Reforming Primary and Secondary Education in Latin America and the Caribbean An IDB Strategy Inter-American Development Bank Washington, D.C. Sustainable Development Department Sector Strategy and Policy Papers Series Cataloging-in-Publication data provided by the Inter-American Development Bank Felipe Herrera Library Reforming primary and secondary education in Latin America and the Caribbean : an IDB strategy / [written by … Claudio de Moura Castro … [et al.]]. p. cm. (Sustainable Development Dept. Sector strategy and policy papers series ; EDU-113) “This strategy (GN-2067-3) was favorably considered by the Bank’s Board of Executive Directors on March 1, 2000”— t.p. verso. 1. Educational change--Latin America--Economic aspects. 2. Educational change-- Caribbean Area--Economic aspects. 3. Education, Primary--Latin America. 4. Educa- tion, Secondary--Latin America. 5. Education, Primary--Caribbean Area. 6. Educa- tion, Secondary--Caribbean Area 7. Education and state--Latin America--Economic as- pects. 8. Education and state--Caribbean Area--Economic aspects. 9. Inter-American Development Bank. I. Castro, Claudio de Moura. II. Inter-American Development Bank. Sustainable Development Dept. Education Unit. III. Series. 371 R457--dc21 This report was written by the following staff of the Education Unit, Sustainable Develop- ment Department: Claudio de Moura Castro, Senior Education Advisor; Juan Carlos Na- varro, Education Specialist; Laurence Wolff, Education Consultant; as well as Martin Car- noy, Professor of Education at Stanford University. The authors would like to thank Marcelo Cabrol (Region 2), and Ruthanne Deutsch (Economist, Social Development Divi- sion) and Aimee Verdisco (Education Consultant) for their assistance. This strategy (GN-2067-3) was favorably considered by the Bank's Board of Executive Di- rectors on March 1, 2000. May 2000 This publication (No. EDU-113) can be obtained through: Publications, Education Unit Inter-American Development Bank 1300 New York Avenue, N.W. Washington, D.C. 20577 E-mail: sds/edu@iadb.org Fax: 202-623-1558 Web site: www.iadb.org/sds/edu Foreword Children in Latin America and the Caribbean attending primary and secondary schools, especially those in rural areas and in urban slums, achieve levels of school completion, knowledge and skills that are inade- quate to compete in a world where mathematics, language, and communication skills as well as flexibility, creativity, and the capacity to work cooperatively are increasingly fundamental for economic growth and social development. Within this context, this paper summarizes experience to date and provides a set of objectives and strate- gies to guide the IDB in its support of primary and secondary education over the next decade. The paper defines five critical areas of reform— making teachers partners in reform, providing more and better text- books and teaching materials, selectively supporting the use of technology in education, decentralizing edu- cational management and increasing accountability of key stakeholders, and strengthening early childhood education and development. It also summarizes the new challenges facing secondary education in the re- gion— to meet increasing quantitative and qualitative needs while at the same time serving a more diverse clientele. This report sets out the information and management tools needed to meet long-term goals and confirms the fundamental importance of engaging in a long-term and consensual reform process. To prepare this strategy, the authors distilled the experience of many years of educational reform efforts in the region and elsewhere. As a result, it is hoped that this paper will also prove useful to policymakers and stakeholders throughout the region who are committed to educational reform and development. Antonio Vives Acting Manager Sustainable Development Department Contents Executive Summary 1 The Need for Education Reform 5 Five Critical Areas of Reform to Improve Quality and Increase Equity 14 The Challenge of Secondary Education 22 Navigating and Managing the System: Tools to Achieve Reform Goals 27 The Art of Implementing Education Reform 35 Looking to the Future: Developing an IDB Action Plan 39 References 45 Annexes 51 Blank page Executive Summary This paper summarizes what is known about how tralizing educational decision-making, encourag- to improve primary and secondary education in ing school autonomy and increasing parental and Latin America and the Caribbean and provides a community participation. There is evidence that set of objectives and strategies to guide the IDB these reforms, especially those leading to school in its support of primary and secondary educa- autonomy, have had a positive impact. Based on tion over the next decade. It is based on six back- experience to date, success requires consensus ground papers prepared for this purpose, as well among the key stakeholders, stable or increasing as a review of the lessons that stem from recent educational spending, local capacity building and regional and country experience. a new set of roles for ministries of education at the central or state government levels. These Inadequate education is a critical impediment to goals promote equity, provide technical assis- long-term economic growth in the region. In spite tance, monitor progress and provide incentives of substantial gains, the percentage of children for improved performance. who fail to complete primary schooling is almost twice that which would be expected on the basis Only a few countries to date have turned their of the region's income level. Children across the attention systematically to teacher education is- region consistently score lower on international sues and there is inadequate documentation of tests than do children in developing countries successful experiences. Nonetheless, there is con- with similar per capita incomes. Large differ- sensus that the reform of teacher education in the ences in educational achievement between the region should emphasize classroom management poorest and richest segments of society exacer- skills and an early involvement in classroom bate the income inequality endemic to the region. practice. In addition, the subject content taught to prospective teachers should be increased. Better CRITICAL ACTIONS TO working conditions, new performance incentives IMPROVE PRIMARY and increased time devoted to real classroom