DECENTRALIZATION AND EQUITY: OPPORTUNITIES AND LIMITATIONS OF CONTEMPORARY EDUCATIONAL POLICIES

International Conference on Federalism Mont-Tremblant, October 1999 BACKGROUND PAPER DECENTRALIZATION AND EQUITY: OPPORTUNITIES AND LIMITATIONS OF CONTEMPORARY EDUCATIONAL POLICIES Cecilia Braslavsky Professor, Department of Education Sciences University of Buenos Aires SUMMARY The purpose of this paper is to promote discussion of ways to reconcile the processes of educational “decentralization” under way in many countries of the world with the objectives of greater equity. For this, we begin by showing the ambiguity of the concept of “decentralization” and its limitations as the single guiding concept for configuring and regulating all educational services. We put forward various rationales for and types of decentralization. Secondly, we state that educational decentralization has five paradoxical effects and we maintain that they all accentuate inequalities, especially social inequalities, but also variations between municipalities, provinces and countries. Thirdly, we propose some ways to avoid these paradoxical effects or at least to prevent them from gaining a hold. 1. The Ambiguity of the Concept of Educational “Decentralization” In the last few decades, countries as different as Germany, Great Britain, Nigeria, Argentina, France and Chile have encouraged or accelerated changes in decision making and management in the provision of educational services. In all countries, even large unitary nation-states like France, stable federations like Australia or ones racked with internal conflict like Argentina, moves to “decentralize” are a fundamental part of this process. Educational “decentralization” refers to a series of quite diverse processes that may take place within large unitary states or within each of the components of a federal state. Within this diversity a set of typical situations may be distinguished around two variables. The first variable concerns the supposedly geographical location, but in reality it is political, namely the location where decisions are made or the day-to-day management of educational services is carried out. The second concerns the relationship between the political authority (also considered the “centre”) and the professionals and social interest groups involved in education. From the perspective of the first of these variables, one type of decentralization is the promotion of administrative restructuring that does not change the underlying logic of the system and the exercise of political power on educational issues. It is concerned instead with promoting strategies to move decision-making centres lower in an existing hierarchy. It tends to operate by organizing a territory in zones or regions in which delegates of the central authority are responsible for making everyday decisions. This type of “decentralization” is supposed to make large educational organizations less bureaucratic and remote, but in reality it often generates new bureaucracies, although they are distributed differently over the territory. From this same perspective, the second type of decentralization involves redistributing power within the various levels of government, as from the national government to the provinces or municipalities, depending on the legal system in each country. In Argentina, for example, the national government historically competed with the provinces in providing educational services. Through a long, uneven and complex process from 1968 to 1992, the services were transferred from the national government to the 24 provinces in the federation. In Chile, however, the national government was historically responsible for regulating and providing all educational services. In a rapid process during the military dictatorship of Augusto Pinochet, responsibility for providing these services was transferred to the municipalities, while regulation remained with the national government but was limited to the evaluation of results. The redistribution of power from a political centre to a social network may involve decentralization associated with movements involving professional groups, users of the services or interested parties in general. The processes of unbundling functions and of reorganizing power within various levels of government may or may not be a result of the national or provincial government decentralizing its own authority and thus making more room for decision making by administrators and teachers in educational institutions. These shifts may not necessarily produce a “decentralization” in terms of the first variable mentioned. Territorial decentralization is associated with redistribution of power from the political and bureaucratic centre to educational professionals when school autonomy is strengthened as the last stage in political/territorial decentralization. The State of Victoria, Australia, for example, has delegated key issues to the professionals in the schools (Caldwell, 1997). In the last decade, North Rhineland–Westphalia and other German states have been the scene for highly significant debates and experiments on why, how and how much autonomy to give to schools (3), although they have not held equally important debates on whether or not to reorganize the territorial organization of educational authority in each of their highly centralized federal states. The debates in Germany, much of the United States and other countries of the world show the tension between proposals more or less identified with the kind of decentralization that would give more power to professional bodies and those that basically seek to redistribute power in ways that involve parents, business people and other social interests and sectors. The foregoing shows that the type of process meant by the term “decentralization” reflects the cultural traditions of each country, the reasons for their liberalization and their model of political and economic development. In fact, the processes of educational decentralization can be articulated around various traditions. One of them is associated with the desire to