2020. május 25., hétfő , Orbán

1055 Bp., Szalay u. 10–14.

Tel.: (+36-1) 235-7200

Fax: (+36-1) 235-7202

magyar english
Elfelejtett jelszó

Arany János Programok  IKT  OFI  OKJ  SDT  Vizsgacentrum  biztonságos iskola  egészségtudatos iskola  erőszakmentes  kiadvány  konferencia  kétszintű érettségi  letölthető  oktatás  próbaérettségi  pályázat  rendezvény  ÚPSZ  Új Pedagógiai Szemle  érettségi 

Intézeti folyóiratok

Köznevelés
Új Pedagógiai Szemle
Educatio
Könyv és nevelés
Kattintson ide a rendeléshez!
Tudástár >> Hátrányos helyzet, második esély >> Esély az együttnevelésre – Chance for integration

Financing the education of disadvantaged students

2009. június 17.

Zoltán Hermann – Dániel Horn1

Financing the education of disadvantaged students

1. Introduction

The educational policy objective of approximating the school performance of disadvantaged students towards the average is most often sought by using special educational methods and/or increasing school resources. However, these two instrument are not independent of each other: the development and use of new educational methods is usually costly and as such the effective education of disadvantages students involves higher per capita cost (for the estimates of the additional costs required see Downes–Pogue, 1994, Duncombe–Yinger, 1999). At the same time, the increase of resources (smaller class-size, reduction of the number of teachers’ lessons, etc.) promises very limited results without the change of educational methods (see for example Murnane–Levy, 1996). It is a necessary but not sufficient condition for more successful education for these students.

In a decentralized educational system such as the one in Hungary, financial incentives have an extremely important role as central educational policy can have limited direct influence on the actual work of schools , e.g. it cannot prescribe the educational methods for disadvantaged students. So besides creating possibilities for education that compensates disadvantages, financing instruments should encourage schools, local governments (as school owners) and parents to make decisions towards this end.

This paper first summarizes the educational problems of the disadvantaged and Roma youth most frequently discussed in the literature in Hungary today. Then it examines the advantages and disadvantages of the possible financing methods: to what extent the particular methods facilitate or hinder the solution of the problems discussed. Finally, it compares the financing mechanisms, together with the brief discussion of possible additional policy instruments.

The analysis is primarily concerned with primary education and only reflects on questions of kindergarten and secondary education occasionally.

Our analysis focuses on the possible financing methods of educational policy programs aimed at managing the learning difficulties stemming from the disadvantaged situation of families. We believe that equal opportunity programs should target socially disadvantaged students living in poor families, brought up by poorly educated parents. Hence, by disadvantaged student we do not mean (mentally or physically) disabled or Roma students, or those with general learning problems. Disabled students pose a particular challenge to the educational system that is not discussed here.

Supporting the education of disadvantaged students is justified by the fact that according to empirical researches these students perform considerably worse at school than those with more fortunate family backgrounds (for Hungary see Bánfi, 1999 and Vári, 2003). Such education programs should target disadvantaged students rather than those with learning problems for two reasons. First, the coherent assessment and recognition of learning problems is not a straightforward exercise and as such any grant system based on this parameter could lead to mistakes and abuses. Second, the performance of disadvantaged students is crucial factor in the reproduction of social inequalities and social mobility.

As far as Roma students are concerned, the prime cause for their education failure is just as well their social environment,2 nothing justifies the exclusion of non-Roma, ‘only’ disadvantaged students and their schools from the program (Kertesi 2002, Liskó 2002a). Though the two groups partly overlap, they should be treated separately (Kertesi, 2002). On the other hand, socially disadvantaged family background is a much more unambiguous aspect in allocating allowances than ethnicity. The question of who qualifies as disadvantaged in a society is mainly a political (maybe sociological) issue, while Roma identity depends on the self-identification of the individual.

All this does not mean that we would challenge the raison d’ętre of an educational policy program targeting Roma students. However, we believe that programs aimed at preserving Roma culture on one hand and at more effective education of disadvantaged students on the other should be treated separately. Though the teaching of minority culture at schools is beyond the scope of this study, we think that such programs also should be financed by the central government.

A major part of the studies referred to below focus on Roma and not on disadvantaged students. As the conclusions point very often to the socially disadvantaged environment of these students, they will be cited in this sense.

2. Educational problems due to disadvantaged family background

2.1 Kindergarten

Essential conditions for a successful school career are the acceptance of school atmosphere, proper language skills and the existence of the social competencies indispensable for learning. Basic socializing instruments such as flush toilet, books, television or other electronic devices are not available for the disadvantaged students at home. They have language difficulties and insufficient material culture. The institution for overcoming all these disadvantages could be the kindergarten. Experiences show, however, that disadvantaged children make the least use of it. The reasons for this might be:

  • There is no sufficient supply of kindergartens therefore they have to refuse several applicants. In the majority of disadvantaged families at least one of the parents is unemployed or inactive, so the “baby-sitting” function of the kindergarten is not essential. What is more, one of the most frequent arguments for refusing a child is that a parent is unemployed and can take care of the child at home (Vágó, 2002). As a consequence, disadvantaged children are underrepresented in kindergartens.
  • The costs of kindergarten are considerable as well (travel fare, meals, contributions to other costs of the kindergarten…). Costs may be an exceptionally strong restraint in settlements without a kindergarten.
  • Kindergarten supply differs considerably across settlements. Small settlements – where the proportion of disadvantaged families is high – have more limited supply or no kindergarten at all partly due to the settlement size and partly due to the bad financial situation of the local government. Among settlements under 10 000 inhabitants those with high unemployment have the least kindergarten places available, regardless to the settlement size (table 1).
  • Kindergarten does not provide appropriate conditions for Roma children (attention, tolerance, pedagogical competence, etc.) therefore they do not like going to kindergarten (Pik, 1998). There may be cultural differences between disadvantaged especially the Roma - and not disadvantaged children. Kindergarten is the first bureaucratic institution that children have to face.
Table 1. Number of places available per 100 kindergarten aged children according to settlement size and unemployment rate, 2001
  Registered unempoyed relative to 18–59 years old population Total
Low* Medium** High***
Source: KSH T-STAR database
* The rate of unemployed below the 33th percentile of the group of settlements.
** The rate of unemployed between the 33th and 66th percentile of the group of settlements.
*** The rate of unemployed above the 33th percentile of the group of settlements.
Under 1000 84,5 78,5 66,8 76,5
1 000–3 000 116,0 120,5 106,9 114,4
3 000–5 000 114,7 113,4 104,7 110,8
5 000–10 000 112,9 115,9 106,8 111,8
10 000–50 000 117,5 124,3 115,2 119,1
Over 50 000 123,9 128,7 122,1 124,9
Total 99,1 97,7 85,8 94,1

2.2. Segregation

School segregation may be the most important and the most evident problem in the education of disadvantaged students, especially in the case of Roma students. According to 1992 data (the last time ethnicity questions were included in educational statistics) 44% of Roma students and only 6.3% of non-Roma students learned in schools where the proportion of Roma students was higher than 25% or their number was 100 or more (Havas-Kemény-Liskó, 2002). According to a survey carried out in schools educating Roma students, 4% of the 6th grade Roma children surveyed attended a ‘pure’ Roma school, 29% attended a school with Roma majority and only 3% of them attended a school where the proportion of Roma students was small (Liskó, 2002).