A ND SECONDARY EDUCATION learning are also essential to improve learning. Lessons Learned in Key Areas of Action Investing in distance learning by radio and televi- sion has already had a major positive impact on There is now an increased awareness of the need quality, equity and access in several countries in to improve quality and equity in primary and the region. On the other hand, in spite of their secondary education. A wide variety of reforms promise, the large-scale cost effectiveness of and innovations in education, many of them sys- computers and related Internet technologies on temic, many supported by the IDB, are under primary and secondary education has not been way in areas as diverse as decentralization and proven. For these newer technologies, it is best the use of instructional technology, and much has been learned. to invest in a wide variety of pilot programs, in preparation for the time when the cost of infor- For example, throughout the 1980s and 1990s mation technology falls and teachers are more many Latin American countries sought to reduce knowledgeable. At the same time, there is strong excessive centralization in education by decen- evidence that the countries of the region cannot 1 neglect the provision and use of conventional improvement in quality will result in increased learning materials such as books, educational student promotion and achievement; reduced materials and libraries. repetition rates could then result in lower overall costs. Preschool education has a high pay-off in terms of helping low-income children to be more suc- Using Information for Policy Decisions cessful, remain in primary school and, ultimately, do better in the labor market. All governments in Many countries in the region are already making the region are investing in preschooling in one major efforts to improve statistical systems, an way or another. Experience has shown that the essential component of better project design and most cost-effective use of government funds is to improved decision-making and policy dialogue. target efforts to the neediest and to rely on and Most countries have initiated educational as- strengthen non-formal providers. sessment systems. To ensure that assessments truly become a tool for improving learning it will The Challenge of Secondary Education be important to also foster institutional develop- ment, training and appropriate feedback to Many countries in the region, especially the more stakeholders. To strengthen applied research and advanced ones, have begun to pay attention to identify what really works in education, there is a secondary education. Necessary reforms, which need for institutional development, the establish- are already under way in several countries, in- ment of a consensus among all stakeholders on clude the separation of vocational preparation the value and importance of research and the de- from formal secondary education; pushing most velopment of an agenda for research, which will technical education to the post-secondary level; vary from country to country. and, overall, making theory more applied, more concrete and more focused on solving problems. Implementing Educational Reforms There is a consensus that the region’s current enrollment ratio of 55 percent of the cohort will Experience has shown that the critical attributes need to be increased significantly, especially to ensure the success of reforms include leader- through targeting access to the poor. Increased ship, consensus and continuity; adequate social learning goals should include not only improved marketing; feedback on achievement of means higher order skills in traditional academic sub- and goals permitting mid-course corrections; in- jects, but also elements such as increased civic centives to encourage actors and stakeholders to responsibility and an understanding of the role of behave in ways that strengthen the reform proc- technology in society. ess; and training to ensure that key actors have the capacity to implement reforms. A specific Educational Financing and Efficiency area in need of renewed attempts at policy re- search and dialogue is the role of teachers’ un- The countries of Latin America and the Carib- ions, a key player in the dynamics of educational bean have a window of financial opportunity for reform. investment over the next twenty years, as the demographic transition results in smaller in- While all countries in the region need to focus on creases in the number of school-age children and quality and equity, each country will need to pur- raises the percentage of the population in the la- sue its own set of solutions based on its economic bor force. With the right policies, the goals of resources and current level of educational devel- increased quality, quantity and equity need not be opment. In a few lower-income countries, and in contradictory but rather can be self-reinforcing. the poorer regions of others, increasing initial For example, system-wide improvements in edu- enrollment in basic education is still an issue. cational quality, particularly but not exclusively Increases in secondary enrollment in each coun- at the primary level, will overwhelmingly benefit try will depend on current enrollment ratios, so- students from poor families, since any significant cioeconomic conditions and perspectives, and the 2 rigor with which countries identify cost-effective but intense application of the potential of tech- solutions to meet targets. nology to expand coverage and improve quality, especially through the mass media (radio and THE ROLE OF THE IDB television), as well as through pilot programs in AS A LEADER IN THE REGION the use of computers and the Internet. The IDB will especially focus on ensuring sustainable fi- Lending Priorities nancing of the most cost-effective mixes of in- puts, processes and incentives to raise student Since 1965, the IDB has committed, on average, achievement and retention in school; and on in- close to 5 percent of its yearly lending to the edu- creased equity through targeted interventions, cation sector. Lending to education increased including preschooling. The IDB will also act to significantly, to 9.5 percent in the period 1994- strengthen assessment systems, statistics, applied 97, as a result of the new development strategy research, feedback to stakeholders, and social as articulated in the IDB's Report on the Eighth