deal with the variety of actual situations involving social needs, as well as the conviction that people are entitled to propose and manage courses of action to this end (4). Consistent with this tradition, in some cases the processes of educational decentralization were initiated by demands from families for more significant involvement in decisions affecting their children’s education (Munin, 1999). Along with this tradition, in other cases educational decentralization was promoted by teachers who wished to encourage marginally involved groups to make educational innovations beyond the restrictions imposed by rigid government standards (Townsend, 1996). Another tradition supposes that bringing decision making closer to those affected will soon result in better allocation and control of spending on education (5). Along these lines, the World Bank stated in 1988 that decentralizing authority over both expenditure and revenue could improve the allocation of resources in the public sector by more closely relating the costs and benefits of local public services; however, there is little empirical evidence of this. Regardless of the reasons for territorial decentralization, with greater or less participation of education professionals, parents and other social interest groups, the actual processes took place during a period dominated by macro-economic concerns strongly related to the need to increase the cost-effectiveness of enterprises in various national contexts. Issues like the need, or even the possibility, to rethink the contribution of education to national integration and cohesion or the need for, and ability of, education to make national economies more productive as a whole were hidden by fascination with the emergence of globalization as an abstract trend removed from politics, and the local dimension was overridden by nationalism in thought and social action. Therefore educational decentralization had at least two peculiar characteristics, which were much more pronounced in the southern countries (6). The first is a cost-cutting bias. The second is a lack of preparation for the smooth operation of the emerging model of educational governance. In other words, education was decentralized in the belief that it would democratize the exercise of power and/or improve efficiency, but the decentralization occurred during decades that were more concerned with saving public money than with democratization or greater effectiveness. In fact, the search for greater efficiency was seen mainly as an irreplaceable opportunity to reduce the total spending on education and a key to making the adjustments deemed essential to pay down a growing foreign debt. In various Latin American countries, for example, in the first stage of educational decentralization, no attention was paid to such issues as evaluation, information gathering, curriculum design and training of new professionals to implement new educational policies and practices. In Colombia and Argentina, for example, management of education was decentralized and responsibility for the services was transferred to the provinces, with no regard to what may be considered necessary for their proper functioning. Behind this neglect, we see a disregard for the purposes of education. In the medium and long term, educational processes and systems are meant to train everyone so that they can plan and live their lives, take part in creating collective wealth and continually re-establish a democratic political order that peacefully resolves the inevitable conflicts arising from the struggle for their fair share and in adjusting to increasingly diverse cultures. In fact, education was decentralized in this stage not in order to recognize and expand citizens’ opportunities for exercising power, to improve the quality of what people learn or to reflect the growing cultural diversity that accompanies the equally important processes of emergence of a global culture, neither was there interest in the mechanisms closely related to quality such as information gathering, evaluation, curriculum design and training of new professionals. The question of education itself was treated as a black box of uncertain importance. In practice, use was still made of mechanisms and methods that long predated decentralization. As awareness grew of the importance of national competitiveness in the context of globalization, of technical progress as a factor in national competitiveness and of education as a significant element in introducing technical progress, decentralization was accompanied with growing attention to quality control mechanisms in education and to information gathering. Accordingly, where no testing systems or procedures to publish the results existed, they were created. It was believed that if the public had more information, it could make better decisions individually and thus the overall situation would improve (7). In this context, decentralization brought with it a series of opportunities, but also a set of paradoxes. The opportunities arose from the underlying reasons for decentralization as a strategy for redistributing power. They were based on the expectation that bringing the centre of strategic decision making and of management closer to the place where the service is provided and allowing public participation in this decision making and management would better meet educational needs and not be bound by pre-existing forms. In other words, the redistribution of power through decentralization of decisions from a single central point to all the formerly peripheral points, or better, from the apex of a system organized as a hierarchical pyramid to the base and from the summit to the grass roots can release a strong individual and collective creativity that will make it easier to achieve new and better approaches. The paradoxes all revolve around the relationship between decentralization and equity in the quality of education provided. In fact, it gradually became obvious that the strategy of decentralization isolated from systemic processes of educational transformation or only associated with regulatory mechanisms through information and control would not achieve the objectives of quality education for all, especially since nothing guaranteed that decentralized decisions would do more to improve the students’ learning experience, which was increasingly seen as necessary. 