemented. A large body of high-quality empirical research has examined this question, using statistical methods to isolate and measure the impact vouchers have on academic achievement in public schools (see accompanied chart).
In some cases the student improvement gains under vouchers are only moderate. That’s not surprising, given that many existing voucher programs are limited in the number and type of students they’re allowed to serve and the amount of choice they’re allowed to offer. Narrowly constricted programs produce narrowly constricted results. To produce revolutionary results, we would need broad programs—eligibility for all students.
Overwhelmingly, studies have found that vouchers improve public schools. No empirical study has ever found that vouchers harm public schools.
Reviewing the research: systemic effects of vouchers and tax-credit scholarships
|Anna J. Egalite, The Competitive Effects of the Louisiana Scholarship Program on Public School Performance, Louisiana Scholarship Program Evaluation Report 4 (Fayetteville: Univ. of Ark., School Choice Demonstration Project, 2016), http://educationresearchalliancenola.org/files/publications/Report-4-LSP-Competitive-Effects.pdf.||Looked at student-level public school test score data from 2010–11 through 2012–13 for students in grades 3–8 and applied four different measures of competition to estimate competitive effects; found that each additional private school located within a 10-mile radius of a public school is associated with a .0011 standard deviation increase in math scores. The author also found that each additional private school located within a 5-mile radius of a public schools is associated with a .0023 standard deviation increase in math scores.|
|Anna J. Egalite, “Competitive Impacts of Means-Tested Vouchers on Public School Performance: Evidence from Louisiana and Indiana” |
(PEPG-1405, Program on Education Policy and Governance, Harvard Kennedy School, Cambridge, MA, 2014)
|Analyzed student-level public school test score data from 2008–09 through 2011–12 and found that each additional type of private school within a five-mile radius of a public school is associated with .0038 standard deviation increase in language arts scores.|
|Daniel H. Bowen and Julie R. Trivitt, “Stigma Without Sanctions: The (Lack of) Impact of Private School Vouchers on Student Achievement,” Education Policy Analysis Archives 22, no. 87 (Aug. 2014), doi:10.14507/epaa.v22n87.2014.||Exploiting exogenous variation from the Florida Supreme Court’s decision to declare the voucher component of the Florida A+ Accountability Program unconstitutional, the researchers use a regression discontinuity model to detect any impact that threats from school vouchers may have had on public school performance. No evidence was discovered in the data.|
|David N. Figlio and Cassandra M. D. Hart, |
“Competitive Effects of Means-Tested School Vouchers,”
American Economic Journal: Applied Economics 6, no. 1
(Jan. 2014), pp. 133-56,
|Greater degrees of competition are associated with greater improvements in students’ test scores following the introduction of the program. Authors also found that schools expected to be the most sensitive to competitive pressures see larger improvements in their test scores as a result of increased competition cultivated by the tax-credit scholarship program.|
|Cecilia E. Rouse, Jane Hannaway, Dan Goldhaber, and Figlio, |
“Feeling the Florida Heat? How Low-Performing Schools Respond to Voucher and Accountability Pressure,”
American Economic Journal: Economic Policy 5, no. 2
(May 2013), pp. 251-81,
|The study used a regression discontinuity model to compare high-scoring F schools (whose students can receive vouchers) and low-scoring D schools (whose students cannot receive vouchers). It found that receiving an F grade in 2002–03 produced academic improvements in students’ test scores in the next year relative to those in non-F schools, and that these improvements were sustained in future years. They presented their results in terms of standard deviations rather than test score points; they found that the gains were equal to about a tenth of a standard deviation.|
|Rajashri Chakrabarti, |
“Vouchers, Public School Response, and the Role of Incentives: Evidence from Florida,”
Economic Inquiry 51, no. 1 (Jan. 2013),
|Examined the scores of students in F schools whose test scores fell just short of proficiency. These students’ scores improved significantly, suggesting that the schools focused on the failing students whose improvements could most quickly improve the school’s overall standing.|
|Marcus A. Winters and Jay P. Greene, |
“Public School Response to Special Education Vouchers:
The Impact of Florida’s McKay Scholarship Program on Disability Diagnosis and Student Achievement in Public Schools,”
Educational Evaluation and Policy Analysis 33, no. 2