marketing of reform objectives. General Increase of Resources of 1994, which provided a specific mandate for education. While the Bank is pleased to respond to increas- ing requests for support of secondary education, The main objective of the IDB's support for pri- it regards primary education as a long-standing mary and secondary education will be to improve commitment, not a transient preoccupation. The quality and equity. The Bank will also support IDB will, therefore, proactively review progress increased access to schooling where necessary, in primary education and seek follow-on loans to especially at the secondary level. Because of the those now under implementation. The Bank is quantitative needs, lending in secondary educa- also prepared to fund the capital costs associated tion is expected to be high (in terms of amounts, with expanding access to secondary education. if not necessarily in the number of operations). Support for secondary education will always be To achieve these goals, the IDB will examine the accompanied by attention to designing secondary reform issues described in this paper. Nonethe- school models which are relevant for the 21st less, the locus of reform has to be inside the century, as well as to quality improvement, in- country. The Bank will endeavor to understand creased equity and efficiency, sustainable fi- the moment of the country, its institutions and its nancing and better school management. Given the needs, as well as the variety of economic and difficulties of appropriating more public funds to educational conditions within the region. education and considering that private education often brings variety and more cost-efficiency, the In its lending, the IDB will focus on the key ac- IDB will consider funding some private education tions to improve quality and equity that are de- costs, provided equity considerations are met. scribed above, taking into consideration each country's specific situation. This will include Processes reviving the school as an active sphere of man- agement, innovation and social responsibility In all of its efforts in primary and secondary edu- through increased autonomy, intense community cation, the IDB will seek to ensure the quality, participation and, wherever relevant, decentrali- implementability and sustainability of its proj- zation and local government involvement. It will ects. In particular, it will encourage client, bene- also include increased support of pre- and in- ficiary and stakeholder ownership; undertake service training of teachers with a focus on inno- sectoral and economic analysis to learn about the vative programs and the extension of those pro- issues and options of education reform and their grams judged cost-effective, establishing incen- relationships to the larger socioeconomic context; tive programs to attract higher quality teachers analyze institutional capacity and sustainability; and analytical work and programs to improve and devise simple but robust monitoring and teacher incentives and accountability at the evaluation instruments. In its project design the school level. The IDB will also support a prudent IDB will help countries test out new ideas on a 3 small-scale (pilot project) basis; use a process the region to share their experiences; civic edu- rather than a blueprint approach for education cation; and how public support to the private reform projects; and utilize a wider range of sector can extend the reach of public funds, in- lending instruments. The IDB will also reconsider crease competition and ensure equity. the set of financial instruments that can be de- ployed to support education, including the devel- Collaboration opment of a simplified procedure for financing pilot projects, research and participation in inter- The IDB will collaborate with other key players national programs, and financing private educa- in the field, such as the World Bank, ECLAC, tion through support for student loan schemes, UNICEF, the Caribbean Development Bank and lines of credit for private school construction and USAID, by identifying common issues and a re- voucher systems. search agenda, sharing information, jointly sup- porting conferences and analytical work, and Emerging Issues and Dissemination of Best capitalizing on the relative strengths of each of Practices these organizations. The IDB will also make spe- cial efforts to link with and begin a dialogue with The IDB is uniquely positioned to play a role in NGOs working in education, country-level think collecting, systematizing and disseminating good tanks and advocacy groups, and the growing practices in the field and in identifying emerging number of private institutions in education and in issues. Jointly, as well as on its own, the IDB publishing. will support on-site and virtual applied research, conferences, study tours and training programs. Looking Towards the Future The IDB has already identified a number of cru- cial emerging issues in primary and secondary In response to the interest and concerns expressed education where applied research and pilot proj- by participants in the 1998 Santiago Summit, ects are needed. These include instructional tech- Bank management announced a renewed com- nology, where the Bank is helping countries to mitment to education lending and anticipated an undertake pilot and development projects to meet expansion of its activities in the sector. In pursuit long-term educational needs; teacher education, of this goal, the Bank will undertake specific working conditions, teacher careers, in-service country by country programming discussions on training and classroom pedagogy, where in con- increased investment in education and, in col- junction with others the IDB has undertaken a laboration with experts across the region, deepen series of studies; educational statistics and as- its understanding of what works best in education sessment, where the IDB expects to continue its and under what circumstances. support for programs enabling countries across 4 The Need for Education Reform EDUCATION AS A CATALYST At the outset it must be said that the world econ- FOR DEVELOPMENT omy has gone through enormous changes in the past 20 years, and these have made the quality of This paper summarizes what is known about how education systems far more important to the eco- to improve educational quality, efficiency and nomic health of nations. Global