2. The Paradoxes of Educational Decentralization at the End of the Century In the last two decades, awareness has grown of at least five paradoxes of decentralization. In some countries, the five paradoxes occur together. In others, some are more common than others, while in others they scarcely occur. As a hypothesis, one may suggest that paradoxes occur more frequently and with more severity in countries whose educational systems are traditionally undemocratic, where socio-economic and regional inequalities are very pronounced, where regulatory systems are weak, very outdated or simply absent and where professionals are undervalued and badly paid. The first paradox of educational decentralization is the existence of situations of anomie. The decentralization of education was proposed as a way to build a new order, but with an end in mind. Its greatest promise is the building of a more democratic and creative order than the centralized one. In its more radical form, it suggests that only self-imposed standards will exist and therefore they will be much more strongly accepted and implemented by educators. Nevertheless, provinces, municipalities or groups of parents and teachers accustomed to operating in the regulatory framework of pyramidal and centralized hierarchical systems, with standards set in some remote and distant place that they only had to follow, often are disconcerted when the central authority withdraws. In such cases it would appear that the old practices had not been internalized as their own but were only followed for fear of bureaucratic control. Educators are thus unable to maintain the old order or set up a different one. They seem to be going around in circles until they finally demand that the central authority which wanted to distribute power take it back and restore for good the old order that existed before decentralization or some other order that is more accepted because of how it was established and imposed than for its ability to provide a good education. It also happens, however, that other jurisdictions given authority as a result of educational decentralization do not miss the central power; in other words, they do not feel the need to create another order. The second paradox of educational decentralization is the continuation of mediocrity or the maintenance of poor institutional and pedagogical practices learned in previous decades. In fact, decentralization assumes that there are other ways that are more efficient, if not more effective, to organize educational institutions and arrange teaching methods. It also supposes that every place is capable of inventing its own approach. However, some research shows that, far from inventing alternate forms of regulation or organizing institutions and practices, the decentralized authorities may simply follow a routine and lose sight of why these same processes were done in a certain way, even though nothing and no one require them to do so (Braslavsky, 1993). The third paradox of educational decentralization is submission to the will of the community. One of the reasons that groups of parents or teachers want more participation in educational decision making is the need to find other forms of school organization and to vary uniform teaching processes. Pyramidal and hierarchical education systems customarily set rules for the processes, on the assumption that identical processes would yield identical results. Thus, if education systems pursued egalitarian results through the exercise of citizenship and the training of whole persons, or to use more recent terminology, self-actualizing individuals, it was necessary to ensure identical processes. Identical processes would give identical educational results in key aspects for personal development and citizen participation. But even in decentralized education systems, schools tend to standardize aspects of cultural education that need not be uniform while they maintain differences that must be abolished in the search for equity. Children who undergo these identical processes are the same in ways that the communities want them to continue to be different, such as knowledge of languages and customs. However, they do not reach the same standards in other aspects such as mastery of the national language, mathematics or skill in handling information. Decentralization was supposed to strike a different balance, more attuned to the interests and needs of the communities. This different balance was struck, but in many cases at the cost of narrowing the students’ horizon in a way that reflected only the immediate community or environment. In Argentina, many provincial curricula developed between 1984 and 1992, for example, put so much emphasis on local and provincial history that national and world history practically disappeared from the core of compulsory subjects. Some principals and teachers were particularly pleased with this development, because, as one of them said, they could now deal just with “the part of the river that goes by our city.” The fourth paradox of decentralization is perhaps the strangest. It runs counter to the previous one. It involves submitting to diversity by uncritically copying projects designed for very different situations. In this case, the jurisdictions to which decision-making and executive responsibilities were transferred assume new roles and start to innovate. Decision makers look around near and far and try to import policies, strategies and projects developed by other provinces, municipalities or educational institutions. In Victoria, Australia, an analysis of a large number of institutional educational projects showed that the topics and strategies suggested for institutions serving immigrant communities did not reflect diversity and repeatedly referred to the same teaching objectives