(June 2011), pp. 138-58,
|Students in Florida public schools with average levels of McKay competition were 12 percent less likely to be diagnosed with a learning disability, and were likely to see reading and math score improvements of .01 standard deviations.|
|Matthew Carr, “The Impact of Ohio’s EdChoice on Traditional Public School Performance,” Cato Journal 31, no. 2 (Spring/Summer 2011), pp. 257-84, www.cato.org/pubs/journal|
|Carr’s study showed that the voucher threat created by the EdChoice program is associated with a statistically significant increase in the proficiency rates of public schools, for low-and-high performing students, on math and reading scores in both fourth and sixth grades.|
|Nicholas S. Mader, “School Choice, Competition, and Academic Quality: Essays on the Milwaukee Parental Choice Program,” PhD diss., University of Wisconsin – Madison, 2010, ProQuest (756023242).||Found small, positive competition effects for reading and math scores for students in grades three–five from 2000–01 through 2006–07.|
|Greene and Ryan H. Marsh, The Effect of Milwaukee’s Parental Choice Program on Student Achievement in Milwaukee Public Schools, SCDP Milwaukee Parental Choice Program Evaluation Reports 11, (Fayetteville: Univ. of Arkansas, Dept. of Education Reform, School Choice Demonstration Project, 2009), http://www.uark.edu/ua/der/|
|Looking at student-level data, the study concludes that public school scores improve as more private schools participate in the Milwaukee voucher program. Finds that for every 37 private schools that participate in the program, public school achievement is boosted by two NCE points (similar to percentage points). Speculates that the program has historically improved Milwaukee Public School performance by six points.|
|John Merrifield and Nathan L. Gray, An Evaluation of the CEO Horizon, 1998-2008, Edgewood Tuition Voucher Program (Washington, DC: Searle Freedom Trust; Austin, TX: CEO Foundation, 2009), https://www.heartland.org/sites/|
|Looked at year-to-year test score changes from 2003 to 2008 for grades three–11 and found that non-voucher schools in a district that had vouchers overall gained .09 standard deviations in math scores compared to schools in districts without vouchers.|
|Greg Forster, Promising Start: An Empirical Analysis of How EdChoice Vouchers Affect Ohio Public Schools, School Choice Issues in the State (Indianapolis: Friedman Foundation for Educational Choice, 2008), http://edchoice2016.wpengine.com/research/ |
|Examined year-to-year test score changes in schools where students were eligible for vouchers. Forster found positive effects from the EdChoice program in math scores for fourth and sixth grade students and reading scores for sixth grade students, and no visible effect in other grades. The positive effects ranged from three to five points in one year.|
|Chakrabarti, “Can Increasing Private School Participation and Monetary Loss in a Voucher Program Affect Public School Performance? Evidence from Milwaukee,” Journal of Public Economics 92, no. 5-6 (June 2008), pp. 1371-93, doi:10.1016/j.jpubeco.2007.06.009.||Combined with the analysis published in January 2008, the author found that the Milwaukee voucher program improved public schools. The author conducted multiple analyses using different methods for measuring public schools’ exposure to vouchers: Some are similar to Hoxby’s method (below) and others to Greene’s and Forster’s method (also below). In both studies, Chakrabarti found that increased exposure to vouchers improves academic gains in Milwaukee public schools.|
|Forster, Lost Opportunity: An Empirical Analysis of How Vouchers Affected Florida Public Schools, School Choice Issues in the State (Indianapolis: Friedman Foundation for Educational Choice, 2008), http://edchoice2016.wpengine.com/Research/|
|Examined the impact of the A+ program in every year from 2001 through 2006. Forster found that in 2001, before vouchers were widely available, “Voucher Threatened” schools made gains relative to all Florida schools equal to 13 points on Florida’s new developmental scale.” Over the next four years, as the percentage of families using vouchers decreased due to the red tape created by the state department of education, the positive voucher effect was not as large but remained substantial (“Voucher Eligible” schools gained from 20 to 27 developmental points each year).|
|Chakrabarti, Impact of Voucher Design on Public School Performance: Evidence from Florida and Milwaukee Voucher Programs, Staff Reports 315 (New York: Federal Reserve Bank of New York, 2008), http://www.newyorkfed.org/research|
|The author found that implementing a voucher program improves academic gains in Florida public schools. The author conducted multiple analyses using different methods for measuring public schools’ exposure to vouchers: Some are similar to Hoxby’s method (below) and others to Greene’s and Forster’s method (also below).|