At the same time the segregation of disadvantaged or Roma students is a consequence of several processes and not exclusively the effect of conscious decisions of individual school directors and school owners.

2.2.1. Segregation within school

Schools with several parallel classes in a grade have an alleged or real interests, supported by non-disadvantaged parents to organize separate classes for students who progress more quickly, thus providing opportunities for acquiring more learning material and better entrance exam results. Slower students may hinder others in learning more or learning more thoroughly. The ambition of non-disadvantaged parents is clearly to make sure that their children go to schools with students similar to theirs. At the same time it is beneficial for the school as well to organize a ”display” class which helps to attract the non-problematic children of the settlement to the school. An obvious way for that is organizing classes with advanced programs. While doing so they will obviously organize a relatively weaker class as well, that consist – as experience shows – of mainly the most disadvantaged, Roma students. In some cases this may be the only rational way of avoiding the situation when the school becomes a totally “Roma school” and to make sure that non-disadvantaged parents send their children to this school and not to another one (Havas, 2001).

2.2.2. Parental school-choice decisions

Purely disadvantaged schools and settlements develop according to similar processes. If a school tends to satisfy parents less, the easiest choice for better-off parents is to find another school either in the same or a nearby settlement. If many choose the exit option, it will automatically result in a school full of disadvantaged students (whose parents cannot afford the costs of commuting) or students who were not admitted to other schools for other reasons (ethnicity or socialization problems).

2.2.3. Spatial segregation

School segregation is a natural concomitant of the spatial segregation of disadvantaged or Roma families, as well. If employment opportunities are getting scarce or infrastructure is getting insufficient in a settlement or in its region, families that have other opportunities can easily move from the given region thus deteriorating the average standard of living there. This process changes the social composition of schools as well and leads to spatially segregated institutions (Havas–Kemény–Liskó, 2002).

2.2.4. Special schools3

Special schools provide education not only for disabled children, but also for students who are not at the maturity level appropriate for their age group. In these schools they can acquire a more individualized and easier learning material more effectively than in traditional ones. Disadvantaged students with Roma origin are heavily over-represented in these schools (Babusik, 2001).

The problem with special classes – apart from segregation – is that they give a lower qualification, thus excluding these students from further educational opportunities. At the same time research data also indicate that school owners allocate considerably less to special classes thus reducing the effectiveness of special schools (Havas–Kemény–Liskó, 2002).

2.3. Pedagogical problems

Disadvantaged students and those belonging to the Roma minority require special attention at school. Often they cannot meet the requirements of the school, concerning preparation, equipment or any other field. However, schools often have neither the financial/infrastructural background for meeting special requirements nor teachers with appropriate pedagogical qualifications. In schools where the proportion of Roma students is above 75%, the proportion of unqualified teachers is 30,8% on average whereas in schools with less than 25% Roma students, it is17,4% (Liskó, 2001, Table 1).

2.4. Educational career following the primary school

As compared to the early nineties, the proportion of Roma students continuing their studies in secondary schools rose considerably. However, even today there is a significant difference between Roma and non-Roma students in this respect. While almost all non-Roma students continue their studies, nearly 15% of Roma students does not enroll in secondary level education (Havas–Kemény–Liskó, 2002). Moreover, while non-Roma students continue their studies in secondary schools preparing for a school-leaving examination, Roma students rather choose vocational training. Children of parents with lower qualification tend to learn in secondary schools preparing for a school-leaving exam to a lesser extent than children of more qualified parents, so the problem is self-preserving.

Several surveys show that Roma students have considerably less chance for completing secondary education than their non-Roma peers. While in institutions where the proportion of Roma students is more than 20%, the drop-out rate is 17,4%, in schools with a proportion of Roma students less than 10%, the drop-out rate in 9th grade is 12,6%. (Liskó, 2002b)

3. Financing methods

In the following section three possible financing methods of programs aimed at reducing learning disadvantages will be examined. The analysis does not cover all possible methods4 and does not include detailed discussions of possible sub-varieties of the analyzed methods. First the effects of supplementary per student lump sum grants will be analyzed, which will be followed by outlining a possible project financing scheme, and finally the advantages and disadvantages of two market-type solutions, the voucher-system and the so-called quasi-market system will be discussed.

We believe that both of the first two financing schemes can easily be incorporated into the established framework of the Hungarian system of local governments, so these are obvious, viable options in the short run from political aspects as well. On one hand, the market-type solutions are regarded as possible options in the long run and as a good reference point for the evaluation of the other two. Market-type solutions cannot be neglected if for no other reasons than because of previous policy suggestions (concerning the education of disadvantaged students: Kertesi, 2002, and with respect financing system of education: Semjén, 1995, 1997). Following the analysis of the three possible financing methods, the next chapter discusses additional policy instruments, which may further improve the effect of programs aiming at a more effective education of disadvantaged students.

The probable effects of each method are strongly influenced by the detailed regulation of grant programs. On the one hand much depends on the criteria according to which grants are available for local governments, schools or parents, i.e. how the government defines the target group of students. Beyond the direct effect this can also influence the degree local governments and schools can abuse this system by classifying students in an arbitrary way in order to increase financial support.

On the other hand, in the case of the supplementary per student grant and project financing the government requirements towards schools on local education programs is of decisive importance. (In market-type systems improved effectiveness is assumed to be warranted by competition between schools and not by governmental regulation and incentives.) Requirements related to local educational programs may prescribe the minimum amount of school inputs per student (e.g. minimum per capita expenditures, maximum class size and teaching hours, requirements concerning the qualification and in-service training of teachers, employing staff assisting in educational work), may regulate the process of education (e.g. applying requirements concerning the pedagogical work of schools, using given developing methods, organizing integrated classes) or may set outcome objectives (e.g. average academic outcome measured by standardized tests or the improvement of these outcomes between given grades, percentage of those continuing studies, rate of school year repetition and rate of absence). Naturally, various combinations of requirements are possible, and the probable effects of each variation can be very different. The perspective effects of the requirements obviously depend on what incentives or sanctions the regulation attaches to them. One extreme is when school owners are practically free to deviate from the objectives (such is the present regulation concerning average class sizes). A stronger regulation is when the given requirements are the conditions for being entitled for support or when additional grants are connected to the set outcome objectives.

