Balancing School Choice and Equity

Chapter 6. Balancing choice and equity: Considerations for policy and practice

Abstract

This chapter summarises the results discussed in the report. It provides a broad perspective on the issue of school choice, and identifies the types of education policies that can help balance school choice and equity.

A note regarding Israel

The statistical data for Israel are supplied by and under the responsibility of the relevant Israeli authorities. The use of such data by the OECD is without prejudice to the status of the Golan Heights, East Jerusalem and Israeli settlements in the West Bank under the terms of international law.

A note regarding Lithuania

Lithuania became a member of the OECD on 5 July 2018. However, consistent with other publications based on PISA 2015 data, Lithuania is shown as a partner country and is not included in the OECD average.

This document, as well as any data and map included herein, are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area.

Many countries face the challenge of balancing aspirations for greater flexibility and parents’ freedom to choose their child’s school with the need to ensure equity in their school systems.

Education systems pursue very different approaches when it comes to linking school autonomy to school choice, and to reconciling choice with equity. For example, England and Shanghai both emphasise market mechanisms, but while public policy in England mainly operates on the demand side of markets, seeking to improve schooling by enhancing parents’ choice, in Shanghai, the main emphasis of public policy lies in creating a level playing field at the supply side: providing schools in the most disadvantaged areas with the best education resources. While Finland and Hong Kong both emphasise local autonomy, in Finland that autonomy is exercised within a strong public school system, while most schools in Hong Kong are managed by private entities with relatively loose steering mechanisms.

Some countries have strengthened choice and equity-related mechanisms at the same time. England, for example, has rapidly increased the number of academies, schools funded directly by the Department for Education and independent of local authority control. At the same time, England has established a pupil premium that provides schools with additional resources based on the socio-economic composition of their student body. Some countries have also made it possible for private schools to be integrated into the public education system as government-dependent schools or as independent schools that receive a certain amount of public funding.

Proponents of school choice defend the right of parents to send their child to the school of their preference – because of quality, pedagogical approaches, religious denomination, affordability or geographical location – regardless of legal restrictions or financial or geographical barriers. The idea is that, given students’ diverse needs and interests, a larger number of options in any one school system should lead to better value by reducing the cost of failure and mismatch. More options should stimulate competition and, in doing so, prompt schools to innovate, experiment with new pedagogies, become more efficient and improve the quality of the learning experience. Proponents argue that the increasing social and cultural diversity of modern societies calls for greater diversification in the education landscape, including allowing non-traditional providers and even commercial companies to enter the market.

Critics of school choice argue that, when presented with more options, students from advantaged backgrounds often choose to leave the public system, leading to greater social and cultural segregation in the school system. They are also concerned with over-reliance on theoretical models of rational, price-based economic competition as the basis for the allocation of resources.

At the macro level, such segregation can deprive children of opportunities to learn, play and communicate with children from different social, cultural and ethnic backgrounds; that, in turn, threatens social cohesion. To critics, vouchers and voucher-like systems divert public resources to private and sometimes commercial providers, thereby depriving public schools, which tend to serve large populations of disadvantaged students, of the resources they need to maintain the quality of the education they provide.

A closer look at the evidence shows that the arguments are not so clear-cut. For a start, the term “school choice” can be interpreted, and made manifest, in a variety of ways, depending on the local context (Musset, 2012_[1]). For instance, school-choice programmes can simply involve relaxing the rules on geographic assignment in the public sector or promote general choice amongst both public and private providers, as in Belgium and the Netherlands. In other cases, school-choice programmes are restricted to specific types of schools or student groups, such as those involving the use of income-dependent vouchers.1

Greater enrolment in private schools is often referred to as the privatisation of education, and is regarded as a move away from the notion of education as a public good. But that link is not so clear. In many countries where large parts of the school system operate under private legal statutes, such schools are seen as legally private but functionally public. For example, they can partly or completely follow the national curriculum and serve the public mission of education by providing accessible quality education (particularly where they are obliged to apply the same admissions and tuition standards as public schools). There are also many cases in which private schools provide access to education for underserved communities and have equity-related missions.

In addition, the consequences of school choice on education outcomes vary widely. This report provides an international perspective on issues related to school choice, especially how certain aspects of school-choice policies may be associated with sorting students into different schools. It discusses two proxies for school choice: the share of private schools in an education system and the use of residence-based policies in allocating students across schools. Over the past two decades, the number of privately funded secondary educational institutions has increased in only a few countries while residence-based enrolments have decreased in most countries. This suggests that school choice concerns all types of schools, as enrolment in public schools does not depend only on catchment areas.