and proposed the same strategies for achieving them (Townsend, 1997). In some cases, this attitude comes from outside the agencies that make decisions on educational policies, strategies and projects and implement them. In the provinces or states that make up federated Latin American countries, for example, this attitude is promoted by international lending organizations that propose homogeneous solutions for heterogeneous situations without a detailed analysis of their economic and pedagogical characteristics and especially their political and cultural differences. For educational institutions, this attitude is promoted by consulting firms that propose identical educational projects for very different situations, and these institutions, at least in these countries, in turn contribute to the market for the consultants’ services. In some of the northern provinces of Argentina, where teachers earn an average monthly salary of US$500, these projects sell for US$2,500. Unfounded belief in the benefits of competition among educational institutions for scarce funds and fear of public authority in places where memories are strong of irrational behavior by those in charge contribute to these supposedly undesired effects. The fifth paradox is the most serious of all and cuts across the other four. It reproduces the poor quality of education of the whole population in each country affected. The decentralization policies adopted during the cost-cutting period produced a deeply flawed market-based remedy for education, in which the reference parameters did not change and were not understood in the same way by all. Closeness to the centre formerly responsible for making decisions and managing the services, access to information from other centres, the educational history of the province, municipality or institution, and other factors made it possible for some of the newly responsible jurisdictions to have information while others did not. But beyond the internal differences, in countries where the regulatory mechanisms for education, and in particular curricula, plans and programs, were frozen in time as a result of years of dictatorship or lack of thought about building alternatives to reinvent government as an organization to rally the public and support public enterprise, and of disinvestment in education, the opportunity was lost to formulate ambitious new objectives for collective improvement of the quality of education. Accordingly, it may be suggested that beyond reinforcing inequalities that are contrary to the achievement of social justice and consolidation of democracy in the medium and long term, the type of decentralization that occurred had a negative impact on the quality of education and all the services and practices of every country affected, reinforcing trends and entrenching inequalities between countries (Ortiz, 1998). 3. Conditions to Make Decentralization Promote EquityAwareness is growing of the dangers to equity from decentralization strategies isolated from systematic processes of educational transformation. Accordingly, there is a need to promote multidimensional policies of educational change that support the value of decentralization but reconnect it to the two traditions mentioned above (redistribution of power and search for efficiency, broadly defined) and to two other previously neglected questions: improvement of educational quality and reversal of trends to inequality, now aggravated by the introduction of new technologies and by new forms of labor organization. First, it was realized that educational achievement does not always correlate with the degree of decentralization of educational systems. In Japan and Korea, for example, children and youth achieve good educational results in the key subjects of the modern world, especially mathematics, within centralized educational systems. In Latin America, Cuba has the best results in language and mathematics (Laboratorio, 1998). This realization was also based on evidence from new processes in countries with educational systems that arose from a decentralized bias, such as Australia, Great Britain and the United States. There, the imperatives of international competitiveness in a knowledge-based society where quality of education is a must strengthened the processes for developing parameters or standards where they had never existed before and in some cases spurred decentralization. Thirdly, empirical research in countries that had decentralized in various ways showed some of the paradoxes mentioned earlier (8). It was already anticipated that during the cost-cutting period proposals for decentralization of education would be accompanied in some countries by the setting of desirable standards or parameters and the establishment of educational quality evaluation systems, where they did not exist, that would inform a growing number of increasingly varied decision makers of the results achieved in meeting these standards. During the military dictatorship in Chile, for example, an educational quality evaluation system was organized and started to provide parents with information on the results obtained by students in each school and municipality. However, this mechanism of regulation through information and control could not prevent some of the paradoxes mentioned. This situation gives rise to many questions. Here are some obvious ones. What is the point of setting standards if teachers do not share them, understand them or know how to achieve them? What is the point if parents and students do not understand them either? What is the use of competition among schools to improve the quality of education if education in a whole community or province has deteriorated badly and people cannot move from one to another? Setting standards seems to be a necessary but not a sufficient condition for improving the quality of education together with decentralization. It has long been known that the quality of education rises where the method or group of schools where it is given is evaluated (Boudelot and Establet, 1989), and evaluation implies comparison against a