None of the non-market type grants can be expected to have a considerable incentive effect on the success of the program if the requirements are weak or totally absent. It is important to stress that the requirements concerning local educational programs have to be taken into account when evaluating the financing methods: one of the financing methods with weak requirements may seem to have more advantages, while in the context of stronger quality control the other may seem to be more advantageous. In other words, the financing method appropriate for the given educational policy objectives and preferences can only be selected if a possible variation of the requirements related to the local educational programs has been selected.

Finally, the success of a program aimed at reducing learning disadvantages depends obviously on the extent of the program: which institutions participate or may participate in the program, which levels of education the program covers, how much the total amount of grants is.

3.1. Supplementary per student grant financing

Current financing methods belong to the per student grant-type of financing. Until the autumn of 2003, local governments received a supplementary per capita grant for disadvantaged students participating in education in special learning groups and after that date they receive this grant for those participating in integrated education. (Law No. LXII. of 2002 on the Budget of the Republic of Hungary)

The obvious advantage of this financing method is that it is simple and basically does not involve huge administrative costs. Additionally, there are no concerns regarding the distribution of this grant, while in the project system the practice of considering individual demands can always be questioned.

This financial method assigns the greatest role, autonomy and responsibility to school owner local governments. While in the project financing system the agreement between the schools and the central government and the rules of this agreement, in a market type system the competition between schools and the individual decisions of parents are the mechanisms that motivate schools to reduce disadvantages. In the supplementary per student grant system it is the local government autonomy and the local political accountability that may ensure the successful realization of the program. Choice between various types of incentives is partly a question of value judgement. It is evident that in the Hungarian political context local government autonomy is an important value. At the same time, it is worth mentioning that in programs targeting a minority of the local electorate5 – independent of whether the criterion for minority is learning difficulties, ethnicity, or the social situation of families – it is questionable how much performance can be guaranteed by accountability, which is enforced through political elections.

One of the basic problems with the per student grant financing is that the amount of grants necessary for achieving a given objective cannot be easily defined (Varga, 1998), and the financing system cannot manage individual cost differences separately. It is not easy to define – even ex post, based on expenditure data – the additional costs of improving the academic outcomes of disadvantaged students, as this type of reliable analysis requires individual-level longitudinal data related to the outcomes – measured characteristically by standardized tests (Duncombe–Yinger, 1999). It is obviously even more difficult to give a preliminary analysis of what extent of cost increase makes it possible for schools to clearly improve the outcomes of students participating in the program. However, this problem cannot be avoided in the other two financing methods, either.

Individual cost differences are related partly to school size and the number of disadvantaged students and partly to their ratio in the school. Economies of scale characterizing the operation of schools are probably present in the programs aimed at reducing learning disadvantages, as well. In smaller schools managing learning difficulties is expected to be more expensive per capita. At the same time the relationship between average costs and the ratio of disadvantaged students within the school is not trivial, costs may increase with this ratio, e.g. because the positive external effect of the presence of students with better school achievement can be felt to a lesser extent. Per student grants are not suitable for taking such cost differences into account. Project financing provides greater possibilities for this, in principle, but it is uncertain to what extent such aspects can be enforced in the practice of discretionary decisions approving individual school projects.

One of the major problems is that the supplementary per student grant financing transferred to local governments does not necessarily mean that resources of the targeted school increase as expected, since the lump sum grants are not earmarked. The microeconomic analysis of the impact of lump sum grants shows – besides standard assumptions – that decision-makers spend the additional grants on various services independent of the original purpose6 (see e.g. Friedman, 1984). Lump sum grants do not have an effect on the relative prices of services; they only have an effect on the budget constraints. So a part of the unconditional supplementary grants is spent on education, another part is spent on other services, as if the additional grant had been received for supporting local mass sport or the social provision of the elderly. The income effect can be ‘redirected’ if the central government attaches certain conditions to grants:

a) either by earmarking (related pitfalls will be discussed in the analysis of project financing),

b) or by setting certain requirements or objectives, e.g. school achievement of disadvantaged students should be improved to some extent.

The first solution could hardly be incorporated into the financing system of the Hungarian local governments as the base expenditure level cannot be defined. The second solution could be that grants are linked to requirements related to school inputs, to the process of education or to the outcomes of education. In essence such an input requirement is – though rather weak as compared to the objectives – that the condition for receiving per student grant for special learning group education (hereafter special learning group grant) is to have special learning groups. Similarly, for receiving the ‘integration grant’ the condition is that disadvantaged students learn in integrated classes.

We should note, however, a major difference between the previous and present per student grant financing. In principle, providing education in special learning groups requires additional expenditures in schools (due to smaller number of students studying in one group and to special learning groups that increase the overall number of lessons per week), so school expenditures will presumably have to increase where such a program is introduced.7 Education organized in integrated classes, however, does not necessarily involve additional expenditures for schools (if pedagogical methods, class sizes, etc. are the same). Moreover, if it is true that the presence of students with good school achievement have a positive external effect on those with learning difficulties, in principle the costs required for achieving results of a given level could even decrease with the introduction of integrated education.8 That is, the per student grant attached to integrated education could be interpreted as a pure reward: if and when local governments intend to increase the incomes only and do not consider the objectives of central educational policy programs (the reduction of school disadvantages) as important priorities, the grant does not cover the expenditures of some additional service but only rewards the introduction of integrated education. As opposed to this either a grant aimed directly at schools or strengthening outcome requirements may result in an increase of school expenditures.

Supposedly it is true for only some local governments that they are interested mainly in increasing the grant revenues, others have certainly adopted the objective of reducing school disadvantages. In their case the integration grant provides some kind of financial opportunity to improve the education of disadvantaged students. It is also worth mentioning that the ‘integration grant’ presents the realization of integrated education as an objective for local decision-makers, while in fact this is only a possible instrument for reducing learning disadvantages. Thus, due to the lack of outcome criteria this financing method does not imply a strong incentive for applying appropriate pedagogical methods in smaller classes in order to reduce learning difficulties.

The advantage of a supplementary per student grant encouraging integrated education is that it rewards the local governments for stopping the practice of school segregation with additional grants. The ‘special learning group grant’ supported a form of education that could both serve the reduction of the problems of disadvantaged students and the segregation and exclusion of the same students from a better education. It is worth mentioning that supporting education in special learning groups could encourage education of disadvantaged students in separate classes, schools even without the intention of negative discrimination: it is easier to fit special lessons in the timetable if the students that take part in them learn other subjects in the same group as well, teaching these special groups may be cheaper and easier if these groups are not scattered or small-sized but such students attend the same school.