One concern associated with increased school choice is the impact on the segregation of students by ability or socio-economic status. This report draws a comprehensive picture of school segregation, using a variety of indicators in order to account for the diversity of the sorting processes of students across schools. While in most countries the indices of academic segregation amongst schools are higher than those measuring socio-economic segregation, the opposite is observed in several countries, notably Latin American ones. These are also those countries where socio-economic differences in enrolment between private and public schools contributes, to a large extent, to the overall level of school segregation.

The organisation of the school system also affects the type of segregation. In some countries and economies, the academic segregation of students is mainly due to the isolation of low achievers “left behind” in some schools, while in other systems, it reflects a significant concentration of high achievers in “elite” schools. These specifics may have distinct consequences for student performance.

Provide the checks and balances that prevent choice from leading to more segregation

Whether greater competition between schools results in sorting students by ability or socio-economic status is a major question in the school-choice debate. Results discussed in this volume suggest that the impact of school-choice policies are ambiguous, and one should look carefully at the effects of the programmes on school diversity. Once the specificities of the school system are taken into account, estimates suggest that within a country/economy, relaxed residence-based admissions regulations is related to an increase in social segregation across schools. This does not mean that one should favour strict residence-based regulations; as discussed below, over the long term, such regulations can create additional residential segregation and thus reinforce school segregation. However, without some constraints in place, relaxing residence-based regulations may result in greater sorting of students by both ability and socio-economic status.

The design of catchment areas is an important tool that can be used to advance both equity and the efficient organisation of the school network. Given its sensitive nature, the definition and reform of catchment areas should involve local actors wherever possible. To avoid segregation, the equitable distribution of students should also be considered, such as by combining districts with different socio-demographic characteristics within a single catchment area (OECD, 2018_[2]).

Relaxing residence-based policies may benefit some disadvantaged students who can then attend better-quality schools that may not be located near their home. However, some students may lose out from such policies, namely low-achieving students who are “left behind” in low-quality schools. At the aggregate level, this may have a negative impact on equity but also, in some cases, on the general performance of the school system if low achievers are more harmed when they attend low-quality schools than high achievers benefit from being in better schools (for an illustration, see for instance (Brunello and De Paola, 2017_[3])). Panel estimates in this volume suggest that an increase in the isolation of high achievers from other students is associated with lower scores in PISA amongst socio-economically disadvantaged students, without any significant impact on advantaged students.

In practice, providing choice to parents without exacerbating segregation may be achieved by introducing specific criteria to the allocation of students across the set of local schools available. Different forms of “controlled choice” have been used to reduce high levels of student segregation, for example by reserving a given number or share of places in oversubscribed schools to students from different socio-demographic backgrounds to maintain a balanced distribution of students. The use of lottery systems to assign places in oversubscribed schools or formulae aimed to maintain a diverse student composition can also be considered (Musset, 2012_[1]). Centralised procedures to match students to schools usually rely on a set of criteria (Abdulkadiroğlu and Sönmez, 2003_[4]) that may include socio-economic status.2 Engaging school communities in defining these criteria and allowing for local variation can ensure that they are sensitive to local contexts; it can also significantly ease implementation of the criteria. Given their complexity, controlled-choice systems may require a certain degree of centralisation in order to minimise administrative costs and avoid problems, like multiple registrations (OECD, 2018_[2]).

In order to mitigate the potential negative effects of school choice and public funding of private schools, particularly segregation and social stratification, various governments have implemented compensatory financing mechanisms. For example the Flemish Community of Belgium (Box 6.1), Chile (Box 6.2) and the Netherlands have instituted weighted student-funding schemes, whereby funding follows the student on a per-student basis, and the amount provided depends on the socio-economic status and education needs of each student. These schemes target disadvantaged students and, in doing so, make these students more attractive to schools competing for enrolment.3

Box 6.1. Controlled-choice reforms in the Belgian Communities

Since passing the 2002 Decree on Equal Educational Opportunities, the Flemish Community of Belgium has implemented a series of reforms to its school-choice system. Following a two-year period between 2008 and 2010 that allowed for local experimentation to test different enrolment systems, a 2011 decree took stock of the lessons learnt and introduced a number of reforms to the controlled-choice system. First applied in 2012-13 (and subsequently adjusted) to admissions to all pre-primary, primary and secondary schools, the reform required oversubscribed schools to assign places to disadvantaged and non-disadvantaged students in proportion to the socio-economic composition of each school’s neighbourhood.

