Gender and Education for All THE LEAP TO EQUALITY

Chapter 5 - From targets to reform: national strategies in action

Donwload this chapter in PDF

Decentralization: is it making a difference?

Meanings and motives Decentralization can make a big difference in the provision of social services. This is a message of the Human Development Report 2003 (UNDP, 2003b) in the context of defining policies that will contribute to the elimination of human poverty. But it requires good governance (Box 5.10), an argument that echoes the World Development Report 2000/2001 (World Bank, 2000): to benefit poor people [decentralization] must have adequate support and safeguards from the centre and effective mechanisms of participation. With or without good governance, one recent survey suggests that 80% of developing countries, plus some of the transitional economies of Eastern and Central Europe, are experimenting with some form of decentralization. In 1999, 96 out of 126 countries had at least one elected sub-national level of government, while 42 countries had two or more levels. In 1997, 52 countries had a measure of fiscal decentralization (Work, 2002). Decentralization for better education rests primarily on the assumption that the quality of education will be improved as a result of greater efficiency in the use of resources and better responsiveness to specific problems. A World Bank definition (Box 5.11) captures this intent for school systems and indicates the implications this process may have for more fundamental shifts in the meaning of education. As Table 5.7 suggests, the motives for introducing decentralization are many and varied. Some are fuelled by donor agencies as a means of promoting local democracy through the work of NGOs (e.g. Netherlands, 2002; Nach Mback, 2001). Within countries there may be a genuine wish to respond to demands for greater participation or to political pressures. For example, in recent years, devolution of authority in the Russian Federation, Papua New Guinea, the Philippines and the Sudan has been a response to the risk of secession (Bray and Mukundan, 2003). Efficiency motives may be influenced by a wish to lessen the financial burden on central government. A historical perspective indicates that decentralization is neither a new idea or process. McGinn (2001) recognizes a number of trends over the past half century. These include: - a move away from decentralization being conceived as a technical instrument to becoming an accepted component of ‘modernization’; - a concern for outcomes as well as inputs; - recognition of diversity – of place, community and need; - a more inclusive approach to stakeholders; - the need to broaden the revenue base; - a shift from local management to local governance. The trends have often involved policy reversals. Thus, in 1973, Bangladesh eliminated local management of schools, passed legislation to restore local control in the early 1980s, reverted to central control in 1990 and, most recently, is moving towards local management again (McGinn, 2001). Colombia demonstrates the inherently political nature of decentralization (Bray and Mukandan, 2003). In the mid–1980s, decentralization was perceived by the government to be a means of promoting stability and political legitimacy. From the early 1990s, resources were transferred from the centre to municipalities, and schools were given direct responsibility for managing personnel, designing aspects of the school curriculum and having some financial control. A greater voice was also given to parents, teachers and students. A voucher scheme for poor students was instituted. However, only 70% of schools had transferred to municipal control by 1993. The enterprise did not have total government support and the teachers’ unions, key to making the reforms work, feared the loss of national bargaining power.