Could the supporting of integrated education reduces the school segregation of disadvantaged students? If and when school segregation is directly a function of local government and school decisions, probably yes. In general, however, it does not necessarily lead to a major reduction as school segregation is not exclusively the result of the decisions of local governments and schools. Most certainly spatial segregation and parents’ choice of school are very important factors. In the absence of such research data it cannot be estimated to what extent local government and school decisions or other factors are responsible for school segregation but we think that the influence of the latter is substantial.9

The greatest danger of supplementary per student grants attached to integrated education is that as a result of spatial segregation a part of disadvantaged schools and local governments owning them get excluded from the grant: in settlements where a great majority of students are disadvantaged there is no possibility for organizing integrated classes. This means that it is exactly those who need support most that may not receive it. Those with lower income may have been forced to move into these villages exactly for this reason and – being far from the labor markets of towns – will have more chances to become unemployed than those living in towns. Moreover, the higher the proportion of the disadvantaged in a settlement, the less the incomes of the local government and the greater the needs are for local government services (e.g. in local social transfers) and the local governments with worse budget situation can spend less on education independent of additional grants as well (Varga, 2000).

This financing method does not offer a solution for the problem of segregation related to the school choice parents. Schools may become excluded from grants attached to integrated education in this case as well.

While the effect of spatial segregation – e.g. in settlements where the composition of those living there makes it impossible for a school to introduce integrated education – seems to be manageable relatively easily (the local government could be entitled to exceptional grant), the problems connected to decisions concerning school choice cannot be managed so simply. It would be difficult to decide whether in a given case schools could raise the proportion of non-disadvantaged students to the level required for the grant.

It has to be stressed that school segregation connected to school choice is a problem that is difficult to solve independent of the financing method. Even if the assumption of net positive external effects of mixing high- and low-achievers holds, i.e. integration is beneficial overall as opposed to segregation (Henderson–Mieszkowski–Sauvegeau, 1978, Schwab–Oates, 1991), from an individual point of view high-achievers face an incentive to choose a school with a similar student composition. This, naturally, does not mean that free school choice dooms the reduction of school segregation to failure. But to achieve this objective, the schools educating disadvantaged students in higher proportion than average should provide educational services of a much higher quality so that it can counterbalance the negative external effect stemming from student composition. This is not inconceivable: the pedagogical methods of schools and smaller class sizes may make the education of all students more efficient. But to achieve this significantly higher expenditures are needed in schools educating disadvantaged students in high proportions than in other schools, though higher expenditures are only necessary but not sufficient conditions for more effective education (Murnane–Levy, 1996). It is difficult to make the schools attractive for high achievers because there is a self-generating process: the higher the proportion of disadvantaged students is, the fewer high social status children will enroll (or the more of them leave the school) and as a consequence the proportion of disadvantaged students will increase further. This process is hardly reversible beyond a certain point.

Above we argued that supplementary per student grant financing of local governments provides a weak incentive for reducing learning disadvantages if accompanied by relatively weak requirements concerning local educational programs. What requirements could central educational policy attach to such financing? We think that requirements related to school inputs are the most compatible with lump sum grants such as smaller than average class-size, participation of teachers in certain training programs, or extra wage for teachers . Such a regulation would be sensible assuming that schools educating disadvantaged students are less attractive for teachers as well: teaching in a school educating students with good school achievement means more peaceful working conditions and higher professional prestige. It is extremely important that the targeted schools are capable of employing well-prepared teachers who can work effectively.10

As opposed to this, per student grant could hardly be made compatible with the requirements related to pedagogical methods, curricula or the improvement of the effectiveness of education. Such stricter criteria seem easier to be enforced in a project financing system. It is worth mentioning, however, that there are possibilities for directly encouraging the improvement of effectiveness besides supplementary per student grants, as well. Financing methods – introduced in some states of the United States – reward the most effective schools ex post (see e.g. Clotfelter–Ladd, 1996, Elmore–Abelmann–Fuhrman, 1996). Student performances are assessed annually by standardized tests, and the effectiveness of schools is defined according to the improvement of the test results, filtering out the effects of social background and prior achievement. The schools being at the top of the rankings (a fourth or a third of schools) receive additional grants that can be used for rewarding the teachers. This solution is compatible with per student grants and can easily be adapted to schools educating disadvantaged students. At the same time it requires the operation of extensive central evaluation system. The introduction of such a financial incentive for schools educating disadvantaged students exclusively would very likely result in major tensions. Therefore a ‘rewarding’ system pertaining to all schools – but treating schools entitled for additional financing separately – seems to be more viable. It has to be stressed that such an ex post financing method may influence the work of teachers but does not provide financial coverage for the reform of the educational program or for extending the staff. So the changes by which the schools educating disadvantaged students could make an attractive choice for all students cannot be financed in this way.

3.2. Project financing

As opposed to the supplementary per student grant, project financing offers a viable alternative even on the short run. Although a system in which school owner local governments could apply for additional financing is also conceivable we think that the introduction of a project system created for schools is more beneficial. Project-based financing is not absent from the practice of Hungarian public education (e.g. Comenius program).

The basis of this financing method is that individual schools receive additional governmental financing by applying for operating programs aimed at more effective education of disadvantaged students (or occasionally for developing or improving such programs). This solution assumes long-term programs designed for several years and predictable financing.

Perhaps the greatest advantage of this system is that it makes possible both the enforcement of the requirements related to school inputs, the process of education and the outcomes , and accommodating the granted sum to individual expenditure needs simultaneously.

Input requirements may be included both in the tenders and can serve as criteria for their evaluation: e.g. the instruments by which they intend to improve the effectiveness of education (pedagogical methods, class sizes, etc.) and the necessary budget plan. At the same time – as opposed to the supplementary per student grants – it is not necessary to enforce totally uniform input requirements. Schools can decide how to use the additional resources, adjusting to the local needs. In the tenders schools should obviously give reasons for their choices. Interim evaluations could be executed during the project period. Exceptionally effective schools could get further financing.

As the project system is aimed directly at schools, the chances that the money is used for its intended purpose – i.e. it can really increase school expenditures – are higher than with local government grants. However, it has to be taken account that local governments may be interested in ‘draining’ a part of the additional financing of schools (the additional financing modifies the aggregate budget constraint of local government decision-makers, who would increase school expenditures only to the extent of the income effect – see above). Local governments can achieve this by reducing the locally approved budget of schools, thus forcing schools to spend a part of the additional financing on basic services of the school. This could be avoided by requiring the endorsement of the local government for launching projects. The positions of schools can be strengthened by requirements related to inputs and effectiveness, which could only be met by additional expenditures. It is also worth mentioning that if the effectiveness of project financing is jeopardized by the intention of the local governments to ‘drain’ financing, the same weak effect can be expected from additional per student grants-type financing as well, since in the latter case local governments can even easier appropriate the financing for purposes other than education.