The reform also defined the criteria that schools could draw on to choose among students within each group when demand for places exceeded supply. Pre-primary and primary schools were allowed to consider the distance between the parents’ home or workplace and the school, the position of the school in the student’s ranking of preferences, or the results of a lottery. Secondary schools were required to operate on a first-come-first-served basis or to make decisions based on the position of the school in the student’s list of preferences. Many elements of the reforms’ implementation were decentralised and assigned to local negotiating platforms (locale overlegplatformen, LOPs). These platforms decided on matters such as the definition of neighbourhoods and the quotas for disadvantaged students, which helped reduce local resistance to the new rules. By 2013, there were 72 such platforms covering most of the territory of the Flemish Community, each of which ensured co-operation among schools, stakeholders and a defined local authority or region, bringing together representatives of the main education stakeholder groups in that area.

The French Community of Belgium reformed its parental choice system starting in 2007 and enrolment in the first year of secondary education has been subject to a 20% quota for disadvantaged students in every school since 2010-11. Enrolment at the pre-primary and primary levels remains largely unregulated. In contrast to that in the Flemish Community, the reform did not provide scope for adjustments at the local level or adaptations to a school’s neighbourhood characteristics, which may help to explain the relatively strong resistance by schools and other stakeholders. Identification of disadvantaged students was also imprecise and based on the average socio-economic status of their primary school, which, in turn, was defined on the basis of the residential area in which the school was located.

Sources: OECD (2018), Responsive School Systems: Connecting Facilities, Sectors and Programmes for Student Success, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264306707-en; Nusche et al. (2015), OECD Reviews of School Resources: Flemish Community of Belgium 2015, OECD Publishing, Paris, http://doi.org/10.1787/9789264247598-en; OECD (2015), OECD Economic Surveys: Belgium 2015, OECD Publishing, Paris, http://doi.org/10.1787/eco_surveysbel-2015-en.

Box 6.2. Regulating publicly funded private schools in Chile: The 2016 Inclusion Law

The Chilean school system is characterised by a large network of publicly funded private schools, enrolling 53% of its students in mainstream basic education (year 1 to year 8) and 51% at the upper secondary level (in 2014). Historically, Chile allowed publicly funded private schools to charge tuition fees, operate for profit and select students based on academic achievement, aptitude tests or parent interviews. This practice contributed to the country’s high level of socio-economic segregation as middle-class students increasingly left the public school system to enter subsidised private schools with admissions requirements that excluded large parts of the population. In order to address these concerns and facilitate the exercise of free school choice, a new law (Ley de Inclusion, Inclusion Law) was adopted in 2016 that imposes new eligibility criteria for public funding in order to restrict selective admissions, for-profit ownership and top-up fees among subsidised private schools.

The new regulations are enforced by the Education Superintendence. In order to remain eligible for public subsidies, private schools will need to phase out their tuition fees and other obligatory parental contributions (e.g. for school materials) over the coming years. They will also have to stop selecting students based on parent interviews or prior academic achievement. In order to facilitate this transition and compensate schools for the loss of revenues from parents’ “co-payment”, the law provides a number of additional subsidies (Aporte de Gratuidad). Notably, the law provides for a 20% increase in the Preferential School Subsidy (Subvención Escolar Preferencial, SEP), which assigns additional resources to schools serving the most vulnerable 40% of students. In addition, schools that abolish co-payments will be eligible to receive a grant amounting to 50% of the SEP for students from the third and fourth quintile of the income distribution. The estimated fiscal cost of these transition arrangements are subject to debate, ranging from the official estimate of USD 914 million per year to about USD 1 170 million (Santiago et al., 2017_[5]).

Sources: OECD (2017), The Funding of School Education: Connecting Resources and Learning, OECD Publishing, Paris, https://doi.org/10.1787/9789264276147-en; Santiago et al. (2017), OECD Reviews of School Resources: Chile 2017, OECD Publishing, Paris, https://doi.org/10.1787/9789264285637-en.

Where there is free school choice, ensure that all parents can exercise their right to choose the school of their preference

The international evidence suggests that schools that are selective in their admissions tend to attract students with greater ability and higher socio-economic status, regardless of the quality of the education they provide. Given that high-ability students can be less costly to educate and their presence can make a school more attractive to parents, schools that can control their intake wind up with a competitive advantage. Allowing private schools to select their students thus gives these schools an incentive to compete on the basis of exclusiveness rather than on their intrinsic quality. That, in turn, can undermine the positive effects of competition (Boeskens, 2016_[6]).