Ways to decentralize In South Africa, mandatory school governing bodies assist principals, teachers and parents. Where, then, do the real opportunities lie for better education through decentralization and where is it working well for EFA? There is no comparable data on which to draw, although there are broad surveys of aspects of decentralization (e.g. Crook and Sturla Sverrisson, 1999) and a growing number of regional and national studies (e.g. Narodowski and Nores, 2002). The decentralization experience of seven countries is briefly reviewed below. Since 1992, Indian decentralization has ‘picked up steam’ (Mahal et al., 2000). Changes in the Indian Constitution (1992) made it incumbent on individual states to set up representative rural (panchayats) and urban bodies. This process was to be accompanied by the establishment of state finance commissions (that would recommend appropriate devolution of resources to the new bodies), and the creation of district planning committees. The exact responsibilities and the specification of the mechanisms to ensure accountability were left to state legislatures to determine. As a result, decentralization finds expression in different ways in different states (Govinda, 2003). In Madhya Pradesh gram panchayats (village-level authorities) take on responsibility for the construction and maintenance of schools, school inspection, the distribution of free textbooks and uniforms, the management of scholarships for Scheduled Castes and Tribal children and for non-formal education. One positive benefit has been the greater willingness of parents to send girls to school. However, in other states – for example, Maharashtra – these types of responsibility lie at the level of the zilla parishad or district level. The recent and varied experience of Kerala is analysed in Box 5.12. This highlights the complexity of the decentralization process where vested interests remain strong. With the planned enactment of the Free and Compulsory Education for Children Bill, 2003, allied to the Sarva Shiksha Abhiyan (SSA) programme for Universal Elementary Education (UEE), designed to attain UEE for 6–14-year-olds by 2010, decentralization should become an even more prominent part of India’s national effort. SSA is defined as a partnership between central, state and local government, involving Panchayati Raj institutions, school management committees, village and urban-slum level education committees, parent-teachers’ and mother-teacher associations, tribal autonomous councils and other grassroots structures in the management of schools (India, 2003b). Sarva Shiksha Abhiyan builds on a number of initiatives from the 1990s. One commentator recognizes at least eight national and state initiatives and programmes, including literacy missions, which have encompassed different approaches to decentralization (Raina, 2002). The debate around these initiatives centres in part on the extent to which the practice of decentralization is about delivery mechanisms rather than learning processes, and decentralized management rather than educational decentralization. Other concerns relate to the power of the national elite being replaced by the power of the local elite: ‘Indian central government plans for universalizing elementary education (grades 1–8) would come to nought unless collective voice and collective action becomes effective in all states…’ (Mehotra, 2001). In South Africa, decentralization and democratic participation has been a live issue for the past decade, particularly in relation to school autonomy. During the apartheid era, the democratic movement sought to develop parent-teacher-student associations (PTSAs). The South African Schools Act of 1996 drew on this tradition, establishing mandatory school governing bodies (SGBs). These bodies were designed to assist school managers and teachers, encourage parents to support their children’s education, and mobilize additional resources. The recent establishment of the National Association of School Governing Bodies offers some evidence of the significance of SGBs, although there is some way to go in building capacities to enable them to function strongly (Nzimande, 2002). But some of these developments have raised questions about the balance between the central and decentralized modes of governance, the regulatory role of the state in pursuit of national norms and educational equality, and strong community control (Sayed, 2003). In Jordan (Work, 2002), the ministry of education has delegated financial and administrative authority to local units, reorganized the ministry to be more responsive to local governments and allowed local decision-makers to promote participatory budget development. District governors advertise, recruit and hire civil service staff through personnel units. In Oman (Oman, 2001), local support councils have been established to contribute to the running of schools, representing the school itself, students and parents. Parents’ councils are empowered to make proposals on admission policies and student achievement. The United Republic of Tanzania’s Primary Education Development Plan 2002–2006 (United Republic of Tanzania, 2001b) states that ‘central ministries will continue to focus on policy development and monitoring. There will be increased delegated authority to local government and schools to manage education provision and development. The ultimate aim is for each district to be able to provide equitable access by children to education services’. The role of school committees is set out in Box 5.13. It includes both planning and financial accountability functions. This represents a very significant attempt to give much greater autonomy to local communities. It requires some investment of people’s time, and comparative experience suggests that unless village plans provide the basis for real action, long-term commitment cannot be assured. These issues are acknowledged in Tanzanian research (United Republic of Tanzania, 2001a) where, for example, it has been noted: ‘There is a significant relationship between village level social capital, of which trust is a major determinant, and parental participation in school-related activities. Although parents retain a degree of confidence in the school committee and the head teacher, there is widespread mistrust of local government as a service provider. This stems from a perception that school fees and other obligatory payments are neither accounted for nor turned into better education. As long as local governments remain under-resourced, it will be a major challenge to turn this situation around’ (Narayan, 1997). It is hard to gauge the relationship between decentralization and learning outcomes. Brazil has one the most decentralized fiscal systems of all developing countries (Crook and Sturla Sverrisson, 1999). It is also characterized by great diversity in its educational structures and systems that are managed by 26 states and 5,561 municipalities where the role of central government in primary and secondary education is mainly redistributive and supplementary (Guimaraes de Castro, 2002a). One expression of these characteristics is the Fund for Primary Education Development and Maintenance and Enhancement of the Teaching Profession (Fundef). Global revenue for education is allocated between states and municipalities to ensure that there is a minimum level of investment for each student. Funds are set aside for states that cannot meet this commitment. To follow up on the availability and use of funds at the community level, committees have been established. Each government (state or municipal) is required to present a monthly report showing how funds have been used. The new Brazilian government plans to strengthen these mechanisms (Lodi, 2003). The Lao People’s Democratic Republic has made significant progress in expanding educational opportunities after long periods of colonialism, conflict and economic crisis (Seel, 2003). Mass organizations such as the women’s and youth unions have traditionally played an important part in community-level development. However recent restructuring and plans for decentralization aim to bring programme implementation to district and village level, and planning and budgeting to the provinces, while the centre retains overall policy direction and monitoring responsibilities. There is an increasing focus on learning outcomes as distinct from educational inputs. School development is being discussed in terms of being responsive to learning needs, with attention to the monitoring of learning and of community participation in schools. The local recruitment of ethnic minority teachers is also being explored. 17. Based on material from the World Bank Global Education Reform website. www1.worldbank.org/education/globaleducationreform