competition and equity in primary and secondary education in the information and telecommunications revolu- Latin America and the Caribbean. It also exam- tions are at the core of these changes. Economic ines issues related to achieving the goals of in- growth under the new conditions requires na- creased quality and equity, summarizes the tional infrastructures that include, among other changes to date and the priorities for the future, elements, a labor force with a solid base of read- and provides a set of objectives and strategies to ing skills, mathematical and scientific knowledge, guide the IDB in its support of primary and sec- problem-solving capabilities and even computer ondary education over the next decade. It is skills, as well as the capacity to communicate based on six background papers commissioned effectively. Political and economic modernization for this study, as well as a wide review of recent in Latin America and the Caribbean has given regional and country analyses, policy statements schools another major responsibility: the devel- and research on primary and secondary educa- opment of a civic culture that stresses tolerance, tion. cooperation, and a broader sense of nation and community. The IDB has been lending to education since 1965. During the 1980s and 1990s the Bank in- The weaknesses of education in the region are creased its focus on lending to primary and sec- well known. Young workers enter the labor force ondary education. Over the same time the IDB’s with fewer years of education, on average, than lending approach has evolved from supporting workers in Southeast Asia and other developing expansion and coverage to concerns for quality regions and the gap is widening (Birdsall, 1998; improvement; from an almost exclusive focus on Filmer and Pritchett, 1998, and Graph 1). In construction and infrastructure to an explicit spite of substantial gains in the reduction of illit- concern for management and pedagogy; from eracy, in gender equality and in access to primary centralized control and monitoring to school- education, the percentage of children who fail to centered, community-based, decentralized ap- complete primary schooling is almost twice that proaches; and from project financing to a concern which would be expected given the region’s in- with system-wide financial sustainability. The come level (IDB, 1996). By the early 1990s, IDB’s involvement in the education sector has workers in Latin America had an average of 5.2 also evolved from top-down methods of program years of education, compared to 7 years for preparation and implementation toward more countries at similar levels of economic develop- participatory approaches to service delivery. ment. Simulations suggest that if the region's educational achievement were equiva- 5 Graphic 1 Years of Education of the Labor Force 9 8 Asia 7 World Average 6 5 Latin America 4 1970 1975 1980 1985 1990 1995 Source: Londoño and Székely, 1997. lent to that of countries in the Far East, per cap- 1990 the gap in years of education between the ita income could be increased annually by 0.5 richest 10 percent of 21-year-olds and their percent. counterparts in the poorest 30 percent was 4 or more in 9 out of 15 countries included in a recent Learning in schools, by all accounts, remains IDB analysis, and for at least 4 countries had strikingly deficient in all but a few elite institu- actually increased when compared to what it was tions. Disappointing achievements in primary in 1980 (IDB, 1998, and Graph 2). school remain the number one educational prob- lem in most Latin American countries. Latin There is widespread support for the idea that a American students included in international test- good education not only includes good subject ing research projects consistently score at the learning but also the encouragement of trustful bottom of developing countries (see Tables 1 and social relationships and an increased awareness 2). The region’s educational institutions are of the basic rules of citizenship in a modern so- among the worst rated by international leaders ciety, the practice of which should start at the and investors (World Economic Forum, 1997). school level. Yet, little is known or energetically Close to 80 percent of low-income students in carried out to inculcate these values in the region. Latin America are not able to understand written messages after six years of schooling, even if After more than a decade of market-oriented eco- they have passed six grades (Schiefelbein, 1996). nomic reforms, the region is undertaking a sec- ond, and perhaps more difficult set of social and The region exhibits the largest income inequal- institutional reforms. Among the most important ity in the world, which at the same time reflects are those related to education. In fact, education and perpetuates disparities in educational op- may well be the region's most critical social issue portunities for different population groups. In over the next decade and its most important 6 Number of Years catalyst for development. The critical need for tives and financing patterns rather than just in- primary and secondary education in the region is creasing resources for institutions or practices "systemic" reform; changing structures, incen- already in place. Table 1 Achievement Scores of Latin America and the Caribbean The IDEA Study of Reading Literacy, 1989 Latin American and Standard Caribbean Countries Mean Deviation Trinidad and Tobago 451 79 Venezuela 383 74 Other Countries Finland 569 70 United States 547 74 Hong Kong 517 71 Singapore 515 72 Spain 504 78 West Germany 503 84 Indonesia 394 59 Source: W.B. Elley, 1992. The International Assessment of Educational Progress Test of Mathematics and Science, 1991 Standard Country Mean Deviation Mathematics Test Korea 73 0.6 Taiwan 73 0.7 Spain 55 0.8 United States 55 1.0 Portugal 48 0.8 São Paulo, Brazil 37 0.8 Fortaleza, Brazil 32 0.6 Maputo and Beira, Mozambique 28 0.3 Science Test Korea 78 0.5 Taiwan 76 0.4 Spain 68 0.6 United States 67 1.0 Portugal 63 0.8 São Paulo, Brazil 53 0.6 Fortaleza, Brazil 46 0.6 Source: Educational Testing Service, 1992. 