|Chakrabarti, Impact of Voucher Design on Public School Performance: Evidence from Florida and Milwaukee Voucher Programs.||Combined with the analysis published in June 2008, the author found that the Milwaukee voucher program improved public schools. The author conducted multiple analyses using different methods for measuring public schools’ exposure to vouchers: some are similar to Hoxby’s method (below) and others to Greene’s and Forster’s method (also below). In both studies, Chakrabarti found that increased exposure to vouchers improves academic gains in Milwaukee public schools.|
|Martin Carnoy, Frank Adamson, Amita Chudgar, Thomas F. Luschei, and John F. Witte, Vouchers and Public School Performance: A Case Study of the Milwaukee Parental Choice Program (Washington, D.C.: Economic Policy Institute, 2007), http://www.epi.org/publication/book_vouchers.||This study used a modified form of the Hoxby/Chakrabarti method. The authors reported that their analysis “confirms the earlier results showing a large improvement in Milwaukee in the two years following the 1998 expansion of the voucher plan to religious schools.” Before 1998, religious schools were excluded from the Milwaukee program, so many fewer students participated. When religious schools were admitted to the program in 1998, participation increased dramatically, and so did public school performance.|
|John W. Diamond, Should Texas Adopt a School Choice Program? An Evaluation of the Horizon Scholarship Program in San Antonio (Austin: Texas Public Policy Foundation, 2007), http://bakerinstitute.org/media/files/Research/d9ce8f94/2007-02-RR03-education-diamond-3-2.pdf.||After the program was introduced, the percentage of Edgewood public school students passing state tests increased at a higher rate, and Edgewood’s graduate rate increased faster relative to the state as a whole.|
|Martin R. West and Paul E. Peterson. “The Efficacy of Choice Threats Within School Accountability Systems: Results From Legislatively Induced Experiments,” The Economic Journal 116, no. 510 (Mar. 2006), pp. C46-62, doi:10.1111/j.1468-0297.2006.01075.x.||Among schools that had not received the lowest possible rating under the state’s previous school evaluation system (which had no voucher component), receiving an F and thus being required to offer vouchers under the new accountability system produced an improvement in students’ test scores equal to about 4 percent of a standard deviation over one year.|
|Greene and Winters, An Evaluation of the Effects of D.C.’s Voucher Program on Public School Achievement and Racial Integration After One Year, Education Working Papers 10 (New York: Manhattan Institute, Center for Civic Innovation, 2006), http://www.manhattan-institute.org/|
|The D.C. program enrolls a relatively small percentage of students within the district, and public schools are “held harmless” to the effects of competition because additional money is used to “compensate” schools that lose students. Unsurprisingly, the authors found no visible effects upon the performance of public schools.|
|Figlio and Rouse, “Do Accountability and Voucher Threats Improve Low-Performing Schools?” Journal of Public Economics 90, no.1-2 (Jan. 2006), pp. 239-55, doi:10.1016/j.jpubeco.2005.08.005.||Examined the period over which vouchers were first being introduced in Florida. If a school received an F grade, its students made gains on the state test that were two points larger in reading and five points larger in math than those of other Florida schools over one year. Scores on the nationally-normed Stanford-9 test also improved. The authors would observe larger effects in subsequent studies, after vouchers had expanded further.|
|Greene and Winters, “Competition Passes the Test,” Education Next 4, no. 3 (Summer 2004), pp. 66-71, http://educationnext.org/competition-passes-the-test.||For both math and reading scores, on both the state test and the Stanford-9 test, so-called “Voucher Eligible” schools made improvements 15 points higher than other Florida public schools, while “Voucher Threatened” schools made improvements nine points higher.|
|Chakrabarti, “Closing the Gap,” in “Competition Passes the Test,” Education Next 4, no. 3 (Summer 2004), p. 70, http://educationnext.org/competition-passes-the-test.||Under the previous state accountability system—which did not include a voucher component for low-performing F schools—putting a school in the F category did not improve its performance relative to D schools in the next lowest performance category. However, three years after vouchers were implemented, the gap between F schools and D schools closed from almost 15 points to about five points.|