standard. But the usefulness of these standards depends significantly on how they were developed, on their suitability for imparting lasting abilities or skills (Braslavsky, C., 1993; Perrenoud, 1998) and the existence of appropriate curricula so that administrators and teachers can effect a transition from rigidly set external practices to ones of their own design. For this, it is essential to combine decentralization and standard setting with empowerment, which means strengthening the ability of educators to make the right decisions to improve the quality of education in their own environment, without losing sight of the challenges and requirements of a globalizing world. In other words, the right information, much more multifaceted than what now exists, can shed light on unknown aspects of the teaching process and the underlying socio-economic, political and cultural conditions. But only educators with power to improve the quality of education and frames of reference and tools for this purpose can make the necessary changes to it. The first condition then for the opportunity provided by the decentralization of education to improve the quality of education for all, that is, equity, is the existence of some commitments to the results of education, which may be called standards, expectations of achievement, general objectives or anything else. The important thing is not so much these “standards” but rather the public commitments they imply for the relatively permanent skills and abilities that education must develop, a contract between the society that requires and demands education, the professionals responsible for providing it and the scientists and entrepreneurs who generate knowledge and appropriate procedures for developing these skills and abilities. Some federal countries that have set standards through an agreement among technical experts have been unable to make these standards guide teaching practice in the same way as standards produced through consultation involving teachers and major social sectors. This observation applies not only to standards but also to course materials and curriculum designs. For example, in some federal countries like Argentina, the impact of new curricula on the quality of education in the provinces seems more closely related to how the curricula were developed than to the quality of the product. It is also essential to propose curriculum development methods that differ from the prescriptive curricula of decades past, typical of centralized and pyramidal educational systems. In decentralized and democratized systems that seek equity, the criteria for action should be made visible and empirically based; they should offer alternatives and allow teachers to develop their own approaches. In the curricula of centralized, hierarchical and bureaucratized systems, supposed experts told supposedly non-expert teachers what they had to do, how to do it and in what order, without explaining why or facilitating horizontal interchange and learning. In the myth-making stage of decentralization, poorly paid teachers are alone with their students, unable to share experience or benefit from collective wisdom. The curriculum materials needed now could explain why various possible strategies can, and not necessarily must, be put into practice, providing specific colleagues in specific and similar situations with results and also inventing other strategies. In some countries in the 1990s, it was thought that these alternatives had to be like a puzzle to put together, that is, a vast wide-open set of unconnected resources from the available supply, out of which the professional team in each educational institution could draw according to their needs. However, lack of training in the exercise of professional freedom and inability to play the role differently made the results poor in some cases. Now a type of curriculum development is being tried called “mapping”. The “map” gives the rules of the road and shows which ways are open or closed from the start and in view of the desired destination and which are possible under certain conditions (time, type of equipment, etc.). The second condition for promoting equity in the context of decentralization is a reconceptualization of the roles and functions of education professionals. First, the issue is raised in relation to the larger number of more diverse officials and individuals involved in the governance of education. Historically, in many countries education was governed exclusively by officials of the national ministry of education. In other countries, it was run by state or provincial officials, who also held all the authority on strictly pedagogical questions. Now, national, provincial and municipal officials, supervisors, school principals, NGOs and businesses are involved in the management of education. To give only a quantitative example, in each of the largest Latin American countries are hundreds of municipalities, most of which have educational programs or projects with officials involved in designing and implementing educational policies and strategies of various kinds. Decentralization can allow decision making and management to be truly democratizing. In relatively small education systems with centralized models of governance, decisions were made at the top of the hierarchy and intermediate officials and school principals had to carry out decisions made by others. They were not expected to report on the results obtained but only to follow the processes laid out. If resources were scarce, it was thought sufficient to inspect the processes. For this, there was a special group of bureaucrats known to all as the “inspectors”. It was not considered so necessary to evaluate results by administering the same costly tests to more and more students. In fact, as expected, in centralized pyramidal systems, the same courses of action and strategies were to be pursued in different contexts. In a decentralized, flattened, open system, this would not happen. Such systems need to be able to identify requirements, formulate alternatives, consider advantages