7 Table 2 Mathematics Achievement in the Third International Mathematics and Science Study Percentage of Students Achieving International Marker Levels, 8th Grade Math Achievement Top 10% Top Half Achievement Differences Country 8th Grade Level Level between 7th and 8th Grade Singapore 643 45 94 42 Korea 607 34 82 30 Japan 605 32 83 34 Hong Kong 588 27 80 24 Belgium (Fl) 565 17 73 8 Czech Republic 564 18 70 40 Slovak Republic 547 12 64 39 Switzerland 545 11 65 40 Austria 539 11 61 30 France 538 7 63 46 Hungary 537 11 60 35 Slovenia 541 11 61 43 Russian Federation 535 10 60 35 Netherlands 541 10 63 25 Bulgaria 540 6 58 26 Canada 527 7 58 33 Ireland 527 9 57 28 Belgium (Fr) 526 6 58 19 Austria 530 11 57 32 Thailand 522 7 54 28 Israel 522 6 56 -- Sweden 519 5 53 41 Germany 509 6 49 25 New Zealand 508 6 48 36 England 506 7 48 30 Norway 503 4 46 43 United States 500 5 45 24 Denmark 502 4 47 37 Scotland 498 5 44 36 Latvia (LSS) 493 3 40 32 Spain 487 2 36 39 Iceland 487 1 37 27 Greece 484 3 37 44 Romania 482 3 36 27 Lithuania 477 1 34 49 Cyprus 474 2 34 28 Portugal 454 0 19 31 Iran, Islamic Republic 428 0 9 27 Kuwait 392 0 3 -- Colombia 385 0 4 16 South Africa 384 0 2 7 Source: Mathematics Achievement in the Middle School Years: IEA's Third International Mathematics and Sci- ence Study (TIMSS) (Chestnut Hill, Mass.: Boston College / TIMSS International Study Center, 1996). 8 Graph 2 Education Attainment of the Richest and Poorest 21 Year Olds in the 1990s 16 (average years of education attained) richest 10% poorest 30% 14 12 10 8 6 4 2 0 Source: Calculated using data from Birdsall, Berhman, Szekely (1998) Education Gaps Between the Richest and Poorest 21 Year Olds in the 1980s and 1990s 8 (average years of education attained) 1980s 1990s 7 6 5 4 3 2 1 0 -1 Note: Gaps are calculated by subtracting the average years of education attained by the poorest 30% from the average attained by the richest 10% . Source: Calculated using data from Birdsall, Berhman, Szekely (1998). 9 An awareness of these accumulated shortcom- must be improved (see Annex Box 1 for a further ings, together with an understanding that educa- discussion and definition of quality). tion is a key element in fighting poverty and er- ducing income inequality and also serves as the All available international comparisons show natural incubator and safeguard of contemporary severe problems of quality in education across citizenship and democratic values, has resulted in the region. The Third International Mathematics the mobilization of constituencies that include and Science Test (TIMSS) of the International parents, students, teachers, governmental Association of Educational Assessment (IEA) authorities, businesses, nongovernmental organi- placed a strong emphasis on higher order skills. zations and the media. These constituencies are Colombia, the only Latin American country vocally demanding more resources, new policies which reported the results of the test given to and better decisions for the education sector; er- eighth graders, scored 41st out of 42 countries sources, decisions and policies good enough to (see Table 2, Mathematics achievement in the turn education systems into institutions capable TIMSS). At least ten of the participating coun- of addressing the economic, social and cultural tries had per capita incomes equal to or lower challenges of the global economy. Education is than that of Colombia. Venezuela and Trinidad not only becoming a real priority in the budgets, and Tobago participated in the 1992 IEA reading but also in the minds, words and actions of po- survey of eighth graders. Venezuela scored lower litical and business leaders ando pinionmakers. than all countries except Nigeria, Zimbabwe, and Botswana. Trinidad and Tobago scored above Fortunately, the region is now characterized by a those countries as well as above Thailand and wide variety of experiments and innovations in Philippines. Brazil participated in an interna- education, some of them systemic, many of them tional mathematics test in 1992 and scored below just getting started, and many already enjoying all participating countries except Angola (see support. There is evidence that public investment Wolff, 1998, for a summary of international in education is increasing from the low levels of tests). The region is also characterized by the the 1980s. This paper incorporates the lessons highest primary school repetition rates in the learned from the reform experience to date in world, averaging more than 40 percent in first areas as disparate as decentralization, educa- grade in some countries. tional assessments, educational technology, and financing. The problem in most countries does not appear to be official curriculum content, especially at the THE TWO OBJECTIVES OF REFORM: primary level, which in many cases is reasonably QUALITY AND EQUITY up to date, but rather a failure to achieve cur- riculum goals because of inadequate teaching, Reforms need to focus directly on improving the too little time on task, lack of teaching materials, quality of education. Quality has many defini- and inadequate management and incentives. tions. The most important definition is "output- and learning-based;" that is, the extent to which The second key goal for education reform is that children attain the knowledge and skills which of equity. Throughout the region there is a wide society wishes to impart to them. These are not discrepancy in educational attainment by social only academic skills. For example, one re- class. In one country, for example, 60 percent of searcher (Gardner, 1993) has identified seven upper-income students aged 15 to 19 have knowledge dimensions: linguistic, logical- reached ninth grade or higher, compared to only mathematical, bodily-kinesthetic, spatial, musi- 20 percent of children in the poorest economic cal, interpersonal and intrapersonal. In principle group. In another country, about 25 percent of all, of these dimensions have a role in ensuring the higher-income age group has achieved ninth social and economic development, and ares elf- grade or higher compared to about 5 percent of reinforcing. To meet these "output" objectives, the lowest-income age group. Despite almost the quality of school "inputs" and "processes" universal access to primary education and rapid 10 expansion of secondary schooling, Latin America plicit equity enhancing programs are needed, still spends about 20 percent of total public re- ranging from school construction in urban slums sources for education on the 5 percent of the total and rural areas to incentives for teachers and student population attending universities. subsidies for students. In