|Greene and Forster, Rising to the Challenge: The Effect of School Choice on Public Schools in Milwaukee and San Antonio, Civic Bulletins 27 (New York: Manhattan Institute, Center for Civic Innovation, 2002), http://www.manhattan-institute.org/html/cb_27.htm.||Found that greater exposure to vouchers had a positive effect on year-to-year changes in public school outcomes; the size of the effect was such that a school with all its students eligible for vouchers could be expected to outperform a school with only half its students eligible by 15 percentile points over four years.|
|Greene and Forster, Rising to the Challenge: The Effect of School Choice on Public Schools in Milwaukee and San Antonio.||Examined the impact of a large-scale privately funded voucher program targeted to the Edgewood school district in San Antonio, Texas. Controlling for demographics and local resources, they found that Edgewood’s year-to-year test score gains outperformed those of 85 percent of school districts in Texas. Given that Edgewood is a high-poverty (93 percent eligible for lunch programs) and high-minority (97 percent Hispanic) district, the study concludes that such a high statewide academic rank for Edgewood suggests that vouchers produced public school improvements.|
|Christopher W. Hammons, The Effects of Town Tuitioning in Vermont and Maine, School Choice Issues in Depth 1 (Indianapolis: Milton and Rose D. Friedman Foundation, 2002), http://edchoice2016.wpengine.com/Research/|
|Maine school districts for decades had the option of “tuitioning” their students—using public funds to pay for their students to attend private schools or nearby public schools —rather than building their own public schools. Hammons measured the relationship between a public school’s academic achievement and its distance from the nearest “tuitioning” town. Using regression analysis, he found a positive relationship. The relationship was strong enough that if a town one mile away from a school began tuitioning its students, the percentage of students at the school passing the state’s achievement test could be expected to go up by three percentage points.|
|Hammons, The Effects of Town Tuitioning in Vermont and Maine.||Vermont school districts for decades had the option of “tuitioning” their students—using public funds to pay for their students to attend private schools or nearby public schools —rather than building their own public schools. Hammons measured the relationship between a public school’s academic achievement and its distance from the nearest “tuitioning” town. Using regression analysis, he found a positive relationship. The relationship was strong enough that if a town one mile away from a school began tuitioning its students, the percentage of students at the school passing the state’s achievement test could be expected to go up by three percentage points.|
|Caroline M. Hoxby, “Rising Tide,” Education Next 1, no. 4 (Winter 2001), pp. 68-74, http://educationnext.org/rising-tide/.||Compared schools where at least 66 percent of the student population was eligible for vouchers to schools where fewer students were eligible for vouchers. Hoxby found that in a single year, schools in the “more exposed to vouchers” group made gains that were greater than those of other Milwaukee public schools by three percentile points in math, three points in language, five points in science, and three points in social studies.|
|Greene, An Evaluation of the Florida A-Plus Accountability and School Choice Program, Civic Reports (New York: Manhattan Institute, Center for Civic Innovation; Tallahasee: Florida State Univ.; Cambridge, MA: Harvard Univ., Program on Education Policy and Governance, 2001).||Schools that had received an F grade, whose students would be eligible for vouchers if the school received another F grade, made much larger year-to-year gains than schools that received a D (18 points in reading and 26 points in math for F schools versus 10 points in reading and 16 points in math for D schools).|
*We define a “study” as a unique set of one or more data analyses, published together, of a single school choice program. “Unique” means using data and analytic specifications not identical to those in previously reported studies. A “publication” is a means of reporting results to the public by report, paper, article, book, or book chapter. By this definition, all data analyses on a single school choice program that are reported in a single publication are taken together as one “study,” but analyses studying separate programs are taken as distinct studies even if they are published together.
“How Does School Choice Affect Public Schools?,” Friedman Foundation for Educational Choice, last modified Feb. 26, 2016, http://www.edchoice.org/school_choice_faqs/how-does-school-choice-affect-public-schools.
1. William G. Howell and Paul E. Peterson, The Education Gap: Vouchers and Urban Schools, ed. (2002; repr., Washington, D.C.: Brookings Institution, 2006), pp. 61-65.