and disadvantages, coordinate action, implement decisions and monitor their impact and finally report to the public on the results. The new professional profiles needed to run education systems and to teach are very different from the previous ones. Among the new abilities required are interdisciplinary and multidisciplinary thinking on future scenarios and strategies for change, the ability to produce and use empirical information on management and teaching processes, understanding of the existence and specific characteristics of various levels of decision making and action (national, provincial, municipal, school) and recognition of various factors (institutional, pedagogical, community) that affect educational success and failure. All this implies ability to work as a team, to handle new communication and information technologies and a high readiness and capacity to deal with continuing change and learning. Some empirical research on recently decentralized systems shows that the difference between good and bad education, after controlling for socio-cultural variables, is due to the kind of teaching given. Principals with more experience, educational officials who spend more time in the institution and personally visit the classes, and teachers who are trained to provide discipline and teach their subjects and not to emphasize their socializing role are able to break the supposedly unbreakable connection between the students’ socio-economic status and what they learn in these institutions (Braslavsky, Gluz and Calvo, 1999). At the same time, many politicians and professional groups began to insist on the following additional conditions: the need to generate a climate of recognition and public trust in education, assurance of basic unconditional funding for all education throughout the country, regardless of the resources in each province, municipality or school, and encouragement of networking among institutions. In fact, at the height of the cost-cutting period, when emphasis was put on controlling results without consideration of the characteristics of the frames of reference, of the tools and of the abilities of educators to obtain these results, everyone mistrusted everyone else. Parents, business people and political leaders doubted that education professionals could obtain good results. Teachers doubted that the government would support their efforts. Students doubted the public’s interest in their education. The senior educational officials in each country and economic leaders seemed to think that if decentralization was a panacea but education did not improve when it was decentralized, it must be the teachers’ fault. It is hard to recognize weaknesses or biases in the strategy pursued, for the politicians and business leaders who promoted it, as well as for the social interest groups and intellectuals who, for other reasons, accepted or supported it. The teachers and their unions are sometimes at a loss. They do not think that it is possible or desirable to go back, but they do not always see the way out. Often attacked and blamed by presidents and ministers, they seek refuge in conservative public attitudes that do not even represent their own personal convictions (9). Thus a profession that is essentially communicative and interpersonal does not improve. In considering the empirical evidence of the conditions just described, we must accept that the persistence and intensification of inequalities has much to do with the lack of public policy and of social movements to generate appropriate frames of reference and to develop tools and strategies to strengthen educators. Decentralization is a gamble for educators. Therefore it cannot be accompanied by unpromoted control processes or by blaming those who are expected to make their teaching practices more dynamic. On the contrary, the decentralization of education can fulfil its promises only if it generates trust and cooperation among the many players involved in remaking education. Therefore decentralization must not be accompanied by policies and strategies that increase competition among schools for the basic resources they need to operate. Decentralization is a way to diversify supply. In this regard, it is desirable to be able to compare different experiences and even to have some competition among provinces, municipalities and institutions, but the various offerings must be equivalent in quality while differing in identity. To be equivalent in quality, all educational services must have a base of assured funding, regardless of the wealth or poverty of the jurisdiction in which they are given, be it a municipality, state or province. This base of funding makes it easier to cooperate on the quality that is essential in a plan for development and democratic living in the long term, and can only be built with equity, on the national and even global level. In some federal countries, per capita GDP varies by as much as a factor of 20 among the states or provinces. Is it morally right and politically desirable that in these cases funding of educational opportunities is related to the ability to pay of the states or provinces? Differences among municipalities are just as great, if not greater. With such differences, is it possible to reconcile decentralization and equity? Everything seems to indicate that they can only be reconciled with compensating policies that ensure a base of funding and also basic conditions of educability for the students. This means that other social policies are needed to ensure that the basic needs of disadvantaged students, especially the very young, and their families for food, health and housing are met. For this to occur, a fifth condition is essential: educational services must network with each other and with other social services in the same territory and elsewhere. The first stage of decentralization resulted in the breaking up or dismantling of an old system, but a system nonetheless. In this system were various types of functional relationships. In a second stage, the fragments lived in mutual ignorance or competed with each