the case of educational services provided to indigenous populations, spe- Because public primary and secondary schools cial consideration is required regarding the are of generally poor quality, pupils from low strengthening of their cultural identity, traditions socioeconomic backgrounds are less likely to and language. This has to be achieved in such a succeed academically, and are greatly under- way that it becomes supportive and mutually re- represented at higher levels of schooling. Indieg- inforcing vis à vis educational objectives, allow- nous peoples and students in rural areas and ur- ing for an adaptation of the educational process ban slums are particularly disadvantaged. Ex- to the special needs of these populations. Box 1: Gender and Education Contrary to a number of other regions in the developing world, education in Latin America and the Caribbean is essentially equitable in terms of gender. But a few exceptions have to be noted.F irst, while coverage at the pri- mary level is close to universal and gender-blind, in the poorest countries, especially those with large populations of indigenous peoples (e.g,. Guatemala, Bolivia), coverage falls short and girls' enrollment continues to lag.S ec- ond, at secondary and higher education, overall enrollment rates for women in the regioenx ceed those for men, at the secondary level by more than four points anda t the tertiary level by more than two points. Rough regional av- erages put women's enrollment at the secondary level at 58 percent and at the tertiary level at 19 percent; for men, these figures are 54 percent and 17 percent, respectively. Differences are even greater in someC aribbean countries where gaps exceed 13 points at the secondary level (e.g., Dominican Republic and Trinidada nd Tobago) and seven points at higher levels of schooling (e.g., Bahamas). Third, girls come out on top in other areas as well. In many countries across the region, dropout and repetition rates for girls fall below those for boys, and female promotion rates are higher. The progress toward equality appears to be notable and fast. The lower schooling levels of women still lingering among older generations show how gender-biased schools were in the recent past. The situation faced by these generations differs significantly from that faced by younger generations: whereas illiteracy tends to be concentrated among older women, the level of education attained by their daughters and granddaughters surpasses that of their sons and grandsons. Such progress, however, has not been uniform. Increases in women’s education take place at different speeds. Improvements in urban areas precede those in rural areas and indigenous women are often last to benefit from better and more schooling. The linkage between education and work is far from linear. Education undoubtedly improves individual producvt-i ity and welfare. Yet, in and of itself, it cannot overcome many effectCs and perhaps, distortionsC in the labor mar- ket. Although women's enrollment in higher education surpasses men's, once in the labor market, women earn less and face higher rates of unemployment. This situation reflects, in part, thec areers women tend to choose. Many enter the teaching or other "pink collar" tracksC professions which, although often requiring university education, command comparatively lower salaries in the labor market. Such "tracking" arises from a variety of factors, ni- cluding the desire to have jobs that accommodate other interests as well as the use of stereotyped messageast ear- lier levels of educationd epicting women in family settings and men as professionals. For the Bank, several implications follow. Gender equity should be promoted in all aspects of education policy. In some cases, this may mean promoting girls' education; in others, it may mean taking action to promote boy's edu- cation. The Bank will also continue its efforts to bring more women into the technical and scientific fields. 11 With the right policies there is no real trade-off ratios, providing the region with a "window of between quality, quantity and equity. In the first opportunity" during the next 20 years (see Graph place, as has been shown (Schiefelbein et al., 3 and Birdsall, 1998) to find adequate funding to 1998), the region spends over $4 billion per year achieve its goals of quality and equity as well as on repeaters, most of whom are from low socoi - to increase enrollment ratios in secondary eduac- economic backgrounds. As a result, system-wide tion (see Table 3 for a summary of enrollment improvement in the quality of education, par- ratios in primary and secondary education). ticularly but not exclusively at the primary level, will overwhelmingly benefit students from poor The rest of this paper enters into greater detail on families, since any significant improvement in these issues and on policies and strategies needed quality will result in noticeable advances in sut- for the next decade. The chapter that follows ex- dent progress and achievement for precisely those amines five critical elements contributing to the students. Quantitative expansion, in turn, will quality of education and to equity— teachers, only be achieved by making ever higher levels of teaching materials, instructional technology, quality instruction accessible to low-income chli- management, and preschooling. Because of the dren, the same children that today leave school increasing importance and a lack, up to now, of a too soon because it has little to offer in all er- systematic analysis of secondary education in the spects. Concrete instances of this general princi- region, a separate chapter examines issues spe- ple abound: very small investments in textbooks cific to this level. Subsequent sections examine have been shown to more than pay for themselves how education financing and information can be in increased learning and school flows H( arbison utilized to achieve education reform goals. A and Hanushek, 1992); and intensive accelerated section then discusses issues related to impel- courses for repeaters have been shown to more menting education reform. The paper ends with a than pay for themselves in reduced repetition summary of the IDB's strategy in light of the re- (Oliveira, 1998). Finally, recent declines in the gion's needs and the Bank's own capabilities, ex- birth rate imply a leveling off in the number of perience and mandate. school-age