Besides these advantages project financing undoubtedly has disadvantages as opposed to the per student grant-type financing. On one hand operating such a system implies significant administrative costs; especially the evaluation of effectiveness based on measuring students’ performance.11 Moreover, the fairness of the selection of projects and the amount of the awarded grants can always be challenged. Additionally, some schools may find it problematic to prepare a proper tender. These two latter problems can be managed but the substantial administrative costs have to be taken into account when considering the introduction of a project scheme. A final disadvantage is that it is hardly acceptable from equity aspects (regarding the situation of the students) if the government excludes the schools educating disadvantaged students with poor effectiveness from those entitled for support. It is worth mentioning that the other two financing methods do not provide a stronger guarantee for all students to get appropriate education either, but such cases are obviously the most spectacular when the entitlement to financing is lost. It seems that it is easier to introduce a system attaching ex post additional financing to effectiveness objectives than a financing system defining the given level of effectiveness as a condition for support.

Project financing and supplementary per student grants differ from each other not only in their expected effect and costs. Project financing is sensible when and as long as the majority of disadvantaged students learn in a relatively small number of schools. Obviously it would not be very sensible to invite applications for programs enhancing more effective education of disadvantaged students if such students learned in nearly every school: the preparation and the selection of projects would require a great amount of additional expenditures from the schools and the government. Moreover, participation would not be essential for schools, either, as only a small number of their students would be concerned. Thus project financing fits better in the present situation characterized by a considerable segregation of disadvantaged students. This, of course does not mean that the objective of reducing segregation should be abandoned. Rather, project financing concentrates the available resources as much as possible so that the targeted schools become suitable and capable for more effective education of disadvantaged students and for providing an attractive educational choice for all students.

As opposed to this, the supplementary per student grant financing starts from the ‘ideal’ situation i.e. the termination of school segregation. This financing method may remain effective even if a relatively even distribution of disadvantaged students among schools is achieved. Moreover it seems to have more advantages in such a situation than the project-financing method. However, supplementary per student grants practically leaves the achievement of the policy objectives to the school owner local government. Project-financing gives more scope for the central government to actively shape the implementation of its policies at the school level while building mainly on school autonomy and initiatives, diminishing the role of local governments.

3.3. Market-type financing

The two financing methods discussed so far seem to be realizable alternatives given the present Hungarian conditions. However, the advantages and disadvantages of market-type methods are worth considering as their problems are often present in the other two as well.

3.3.1. Voucher system

The voucher system builds on the assumption that everybody knows what is the most useful for him/herself. As such, decisions have to be assigned to the individual consumer to the greatest possible extent. The essence of the voucher system is that the government allocates the amount of money earmarked for education directly to the citizens (and not to the local governments) in the form of a voucher. The voucher can be cleared in schools specialized for education in exchange of education services. Schools can submit these vouchers to the state and cash them. Every institution prices its services by estimating the costs, so on the market services of various qualities will appear at various prices (Friedman, 1997, Raganzas, 1997, Manski, 1992).12

The vital interest of schools will be to attract as many students as possible that will create a considerable competition among similar institutions. The institutions will only be able to operate in the long run if they provide high quality education, acknowledged by the consumers, both in pedagogical terms and in terms of infrastructure. The voucher-system gives a considerable leverage for parents, as the potential leave of their children would hit the schools financially.

By issuing vouchers of various nominal values the state can assert its educational policy preferences. A voucher of a bigger nominal value given to the disadvantaged may enhance that students have a real choice when selecting a school as they can obtain a more expensive school service providing quality education.

The obvious advantage of the system is that it requires little bureaucratic coordination. There is no need either for input or outcome regulations, as parents themselves will choose the institution that seems to be the best for them. The only role assigned to the state is to overcome market failures. Most importantly, it has to provide appropriate information concerning the operation of schools, mainly the quality of education so that parents can make befitting school choices. In the case of educational markets the information asymmetry stems mainly from the fact that the observable outcomes of students (e.g. rate of those continuing their studies, rate of admission, test results) are only partly determined by the quality of the service provided by the school. Other factors are the individual characteristics of students (capacities, family background, etc.) and the enforcement of the externalities among students (see e.g. Hanushek, 1986). If parents judge the quality of a school only by student performances they could easily misjudge how advantageous the certain school is for their children. So the administration has to provide parents with detailed and reliable information concerning not only student performance but also the student composition and the quality of education – cleared from the effect of student composition. In the voucher system it is the provision of information that makes assessing student performances and evaluating schools necessary. It is especially important concerning the education of disadvantaged students: a frequent argument against competition among schools is that it is more difficult for parents with lower qualification to obtain the necessary information, or that it is more difficult for them to evaluate this information and thus they are more likely to make ‘wrong’ decisions concerning school choice than parents with a higher social status (or they do not use this opportunity and mechanically send their children to the nearest school).

In settlements where – due to the size of the settlement – there is no competition as only one school may operate economically, the state – similarly to the vouchers for the disadvantaged – may reduce the problem by giving vouchers of higher nominal value.

One of the greatest problems of the voucher system is that it can develop segregated educational institutions along with a high quality educational service favorable for students if the administration does not set the value of the vouchers for the disadvantaged adequately. If this amount is smaller than the additional cost, schools have no interest to enroll these students. On the other hand, if the value of the voucher is too high, some institutions may specialize in educating disadvantaged students thus creating segregation.

It is obvious that the voucher system is unable to increase the educational consumption if it provides no, or very little utility for the families. In other words if the student does not like to study, and the parents also do not find education in schools necessary – which is not at all an unreasonable assumption for the disadvantaged families – it is very likely that the vouchers will not be used at all, or will not be used according to the policy purposes. Because for the schools it is vital to collect the vouchers from the students, a cooperation between the students and the schools are expected, where the school will not hold the lessons, so that children do not have to study, but they will hand the vouchers over to the schools.