other. Now each educational service must reconnect with the rest. Horizontal relationships must be formed with health, early childhood and youth services, with sporting groups and social institutions, with businesses and political parties. This search for partners must not be limited by jurisdictional and political boundaries, but must take into account the needs and abilities of the cultural groups and the players involved. In some countries, certain regions have closer cultural ties with regions of neighboring countries than with other regions in the same country. Decentralization should make it easier for them and so many others to build functional networks that lead to shared solutions for common problems and thus save resources of all kinds. New communication and information technologies also open previously unheard-of opportunities to build such networks even over long distances. Decentralization seen as a deconcentration or delegation of responsibilities from one level of a political system to another does not necessarily make it easier to build networks between institutions to promote educational quality and prevent inequalities due to socio-cultural factors. Neither does it prevent the emergence of new inequalities arising from great imbalances in basic aspects of educational quality. Only decentralization seen as redistribution of power from the top of a hierarchical system to the base and from the centre of a government political apparatus, whatever the size of its territory, to society in both decision making and management can facilitate the establishment of such networks and complement them with connections to the NGO community. The five conditions mentioned above refer to three types of issues: first, a recentralization of certain functions, second, a different way of doing things, and third, more horizontal interconnections. The most complex and most controversial is recentralization. Some authors argue that recentralizing regulatory functions while decentralizing services is a contradiction or betrays a desire to maintain a conservative ideological order (10). No doubt that faced with poverty, poor education and increasing inequalities, the temptation exists to invoke clear frames of reference, effective tools and basic human needs of students (such as good nutrition) for producing policies and teaching practices intended to restore a paternalistic national government or promote an authoritarian recentralization of decision making and management of social practices for disseminating conservative values or those of specific groups. Therefore it is important to consider not only what is required to promote equity but also for whom, where and how it is to be done. Given this challenge, one naturally tends to give answers out of context. National governments must provide this or that, the provincial authorities must do something else, and non-government organizations or universities something else again. It is just such reasoning that was followed in the first stages of decentralization. An attempt was made to find a strategy that would be valid for all times and places, regardless of history or culture. For this, a major advance in educational policy making is possible and would provide greater opportunities for finding better alternatives to promote equity; that is, the answer to these questions must be found in each specific situation. In some federal states, the conditions will be right to develop curricula with the various provinces. Returning to Australia, it is possible to find interesting experiences of cooperation among teams from various Australian states to advance the development of compatible national standards, but in other countries, this is not possible and each province will have to set standards by itself and for itself, and then negotiate equivalence requirements, conditions for entry to the next educational level and other similar issues with the other members of the federation. In the United States, the development of standards, evaluation and production of curricula are concentrated in professional associations, businesses and universities. Such a system in Argentina or Brazil, absolving the national government of responsibility for such things, would only worsen the already gaping inequalities. In these countries and many in Asia and Africa, only the national government can bring together the professional teams and economic resources to ensure certain educational outcomes. This does not mean that they must monopolize and establish a centralizing exclusive statist ideology to achieve the five conditions mentioned. Hybrid models for making educational policy and teaching could be a good strategy (Braslavsky, 1999). In any case, this overview leads us to propose a change in how we seek alternatives for quality education with equity. As Michel Fiol, a recognized specialist in organizational management, said recently, we in the West tend to think in either/or terms. In practice, this means that we feel compelled to choose one of two options that we see as polar opposites and to exclude the other option from our frame of reference. Either we decentralize or we centralize; and if some aspects must be decentralized, it is considered an internal contradiction if others are centralized. But in reality this kind of thinking makes us confuse means and ends. Decentralization or centralization are instrumental options and should not be seen as ends in themselves. Equity, on the other hand, is a matter of ends. People have the same rights to be and educational services are there to help ensure those rights. FOOTNOTES 1. International Conference on Federalism, Mont Tremblant, Canada, 5–8 October 1999. 2. Prof. Cecilia Braslavsky, Ph.D. Scientific researcher with the National Council for Scientific and Technical Research (CONICET). Professor in the Department of Education Sciences of the University of Buenos Aires. Former Director General of Educational Research and Development in the Ministry of Culture and Education of Argentina. 