children and declines in dependency Graphic 3 A Window of Opportunity: Declining Dependency Rates 1.6 Dependency Rates 1.5 1.4 1.3 WINDOW OF OPPORTUNITY 1.2 1.1 1.0 0.9 0.8 0.7 1950 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 12 Table 3 Enrollment Rates in Latin America and the Caribbean Gross Enrollment Ratio Net Enrollment Ratio Preprimary % Primary % of Secondary % of Tertiary % of Primary % of Secondary % of Country of Relevant Relevant Age Relevant Age Group Relevant Age Relevant Age Relevant Age Group Age Group Group Group Group 1995 1980 1995 1980 1995 1980 1995 1980 1995 1980 1995 Argentina 50 106 108 56 72 22 38 -- -- -- 59 Bolivia -- 87 -- 37 -- 16 -- 79 91 16* 29* Brazil‡ 56 98 112 33 45 11 11 80 90 14* 19* Chile 96 109 99 53 69 12 28 -- 86 -- 55 Colombia 28 124 114 41 67 9 17 -- 85 -- 50 Costa Rica 70 105 107 48 50 21 32 89 92 39 43 Cuba 89 106 105 81 80 17 14 95 99 -- 59 Dom. Rep. 20 118 103 42 41 -- -- 73 81 -- 22 Ecuador 49 117 109 53 50 35 20 88 92 -- -- El Salvador 31 74 88 25* 32* 13 18 -- 79 -- 21 Guatemala 32 71 84 18 25 8 8 58 -- 13 -- Haiti† -- 76 -- 14 22 1 -- 47 26 3 -- Honduras 14 98 111 30 32 8 10 78 90 -- 21 Jamaica 81 103 109 67 66 7 6 96 100 64 64 Mexico 71 120 115 49 58 14 14 -- 100 -- 46 Nicaragua 20 98 110 43 47 13 9 98 83 23 27 Panama 76 106 106 61 68 21 30 89 91 46 51 Paraguay 38 106 109 27 38 9 10 89 89 -- 33 Peru 36 114 123 59 70 17 31 86 91 -- 53 Trin. & Tob. 10 99 96 70 72 4 8 90 88 64 64 Uruguay 33 107 11 62 82 17 27 -- 95 -- -- Venezuela 43 93 94 21 35 21 29 82 88 14* 20* Notes: Italicized numbers indicate 1992 statistics. * Refers to higher secondary only: Bolivia - 4 years; Brazil - 3 years; El Salvador - 3 years; Venezuela - 2 years. † Due to its low enrollment ratio, Haiti could also be considered Pre-Stage ;I ‡ Numbers refer to some states in Brazil. Definitions: Gross enrollment ratio is the ratio of total enrollment, regardless of age, to the population of the age group that officially corresponds to the level of education shown. Estimates are based onU NESCO´s classification of education levels, as follows. Preprimary provides education for children not old enough to enter school at the primary level. Primary provides the basic elements of education at elementary or primary schools.S econdary provides general or specialized instruction at middle, secondary, or high schools, teacher training schools, and vocational or technical schools; this level of education is based on at least fours years of instruction at the primary level.T ertiary requires, as a minimum condition of admission, the successful completion of education at the secondary level or evidence of atnta- i ment of an equivalent level of knowledge and is provided at universities, teachers colleges, and higher-level professional schools. Net enrollment ratio is the ratio of the number of children of official school age (as defined by the education system) enrolled in school to the number of children of official school age in the population. Sources: UNESCO, World Education Report, 1995; UNESCO, Statistical Yearbook, 1997. World Bank, World Development Indicators, 1998. 13 Five Critical Areas of Reform to Improve Quality and Increase Equity This section examines five critical areas of er- to pupils is limited in all but schools for children form where action is needed to deal with the of higher educated parents, effectively shortening problems of low quality, especially inadequate the average school work day and the amount of educational achievement, and equity: 1) changing time that children spend on schoolwork during the way teachers are trained, operate and are re- the school year. At the secondary level especially, warded within schools; 2) reforming school mna- teacher subject matter knowledge is inadequate. agement to emphasize autonomy, accountability While there is discussion of new pedagogical and teamwork at the school level; 3) providing trends, such as constructivism, classroom prac- adequate learning materials; 4) using information tice benefits little from this approach except in a technology appropriately to improve learning and small number of schools.2 meet new labor market demands; and 5) targeting preschooling, especially for underprivileged Teacher Education populations.1 While there is much progress in these areas throughout the region, much still re- Teacher education in the region is characterized mains to be done. by low standards, inadequate learning of subject matter and lack of practical hands-on activities. TEACHERS: Recently, attempts have been made to raise edu- PARTNERS IN REFORM cational requirements of teachers from compel- tion of secondary level to normal education at the Pedagogy throughout the region’s classrooms, tertiary level. But while teachers' subject knowl- even in the more economically advanced conu- edge does increase, low quality, relatively inxe- tries, leaves much room for improvement. Chli- pensive secondary-level teacher education often dren are told what to write in their notebooks, turns into low-quality, expensive tertiary-level either arithmetic examples or short essays, and teacher education. Generally, the practical con- are tested on such exercises rather than problem- tent of teacher training gets weakened, removing solving. In a fifth grade classroom, only a few or delaying contact with real classroom environ- children may understand fractions and how to ments for prospective teachers. add and subtract them. While they can read, their written work is simple. In the more advanced Teacher education in the advanced countries ni- countries, teachers will ask pupils questions and tegrates content (in science, mathematics, social get interaction, but the teacher works directly studies, language, etc.) with developing flexible from the curriculum guide, and rarely goes out- pedagogical skills adapted to children’s multiple side the examples and explanations there. Group ways of learning and problem-solving. work on problem-solving and other innovative approaches using peer tutoring are only begni - ning to be tried. The amount of homework given 2 In many faculties of education in Latin America there appears to be a misunderstanding of the con- cept of "constructivism," leading them to believe 1 The reader will find an extensive discussion of that it requires that individual teachers invent their these issues in the companion background papers own pedagogical approaches rather than identify, listed in the annex. A later chapter discusses how to adapt and utilize best practices to meet students' use financing and information to achieve these goals. needs. 