Developing a balance sheet This set of examples points primarily to work in progress, but a number of trends can be detected. These include: - a new balance between the authority and responsibility vested in different levels of government – in federal as well as unitary states; - new responsibilities for schools, parents and communities; - greater devolution of financial authority and personnel management; - some evidence of local – school and district – planning. These are primarily shifts in the locus of management responsibility. There is much less evidence of the decentralization of the processes that define and monitor teaching and learning activities. There is even less evidence of decentralization developing as the result of local pressure rather than through centrally determined political decisions. It is also difficult to gauge the relationship between decentralization and learning outcomes, even though this is the real test. There are circumstances in which decentralization can increase disparities. For example, in Burkina Faso, Mali and Senegal, where increased levels of authority have been given to communities and NGOs, tensions have been evident between public and non-governmental schools, including different and uneven interpretations of what should be taught (Charlier and Pierrard, 2001). Other challenges associated with the effective implementation of decentralization are highlighted in Box 5.14.

Industrialized countries – towards greater flexibility? For education systems in industrialized countries, decentralization ‘has variously meant devolving power to the regions, the regional outposts of central government (deconcentration), the local authorities, the social partners and the institutions themselves’ (Green, 1999). As a result, a variety of governance models exist in practice. In some countries, for example France and Japan, most power lies at the centre. In Germany and Switzerland regional control is strongest, while the Nordic countries are known for their emphasis on local control. In the United Kingdom and the Netherlands substantial power has been devolved to schools themselves, as well as to the educational marketplace (Green, 1999). Notwithstanding these differences, industrialized countries share a common interest in innovation for the better governance of education. In 1996, OECD education ministers expressed concern at the ability of their education systems to adapt quickly. They called for more flexible frameworks (OECD, 1996). In 2001, they noted the increasing diversity of learners’ needs and the importance of maintaining cultural diversity while improving quality. Strengthening the connections between schools and their communities was seen as an important strategy (OECD, 2001b). The rapidly changing environment of schools is invariably mentioned as the main driver of governance innovation. Systems built on the need to prepare people for repetitive, obedient work and to shape national identity (Barber, 2003) are being replaced by education that is responsive to fast-changing technologies and markets, that not only call for different learning outcomes (problem-solving, networking, communication and learning skills, flexibility, mobility) but also for changes in the organization of the learning process (Carnoy and Castells, 1996). Schools need more autonomy in order to be more responsive to the changing needs of the workplace, the increasing heterogeneity of student populations and the growing complexity of the learning market (Halasz, 2003). Decentralization is not seen as merely the unconditional delegation of authority to the school level. Even in countries where school autonomy is greatest, there are mechanisms designed to link school improvement with systemic improvement, while rigorous procedures are put in place to ensure accountability. The former Swedish education minister, Ylva Johansson (2000), suggests that ‘experimentation should be fostered within the broad frameworks of national goals, with imaginative solutions devised for the real challenges being confronted on the ground. Evaluation and feedback are critical. Some ‘failures’ are inevitable and must be accepted in order to encourage risk-taking; valuable lessons can be learned from them as well as from the successes. These practices should not remain isolated examples, but be disseminated so that they can enjoy a much broader impact’. The United Kingdom, for example, applies the principle of intervention in inverse proportion to success. Initially, schools are granted a large amount of freedom and power. But there are national standards18. Schools that do well