As opposed to supplementary per student grant-type and project-type financing the voucher system cannot be applied as a financing instrument of a single educational policy program or as a complementary element of the present financing system. It is inconceivable that the government will finance programs aimed at more effective education of disadvantaged students by educational vouchers while the present two-level (central–local, school owner–school) financing system is retained for the whole system of education. The education of disadvantaged students cannot be detached from the basic education as it is done e.g. in the framework of afternoon programs and in the case of music education that is provided in a separate institutional system. If parents received vouchers exclusively for disadvantaged students in the present framework of financing, this would lead to a situation very similar to the introduction of the supplementary per student grant, the beneficial effects expected from the strengthening competition would be limited. The competition would be limited by the fact that the school owner local governments would be interested in taking away a part of the voucher-income of the schools (see above). The competing behavior of schools would be restricted by the fact that the basic characteristics of their activities (number of jobs, number of learning groups, pedagogical program) are set, approved of or financed by the local government. In the case of the non-state sector – similarly to the present state of affairs – entering the market would be restricted by partial nature of state financing: the per student grant that the school is entitled to for each student does not cover the full cost13 and the situation would not be different even if the vouchers of disadvantaged students covered all the additional costs of such education. The higher level of financial support would increase the market supply only with schools specializing in educating disadvantaged students.

This educational voucher system could be efficient only if introduced in the whole education system. In this case any actor could enter not only the market for the ‘disadvantaged’ but any segment of the market and thus integrated education could be offered – with an appropriate supply. This, however, does not seem to be a realistic alternative.

3.3.2. Quasi-market system

The quasi-market system, similarly to the voucher system builds upon the student number based school financing system and on the competition among schools. At the same time the two systems offer different solutions in at least three important aspects. First of all, the quasi-market system assumes competition within the state sector while the voucher system usually attribute a key role to the private sector forcing state schools to compete. Second, there is no price competition among schools in the quasi-market system, so none of them can dictate a price above the state financial support and make the parents pay the difference. Finally, the school owner local governments have an active role: they can provide local rules within the national framework. One of the most frequently mentioned examples of the quasi-market systems is the British Local Management of Schools (LMS). (The more detailed description of the quasi-market system see: Semjén, 1999.)

The three most important characteristics of the system are: the free enrolment in schools per student financing of institutions, possibility of opting out.

The free enrollment does not only mean that anyone can apply to the school of his/her choice but also assumes his/her acceptance, up to the capacity limit of the schools. This would eliminate the possible segregation.

All local school owners (local governments) are obliged to distribute educational contributions according to the number of students. Local governments have sources at their own discretion but they have considerably less influence on financing schools than in the case of per student grant financing.

All institutions have the opportunity to decide whether they want to be financed by the local or directly by the central government. If the local rules of financing do not suit the schools, they have the opportunity to leave (to opt out) the system and choose the financing guaranteed by the state. They would still receive per student grant but more, as these schools have to be compensated for the loss of local services.

Everything that is applicable to the market system is relevant for the quasi market system as well. So it is in the interest of the schools to attract more and more students to the school by improving quality (teaching, infrastructure or any other way). As the institutions depend directly on the per student grant, each school will develop an educational strategy by which they try to attract students from other schools, which will lead to a diversity of services. Segregation is expected to be reduced by competition among the institutions, free school choice and additional sources guaranteed by the state.

The problems are also similar to those of the voucher system – limited possibility of competition in small settlement schools, informational asymmetry, different valuation of the educational utility. It can only be applied to the whole educational system, i.e. not for a single educational policy program. The difference is that in this case we only speak of centrally financed public institutions.

In England (like in Hungary) teachers’ salaries are regulated by the state. As this sum constitutes the biggest proportion of the school budgets, financial management does not have too much freedom of choice which – in itself – restricts competition among schools.

The present financing system cannot be changed into a totally market-type system overnight but several lessons can be drawn that can help non-market based systems as well, such as the problem of information flow, for example. Market-type systems assume perfect (or nearly perfect) information flow. The state has a major role in providing information related to the quality of schools, regardless the system of finance. To ensure equal chances, it is essential that parents and students get the most easily comprehensible information. Those using educational services have to be able to size up what kind of education they participate in.

A similar issue is that of economies of scale. A great disadvantage of market-based economies is that with low demand (e.g. in small settlements) competition does not necessary develop, since per capita costs can be so high that it is not worth entering the market. In the case of other financing systems economies of scale also have to be taken into account. Otherwise sources are not guaranteed for quality education.

Market based systems may be criticized for their conception of human capital, one of the main issues of any system is the criteria of efficient education. Market based financing systems acknowledge it in an indirect way via the labor market. In a public system, however, the criteria for evaluating the education have to be set clearly. Based on this the state has to decide what forms of regulation are introduced and how are their effectiveness measured (input or outcome monitoring).

A quasi-market system may involve considerable central bureaucratic costs. An important aspect related to each financing system is what operational costs are required from the state. The smaller the costs of introducing and maintaining a system the more acceptable it is.

The inflow of private capital into public education is worth supporting whatever the financing method is. At present, school owner local governments may find it sensible to displace sources from institutions that are capable to operate without it (either in the form of a foundation or any other kind of support). In the long term, however, it may lead to a situation where private capital finds that its contribution does not improve the effectiveness of the institution (outcomes are the same as previously regardless the involvement of private capital). In this case, it will stop financing the institution and shifts the responsibility of operation back to the state. Private capital in all financing forms – with appropriate regulations – may have a proper role. National and local governments should encourage its involvement.

3.4. Comparison of financing methods

The basic features of the three financing methods are summarized in Table 2. It has to be stressed though that the choice among the financing methods is not an easy task, we cannot say that any of the discussed methods is more advantageous than the others – from all aspects and independent of the detailed regulation of the program. We think that working out requirements concerning local educational programs is a key issue deciding on the success of the program aimed at the education of disadvantaged students.

Finally it is worth considering the issue what resources can be used for the realization of the program aimed at more effective education of disadvantaged students. In the present education system HUF 410 billion was spent on primary and secondary education from the central budget in 2001 (Halász–Lannert, 2003). In the past years there was a considerable decrease in the number of students in primary schools. It means that more money can be spent on one student without a major change in the real value of total expenditures and that the student/ teacher ratio decreases. Moreover, the drastic decrease in student number will inevitably lead to school closures (schools with the highest per capita expenditures) which will release considerable sources in the public education system. The present financing system allocates HUF 17,000–95,000 additional per student grant for the education of the disadvantaged (above basic per student grant). Independent of the financing method redistribution of the present resources may cover the additional grants. We may therefore agree with the remark of Kertesi (2002) that a more effective education for disadvantaged students does not require an increase of total educational expenditures or the number of teachers employed, it can be solved by the redistribution of resources within the educational sector. Such redistribution, however, is not a simple task. On the one hand, in the short run the introduction of a new supplementary grant related to any new educational program increases educational expenditures. Redistribution presumes the reduction of the expenditures and the number of teachers of schools not involved in the program that can be attained by reducing the amount of the per-student-grant. This, however, is hardly executable overnight.