Evidence also shows that selective admissions can be a source of greater inequality and stratification within a school system. However, there are few studies that have investigated whether these effects vary, depending on the selection criteria – for example, interviews with parents compared to results of aptitude tests (Boeskens, 2016_[6]). It is also important to keep in mind that students are selected not only based on explicit admissions criteria but also because of parents’ self-selection, selective exclusion and more subtle barriers to entry. Policies that aim to reduce segregation in a school system should therefore also identify and address overly complex application procedures, expulsion practices, lack of information and other factors that prevent some students and parents from exercising their right to choose a school. To ensure that the advantages of school choice accrue to families across the socio-economic spectrum, the criteria used by oversubscribed schools to select their incoming students should be monitored and regulated to prevent “cream skimming” (OECD, 2018_[2]).

Evidence has repeatedly shown that it is usually the most highly educated and well-off parents who opt out of the assigned public schools. Low-income families may be unable to send their child to the best schools if those schools are privately funded and if the parents cannot afford tuition fees. They may prefer to enrol their child in the schools closest to their home, in order to avoid a long and costly commute. Public and school transport systems are another factor moderating the link between residential segregation and diversity in schools. Weakly developed or prohibitively costly transport networks can limit the extent to which lower-income families benefit from school choice. Effective school transport arrangements can ensure that all students are sufficiently mobile to benefit from the expansion of parental choice (OECD, 2018_[2]). Policies should thus provide targeted support to these families, not only by offering financial assistance but also by promoting mobility through adequate public transportation. Pioneered in the United States to overcome the legacy of racial segregation in the 1950s, busing schemes have since been used in multiple countries with the explicit aim to achieve a more socially balanced distribution of students across the school network and reduce the impact of residential segregation (Brunello and De Paola, 2017_[3]).

Mobility may also be reinforced through policies aimed at reducing urban segregation. School and urban segregation are two mutually reinforcing phenomena, as the social diversity in schools partly reflects that observed in nearby neighbourhoods, and will also determine where families choose to reside (if they have a choice).

School systems should ensure that all parents are provided with the relevant information. Efficient school choice implies that parents select the best school for their child – the school that can fulfil their education needs and improve their performance. However, a growing body of research suggests that parents have a clear preference for schools with better performance at test scores.4 Good test scores may be an indication that a school is doing a very good job of educating its students; they may also reflect the fact that the school teaches only the (already) best students. Absolute performance, as measured by test scores, does not mean that a school has a high “value-added” component. Attending a school with a large proportion of high achievers does not always result in individual improvements in performance (Abdulkadiroğlu et al., 2017_[7]; Dobbie and Fryer, 2014_[8]; Lucas and Mbiti, 2014_[9]). Parents would benefit from a measure of the actual “value-added” of schools, meaning whether those schools succeed in improving the performance of all of their students.5 Evidence shows that the quality of teachers is the most important element (Hoekstra, Mouganie and Wang, 2018_[10]; Pop-Eleches and Urquiola, 2013_[11]), suggesting that another way of avoiding additional stratification between schools is to allocate the best teachers across all schools, including disadvantaged ones.

Well-crafted school-choice policies can help school systems deliver education tailored to a diverse student population, while limiting the risk of social segregation. When market mechanisms are introduced or expanded in education systems, the role of public policy needs to shift from overseeing the quality and efficiency of public schools to ensuring that oversight and governance arrangements are in place to guarantee that every child benefits from accessible, high-quality education.

The conditions under which private schools are eligible for public subsidies influence the ways in which school-choice programmes affect the accessibility, quality and equity of the school system. Risk to equity can be mitigated if all publicly funded providers are required to adhere to the same regulations regarding tuition and admissions policies, and compliance with these regulations is monitored. Adequate accountability and transparency requirements are also important to ensure that subsidised private schools serve the public interest in providing high-quality education, and to provide parents with the information they need to evaluate different schools’ processes and outcomes (OECD, 2017_[12]).

References

[13] Abdulkadirog˘lu, A., J. Angrist and P. Pathak (2014), “The Elite Illusion: Achievement Effects at Boston and New York Exam Schools”, Econometrica, Vol. 82/1, pp. 137-196, http://dx.doi.org/10.3982/ECTA10266.

[7] Abdulkadirog˘lu, A. et al. (2017), “Do Parents Value School Effectiveness?”, NBER Working Paper, Vol. 23912, http://dx.doi.org/10.3386/w23912 (accessed on 19 July 2018).

[4] Abdulkadirog˘lu, A. and T. Sönmez (2003), “School Choice: A Mechanism Design Approach”, American Economic Review, Vol. 93/3, pp. 729-747, http://dx.doi.org/10.1257/000282803322157061.