retain this amount of autonomy. Schools that do particularly well receive even greater autonomy, can be financially rewarded and will serve as examples for other schools. Those schools that perform poorly may loose autonomy and eventually risk direct intervention from central government, a ‘takeover’ by the private sector or closure. But schools are not left on their own. A well-defined framework is put in place for the early identification of poor performance by means of regular assessments and inspection. Inspection results are disclosed publicly, both to generate the pressure to improve and, in theory, to help schools to identify other schools from which they can learn. The creation of a National College for School Leadership is one of various measures to strengthen schools so that they are able to respond to this challenging framework (Barber, 2003). In Hungary, involvement of the private sector and the application of business models (benchmarking, quality management) are important in school improvement strategy. The rapid transition from central control to community-owned schools has highlighted the importance of school heads and teachers having the ability to fulfil their new roles. Schools are assisted by consultants from various sectors in defining their goals, in partnership with the local community. Schools are also supported in their development into learning organizations, implementing ‘Total Quality Management’ and disseminating the best of their experience throughout the education system (Halasz, 2003; Hirsch, 2003). The Netherlands introduced ‘freedom of education’ in 1917: parents and civil society organizations were given the right to establish schools according to preference and receive public funding if a minimum set of general conditions were met. This tradition has caused a deeply rooted belief that ‘it is not the government’s job to intervene at (school) level’ (Netherlands, 2003b). In this culture, the school itself is the primary actor in the system of quality assurance and accountability. ‘Self-evaluation’ by the school is the core instrument. The school inspector assumes the role of ‘critical friend’; on the one hand, he/she assesses the performance of a school with, as in the United Kingdom, the closure of the school as ultimate sanction in case of lasting under-performance. On the other hand, the inspector seeks to prevent this from happening by supporting the school in its process of improvement. As in Hungary and the United Kingdom, publicly disclosed performance data play an important role (Netherlands, 2001). Networks of schools and teachers are important vehicles for the exchange of good practice and for professional development. In Portugal, the proliferation of such networks is one of the core strategies in the government’s policy of decentralization. In order to ensure the dissemination of successful innovations and a professional consultation among schools and among teachers, Portugal has initiated the ‘Good Hope Programme’19: a government-funded scheme to establish links between schools (Céu Roldao, 2003). More classic, centrally-driven reform policies may be insufficiently informed by the body of knowledge about school effectiveness (Hargreaves et al., 1998), and therefore inadequately tailored to local contingencies. As a result, they have generally failed to improve student achievement (Hopkins and Levin, 2000). Networks may offer a more influential way of exchanging knowledge about what works for schools. This is important in highly decentralized systems, where channels for top-down dissemination of knowledge may be absent in the first place. Networks may offer a better way of exchanging knowledge about what works for schools. 18. www.standards.dfes.gov.uk 19. Apart from the Portugese Programa Boa Esperança, Boas Práticas (www.iie.min-edu.pt/proj/boa-esperanca/index.htm), there are many other schools networks, e.g. Improving the Quality of Education for All (IQEA) project (www.nottingham.ac.uk/education); Bertelsmann Foundation’s Network of Innovative Schools, Germany (www.inis.stiftung.bertelsmann.de/set.htm); Accelerated Schools Project, United States (www.acceleratedschools. net/); Network of Agenda 21 Schools, Germany (nibis.ni.schule.de/agenda/projekt.htm); Network of Core Knowledge Schools, United States (www.coreknowledge.org). Networks with individuals (teachers, teacher trainers) as members include: European Observatory on Innovation in Education and Training (www.inrp.fr); International Network of Innovative School Systems, Germany; The Learning Consortium, Ontario, Canada (fcis.oise.utoronto.ca/~learning).