Table 2. Comparison of financing methods
  Supplementary per student grant School level project financing Market-type financing
Who is the addressee of the grant? School owner local government School Parents
What simulates schools to improve effectiveness? Local political accountability + central regulation Central regulation + school autonomy Competition among schools
Does additional grant reach schools that educate disadvantaged students? Presumably only partly Presumably mostly Yes
To what extent is it expected to contribute to the reduction of segregation? Special learning group grant ? not at all integration grant ? partly Depending on central regulation (competition requirements), imilarly to integration grant Depending on the amount of additional grants for disadvantaged students
Does it encourage the application of new pedagogical methods? Depending on central regulation, difficult to enforce Depending on central regulation, easy to enforce Yes
What requirements concerning local educational programs can be attached to it? Input requirements + ex post rewarding of outcome objectives Input requirements concerning the process and outcomes of education Nothing concerning the given program (in general: minimum conditions for operating school)
Monitoring With ?weak? requirements checks are not necessary Checking the realization of local programs is essential Only information collection
How much is it suitable for taking individual cost differences of schools into account (e.g. small settlements)? To a limited extent To a great extent Practically not at all
The transparency of distribution of grants Cannot be questioned Doubtful Cannot be questioned
The amount of the grants Any amount of grants can be distributed this way Grants that make viable local programs possible are worth distributing this way Only viable with grants covering costs
Could it fit as a complementary element into the present system of financing education? Yes Yes No
What is the major role of the central government? Setting generally enforceable, simple requirements concerning educational programs Enforcing the requirements related to the local programs in the evaluation and checking competitions Assessing the quality of schools, providing reliable information
Administrative costs Very low High If assessing the quality of school is taken into account: relatively high
The most important advantage Simple, transparent Requirements related to local programs are easy to enforce Builds on consumer decisions
The most important disadvantage Requirements related to local programs are more difficult to enforce, without them: weak effect High administrative costs + lack of transparency Small settlements + providing reliable information to parents

In practice a much more plausible solution is the gradual reduction of per student grants (or possibly keeping them at the nominal level). On the other hand, the more the central government wants to find resources for the program by such redistribution within the educational sector, the more likely is that local governments try to take away the additional grants from the targeted schools (see above) and redistribute them to the others, adjusting to the inflexibility of the local system of institutions.

4. Additional policy instruments

Financing methods presented in the previous chapter can be applied in the distribution of the grants necessary for the operation of schools. The successful realization of a more effective education of disadvantaged students may require several further policy measures. These are summarized briefly below without discussing their effects in detail.

In order that schools can use new pedagogical methods the central governments has to support the development, improvement and implementation of developing educational methods in institutions specialized in this area. Project and market-based financing (buying services by schools) are both conceivable.

The key issue of educating disadvantaged students is in-service teacher training. Besides general grants, direct financing of in-service training institutions (or a part of them) may also be important. Reliable assessment of the programs aimed at more effective education of disadvantaged students, introducing outcome incentives and providing appropriate information for parents assumes the operation of an individual-level evaluation system. To make minority culture present in school education additional grants are required. In the supplementary per student grants system it could be financed by a further per student grant, under project financing either by a further per student grant or on a project basis (in this case the latter seems to be simpler and more effective).

It looks as if the appropriate instrument for helping disadvantaged students to continue their studies at upper secondary level is not the financial support given to secondary schools but individual scholarships. The financing system has to create the possibility for further studies; the government has to support individual willingness to continue studies. In most cases it seems to be unnecessary to finance students either through local owners or institutions, in secondary and especially higher education the introduction of an individual-level grant system seems to be sensible. Experience shows that such a system can work well (e.g. József Attila Program). Chances of disadvantaged students for continuing studies may be improved by providing boarding school/dormitory places at a reduced price.

Extending kindergarten supply is one of the key issues of more effective education of disadvantaged students. Besides financing the maintenance of kindergartens investment grants may also be necessary by which local governments can expand existing kindergarten buildings or establish new ones in settlements where there is none at present.

5. Conclusions

In this paper three possible financing methods aimed at a more effective education of disadvantaged students were discussed. Lump sum supplementary per student grants and school level project financing can both fit into the present system, market-type solutions could be realized only by transforming the whole system of education finance.

Considering the expected effects of these methods, none of them is optimal in all contexts. On the one hand, each method has its disadvantages, on the other, the incentive effect of the financing methods depends considerably on the regulatory instruments attached to them. Even a well-functioning financing system creates only possibilities for developing pedagogical practice appropriate for the given context. More resources do not lead to better pedagogical work automatically. To make institutions capable for educating disadvantaged students in a more effective way, substantial reforms are required in pedagogical work. The central administration may encourage this by setting requirements for local programs and by additional policy measures, independent of using supplementary per student grants or project financing.

The feasibility of the financing methods discussed is different at the various levels of education. We think that providing grants for schools or school owners is an appropriate instrument for primary schools and kindergarten, continuing education at secondary and higher levels is worth being encouraged by individual-level grants.

At the same time the problems of educating disadvantaged students in kindergartens and primary schools are also different, so the same financing method might not produce the same result in the two institution types. In case of kindergartens the primary objective is not a more effective education of disadvantaged students but an increase in their enrolment rate. To achieve this objective per student grant financing seems to be more suitable than project financing. As kindergarten applicants are rejected not because the child is disadvantaged but due to the insufficient infrastructure of the institutions (Vágó, 2002), it seems to be a sufficient incentive if the grant per student equals the amount of the average costs per child in the institutions. Thus, on one hand these kindergartens will not have to bear disadvantages if they admit another child, while grants will cover the total costs of supply, therefore there will be no additional expenditures for disadvantaged parents. The elimination of segregation can be set as one of the conditions for receiving such grant, similarly to the ‘integration grant’ introduced for primary education. It would be advisable to attach a simple outcome requirement to it. Local governments would be entitled to additional grants if the percentage of disadvantaged student in kindergartens reaches a certain level. That would encourage kindergartens to contribute to the solution of the problems emerging from the parents’ part. In the settlements (regions of a settlement) where there is no kindergarten at present, although there would be market demand for it, it is worth considering the introduction of investment grants that make opening new kindergartens easier.

For problems emerging in primary schools the various financing methods give various solutions depending on what regulation context the policy makers have in mind. Without a fundamental transformation of the present financing system the introduction of market-based financing systems is inconceivable, therefore they are not considered to be real alternatives in short term. Project and supplementary per student grant financing, however, are both viable options.