3. See for example the recommendations of the commission on the future of education and the school of the future presented to the minister of North Rhineland–Westphalia (Bildungskommission NRW). Germany’s intellectual tradition explains why these recommendations make almost no reference to “decentralization” or “autonomy”. Nevertheless, analysis of the proposals concerning “self-administration” (pages 63, 64, 65, 167, 170, 191, 267, 294, 299) of educational institutions shows that these proposals involve decentralization and autonomy within a context in which the government has historically played a very different and highly valued role in regulating and providing educational services. For more on the German debate, see Munín, 1999. 4. A brief description and a preliminary critique of some parts of this tradition as a more general framework for public policy can be found in Mattos (1989). A more recent text in this tradition states that “participation is the primary basis of decentralization” (Fernández-Soria, 1996). 5. See for example the positions of Roland Glaser presented by Hermann Avenarius (1999). For the United States, see also the positions of Lewis J. Perleman (1992) and Gerstner et al. (1996). For Latin America, see for example the material produced by FIEL-CEA (1993) and Juan Llach (1997). This last one is particularly important since he is an influential and respected former minister of economic affairs of Argentina. 6. See various articles compiled by Miguel A. Pereyra et al. (1996). 7. For Latin America, see a greater development and a critique of these positions in Cosse (1999). 8. See for example a series of research studies and analyses reporting the growing inequalities during the period of transfer of education to the municipalities under the Chilean dictatorship and the subsequent narrowing of the gap through active national policies (Espínola, 1993; Brunner and Cox, 1995; Carnoy and McEwan, 1997). 9. The lack of policies concerning the professional staff leads to the statement that in many of the latest educational reforms in the southern countries, the teachers themselves are notably absent (Villegas de Reimers and Reimers, 1997). 10. See R. Jonathan (1997) and Geoff Whitty (1998), among many others. BIBLIOGRAPHY Avenarius, H. (1999), “La autoadministración escolar. Límites y posibilidades”, in Munin, H. (comp.), La autonomía de la escuela: ¿libertad y equidad?. Un recorrido por la discusión alemana de los años noventa, ed. Aique, Buenos Aires. Baudelot, Ch. and R. Establet (1989). Le niveau monte. Réfutation d’une vieille idée concernant la prétendue décadence de nos écoles, L’Épreuve des Faits/Seuil, Paris. Bildungskommission, NRW (1995). Schule der Zukunft. Zukunft der Schule, Bochum. Braslavsky, C. (1993). Autonomía y anomia en la educación pública argentina. FLACSO. Buenos Aires (Serie Documentos e Informes de Investigación No. 142). Braslavsky, C. (1993). “Una función para la escuela: formar sujetos activos en la construcción de su identidad y de la identidad nacional”, in Filmus, D. (comp.). Para qué sirve la escuela, ed. Tesis-Norma, Buenos Aires. Braslavsky, C. (1999). “Estudios de casos sobre la reforma de la gestión en América latina”, in UNESCO, La gestión: en busca del sujeto, Santiago, Chile. Braslavsky, C., N. Gluz and M. Calvo (1999). “Realidades y mitos de la autogestión: apuntes sobre su impacto en los aprendizajes”, Paper presented to the Third International Congress on Education Management, General San Martín University, Argentina. Caldwell, B. (1997). “Local management of schools and improved outcomes for students”. Paper presented at the Annual Meeting of the American Educational Research Association, Chicago, March 24–29. Carnoy, M. and P. McEwan (1997). “Public investments or private schools? A reconstruction of educational improvements in Chile”, Stanford University. Cosse, G. (1999). “La propuesta del voucher educativo: ¿una nueva panacea para América latina?”, in FLACSO, Propuesta Educativa, No. 20, Buenos Aires (at press). Cox, C. et al. (1995). “Calidad y equidad de la educación media en Chile: rezagos estructurales y criterios emergentes”, in CEPAL, Políticas sociales, No. 8, Santiago, Chile. Espínola, V. (1993). “La descentralización educativa en Chile: continuidad y cambio de un proceso de modernización”. Paper presented to the International Seminar on Deconcentration and Decentralization of Education, Asunción, Paraguay. Fernández-Soria, J. M. (1996). “Descentralización y participación de los padres y las madres en el sistema educativo”, in Pereyra M. A. (op. cit.). Gerstner, L. V. et al. (1996). Reinventando la educación, Ed. Paidós, Spain. Jonathan, R. (1997). Illusory Freedoms, Blackwell Publishers, USA. Llach, J. (1997). Otro siglo, otra Argentina, Ariel, Buenos Aires. Mattos, de, C. A. (1989). “La descentralización: ¿una nueva panacea para enfrentar el subdesarrollo regional?”, in Revista Paraguaya de Sociología, Vol. 26, No. 74, Asunción. Munín, H. (1999) (comp.). La autonomía de la escuela: ¿libertad y equidad?. Un recorrido por la discusión alemana de los años noventa, ed. Aique, Buenos Aires. Ortiz, R. (1998). Otro territorio, Convenio Andrés Bello, Bogotá. Pereyra, M.A. et al. (1996). Globalización y descentralización de los sistemas educativos. Fundamentos para un nuevo programa de la educación comparada, ed. Pomares-Corredor, Barcelona. Perrenoud, Ph. (1997). Construire des compétences dès l’école, ESF, Paris. Townsend, T. (1996). “The self-managing school: miracle or myth?”, mimeo. Townsend, T. (1997). Restructuring and Quality. Issues for Tomorrow’s Schools, Routledge, New York. Villegas Reimers, E. and F. Reimers (1996). “Where are 60 million teachers? The missing voice in Educational Reforms around the world”, in UNESCO, Perspectivas, Vol. XXVI, No. 99, Geneva. Whitty, G. (1998). “Controle do Currículo e Quase-Mercados: A recente reforma educacional na Inglaterra e no país de Gales”, in Warde, M. (comp.). Novas Políticas educacionais: críticas e perspectivas, Programa de Estudos Pós-Graduados em Educação: Historia e Filosofia da Educação, PUC-São Paulo. Forum of Federations / Forum des fédérations forum@forumfed.org