[6] Boeskens, L. (2016), “Regulating Publicly Funded Private Schools: A Literature Review on Equity and Effectiveness”, OECD Education Working Papers, No. 147, OECD Publishing, Paris, https://dx.doi.org/10.1787/5jln6jcg80r4-en.

[3] Brunello, G. and M. De Paola (2017), School segregation of immigrants and its effects on educational outcomes in Europe - Publications Office of the EU, European Commission, http://dx.doi.org/10.2766/045520.

[17] Davezies, L. and M. Garrouste (2018), “More Harm than Good? Sorting Effects in a Compensatory Education Program”, Journal of Human Resources, pp. 0416-7839R1, http://dx.doi.org/10.3368/jhr.55.1.0416-7839R1.

[8] Dobbie, W. and R. Fryer (2014), “The Impact of Attending a School with High-Achieving Peers: Evidence from the New York City Exam Schools”, American Economic Journal: Applied Economics, Vol. 6/3, pp. 58-75, http://dx.doi.org/10.1257/app.6.3.58.

[14] Fack, G. and J. Grenet (2014), Can we increase the social and academic mix in the education system? Theimpact of allocation procedures on Lycée students in the Paris region, IPP, Paris, http://www.parisschoolofeconomics.eu (accessed on 4 January 2019).

[10] Hoekstra, M., P. Mouganie and Y. Wang (2018), “Peer Quality and the Academic Benefits to Attending Better Schools”, Journal of Labor Economics, Vol. 36/4, pp. 841-884, http://dx.doi.org/10.1086/697465.

[9] Lucas, A. and I. Mbiti (2014), “Effects of School Quality on Student Achievement: Discontinuity Evidence from Kenya”, American Economic Journal: Applied Economics, Vol. 6/3, pp. 234-263, http://dx.doi.org/10.1257/app.6.3.234.

[1] Musset, P. (2012), “School Choice and Equity: Current Policies in OECD Countries and a Literature Review”, OECD Education Working Papers, No. 66, OECD Publishing, Paris, https://dx.doi.org/10.1787/5k9fq23507vc-en.

[2] OECD (2018), Responsive School Systems: Connecting Facilities, Sectors and Programmes for Student Success, OECD Reviews of School Resources, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264306707-en.

[12] OECD (2017), The Funding of School Education: Connecting Resources and Learning, OECD Reviews of School Resources, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264276147-en.

[15] OECD (2008), Measuring Improvements in Learning Outcomes: Best Practices to Assess the Value-Added of Schools, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264050259-en.

[11] Pop-Eleches, C. and M. Urquiola (2013), “Going to a Better School: Effects and Behavioral Responses”, American Economic Review, Vol. 103/4, pp. 1289-1324, http://dx.doi.org/10.1257/aer.103.4.1289.

[16] Raudenbush, S. and J. Willms (1995), “The Estimation of School Effects”, Journal of Educational and Behavioral Statistics, Vol. 20/4, pp. 307-335, http://dx.doi.org/10.3102/10769986020004307.

[5] Santiago, P. et al. (2017), OECD Reviews of School Resources: Chile 2017, OECD Reviews of School Resources, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264285637-en.

Notes

← 1. For instance, in the voucher programme implemented in the district of Milwaukee (state of Wisconsin in the United States), eligibility is restricted to students from low-income families.

← 2. For instance, in the city of Paris (France), the allocation of students to high schools relies on a matching procedure that has, for more than 15 years, integrated criteria favouring students from low-income families. This has resulted in large reductions in academic and social segregation across Parisian high schools (Fack and Grenet, 2014_[14]).

← 3. Specific area-based support schemes, such as the “zones of educational priority” found in France and Greece, are observed in school systems with large between-school variations in performance and a concentration of low-performing schools in certain locations. In the case of France, this scheme has been shown to do more harm than good (Davezies and Garrouste, 2018_[17]): the most advantaged families tend to disproportionally opt out from schools labelled as disadvantaged.

← 4. This is observed, for instance, in China (Hoekstra, Mouganie and Wang, 2018_[10]), New York City and Boston (Abdulkadirog˘lu, Angrist and Pathak, 2014_[13]; Dobbie and Fryer, 2014_[8]), Romania (Pop-Eleches and Urquiola, 2013_[11]) and Paris (Fack and Grenet, 2014_[14]).

← 5. See (OECD, 2008_[15]) for a practical guide on how school effects were computed. The evaluation of school or teacher effects per se may be complex (as they are often intertwined with the impact of school composition on achievement); but under certain conditions these indicators may be relevant for parents choosing a school for their child (Raudenbush and Willms, 1995_[16]).