If the government wants to apply stricter regulatory mechanisms and/or ones related to the process of education, project financing seems to be more appropriate. However, if it wants to empower institutions and local communities to control (it sets weak central requirements for schools) the per student grant financing seems to be a more sensible solution. In this case project financing would hinder the operation of the institutions by unnecessary bureaucratic difficulties. From the aspect of schools along the present per student grant mechanism it seems important to take the prevailing disparities in the local tax base into account. Under project financing this can be taken into account discretionally when deciding on the selection of projects and the approval of budgets, while in the case of supplementary per student grants special state regulation seems to be necessary for the additional financing of small settlements.

References

Babusik, Ferenc (2001): Roma gyerekeket képző általános iskolák speciális kínálata (Special supply of schools educating Roma children). Új Pedagógiai Szemle, 2.

Bánfi, Ilona (1999): Az iskolai teljesítményt befolyásoló háttértényezők. Új Pedagógiai Szemle, 6. (Factors influencing school achivement).

Clotfelter, C. T.–H. F. Ladd (1996): Recognizing and rewarding success in public schools. In: H. F. Ladd (ed.) (1996): Holding schools accountable. The Brookings Institution, Washington, D. C.

Downes, T. A.–T. F. Pogue (1994): Adjusting school-aid formulas for the higher cost of educating disadvantaged students. In: National Tax Journal, 47.

Duncombe, W. D.–J. M. Yinger (1999): Performance standards and educational cost indexes: you can’t have one without the other. In: H. F. Ladd–R. Chalk–J. S. Hansen (eds.) (1999): Equity and adequacy in education finance. National Academy Press, Washington.

Elmore, R.–C. H. Abelmann–S. H. Fuhrman (1996): The new accountability in state education reform: from process to performance. In: H. F. Ladd (ed.) (1996): Holding schools accountable. The Brookings Institution, Washington, D. C.

Friedman, M. (1997): Public Schools: Make Them Private. Education Economics, vol. 5, No. 3, 1997.

Friedman, L. S. (1984): Microeconomic policy analysis. McGraw-Hill.

Halász, Gábor – Lannert, Judit (2003) (eds.): Jelentés a Közoktatásról (Education in Hungary). National Institute of Public Education, Budapest.

Hanuschek, E. (1986): The economics of schooling. In: Journal of Economic Literature, 24.

Havas, Gábor–Kemény, István–Liskó, Ilona (2002): Cigány gyerekek az általános iskolákban (Roma children in primary schools). Oktatáskutató Intézet, Új Mandátum Kiadó.

Havas, Gábor (2001): Elkülönítés és következményei, a cigány gyerekek az általános iskolában (Segregation and its consequences, Roma children in primary schools). In: Semjén András (ed.): Oktatás és munkaerő-piaci érvényesülés (Education and assertion on labour markets). MTA Közgazdaságtudományi Kutatóközpont.

Henderson, V.–P. Mieszkowski–Y. Sauvageau (1978): Peer group effects and educational production functions. In: Journal of Public Economics, 10.

Horn Dániel (2003): Az oktatási szolgáltatások összetételének elemzése. Kézirat. (Analysis of the composition of educational services.) Manuscript. National Institute of Public Education, Budapest.

Kertesi, Gábor – Kézdi, Gábor (1999): A cigány népesség lélekszáma Magyarországon a kilencvenes évek elején (Figures of Roma population in Hungary in the early nineties). In: Glatz Ferenc (ed.): A cigányok Magyarországon. (Roma in Hungary.) Magyar Tudományos Akadémia, Budapest.

Kertesi, Gábor (2002): Oktatási reformterv a tanulási problémákkal küszködő, hátrányos családi hátterű gyermekek megsegítésére az alapfokú oktatásban (Reform curriculum for helping disadvantaged children with learning difficulties in primary education). Esély.

Liskó, Ilona (1999): Az oktatás minősége és az iskolák presztízse (The quality of education and the prestige of schools). Educatio, 3.

Liskó, Ilona (2001): A cigány tanulók és a pedagógusok (Roma students and the teachers). In.: Romák és az Oktatás (The Roma and Education). Iskolakultúra.

Liskó, Ilona (2002a): A hátrányos helyzetű tanulók oktatásának minősége (The quality of the education of disadvantaged students). Új Pedagógiai Szemle, 2.

Liskó, Ilona (2002b): Cigány tanulók a középfokú iskolákban (Roma students in secondary schools). Új Pedagógiai Szemle, 11.

Manski, Charles F (1992): Educational Choice (Vouchers) and Social Mobility. Economics of Education Review, Vol. 11, No. 4.

Murnane, R. J.–F. Levy (1996): Evidence from fifteen schools in Austin, Texas. In: G. Burtless (ed.): Does money matter? The effect of school resources on student achievement and adult success. Brookings Institution, Washington.

Pik, Katalin (2002): Romák és óvodák (Romas and kindergertans) Kisebbségkutatás, 11/1.

Raganzas, P (1997): Competition and Private School Vouchers. Education Economics, vol. 5, No. 3.

Schwab, R.M.–W.E. Oates (1991): Community composition and the provision of local public goods. In: Journal of Public Economics, 44.

Semjén, András (1995): Miért jobb? Normatív iskolafinanszírozási modell a Józsefvárosban (Why is it better? Per student grant school financing model in Józsefváros). Iskolavezetés, 12.

Semjén, András (1997): Állami szerepvállalás és finanszírozás a közoktatásban: merre tovább? (The roles of state and state financing in public education: how to go on?) Új Pedagógiai Szemle, 2.

Semjén, András (1999): Hatékonyság az oktatásban (Efficiency in education). Educatio, 3.

Vágó, Irén (2002): Óvodai intézményrendszer, óvodai nevelés az ezredfordulón (Kindergarten institutions, kindergarten education at the millennium). Új Pedagógiai Szemle, 12.

Varga, Júlia (1998): Oktatás-gazdaságtan (Education economics). Közgazdasági Szemle Alapítvány, Budapest.

Varga, Júlia (2000): A közoktatás-finanszírozási rendszer hatása az egyenlőségre, 1990–1997 (The impact of the financing system of education on equality, 1990–97). Közgazdasági Szemle, XLVII.

Vári, Péter (ed.) (2003): PISA-vizsgálat 2000. (Pisa-survey 2000.) Műszaki Kiadó, Budapest.

A honlapon található tanulmányok, egyéb szellemi termékek, illetve szerzői művek (a továbbiakban: művek) jogtulajdonosa az Oktatáskutató és Fejlesztő Intézet. A jogtulajdonos egyértelmű forrásmegjelölés mellett felhasználást enged a művekkel kapcsolatban oktatási, tudományos, kulturális célból. A jogtulajdonos a művekkel kapcsolatos anyagi haszonszerzést azonban kifejezetten megtiltja.