14 The main focus in reformed teacher education in focus almost exclusively on salary issues, often the region should therefore be a greater emphasis have a political agenda and rarely pay attention on classroom management skills, an increase in to professional development or improving student subject content taught to prospective teachers, learning. and an early involvement in classroom prac- ticeC in short, increasing teachers' subject knowl- There is much evidence that the current condi- edge and providing them with more hands-on tions are unsatisfactory. For example, many practical knowledge and approaches. countries in the region report increasing difficulty in recruiting young people into teaching despite, Most teachers in Latin America have been poorly in some cases, high levels of unemployment. trained and are already in the system. Studies Teacher absenteeism is a serious problem in most have shown that good in-service training can countries. The average number of days of teach- have an impact (Lockheed and Verspoor, 1991). ing in one country's rural schools is less thano ne- But in-service training tends to be used to teach half the 200 days scheduled for the legal school teachers the latest pedagogical fad or the latest year. Since effective mechanisms for hiring sub- curriculum reform rather than to help them mi - stitutes are the exception rather than the rule in prove teaching in their classrooms based on an Latin America, teacher absenteeism generally assessment of what they need to become better means a classroom without a teacher for that teachers. A particularly challenging but often day. In addition to teaching for fewer days, lower overlooked problem is training for teachesr in effort is also evident in the reduced number of isolated and single-teacher schools in rural areas. hours of contact with students. Finally, in many This is an issue that several countries (Guate- Latin American countries, a significant number mala, Chile and Ecuador, among others) have of school days are lost because of disputes be- been addressing recently by relying on "quality tween teachers and administrators that result in circles" and "training networks" that allow teach- strikes. In a number of countries, payment poli- ers to meet and share experiences with others in cies for secondary school teachers (such as pay- similar situations. Many of these programs build ment on the basis of hours taught), explicitly dsi- on the experience of the Escuela Nueva in Co- courage a teacher from developing a commitment lombia, a long running rural education program or identification with a particular institution. that has many times demonstrated its effective- ness. Focused, linked and context-sensitive Several studies of teachers' salaries show that, teacher in-service training has to be distinguished along with other civil servants, they experienced from teacher in-service in most countries. The significant salary declines during the 1980s. A latter is usually not linked directly to school mi - comparison of teachers' salaries with nonpublic provement programs and is therefore often never sector jobs that require a comparable level of implemented in clas rooms. education yields no clear trends in the region, particularly after taking into account the often Teacher Working Conditions and Incentives part-time nature of teachers' work (see IDB, 1996, pp. 289-292 for a detailed discussion of Improved teacher training will be irrelevant if this issue). A study of teachers’ salaries in Bo- the conditions under which teachers work are livia (Savedoff and Piras, 1998) showed that they not improved. Worsening working conditions, were higher than in the private sector but lower inadequate attention to professional develop- than in the unionized public sector. At the same ment, poorly designed performance incentives, time, throughout the region many of those at- distorted supervisory practices and rundown tracted into the education profession score lowest physical facilities have contributed to the low on university entrance and other achievement quality of teachers, excessive absenteeism and examinations. This suggests, first, that teacher the inadequate amount of time devoted to real salaries should be increased at least in accor- classroom learning. In several cases, teachers dance with economic growth, and that in some unions can exacerbate these problems since they countries major efforts are needed to make up for 15 recent relative declines in salary competitiveness. way to counter teacher shortages but also in a In the long term, major increases in teachers' deliberate attempt at improving the quality of salaries could lead to a better and more commti- teaching in certain subjects. ted performance, but only if such increases are accompanied by increased responsibilities and MORE AND BETTER TEXTBOOKS accountability, rewards for superior performance AND TEACHING MATERIALS and penalties for gross misconduct such as con- tinued absenteeism. All of these actions will need With a few exceptions, Latin American countries to be targeted to schools located in less affluent have policies to provide books free of charge to areas. primary school pupils, but only in exceptional cases to secondary students. However, except in An integral policy response to these complex si- the richest of the region’s educational systems, sues implies gaining insight in matters such as not every pupil has a book, and in many teacher career decisions, recruitment and selce- schools, the books are kept in the school office. tion patterns, incentives and labor markets. Inno- Because book deliveries are irregular (in many vative approaches will be needed in the areas of countries textbooks are sent to schools only once incentives, accountability, career regulations and every few years), principals often tend to hoard mechanisms for recruitment, and selection and their books, either keeping them in the main fo- promotion of teachers. Supervision will need to fice, or making pupils return them at the end of be put on a more systematic basis by utilizing the school day. Although books are, in theory, objective information on student performance to available for every pupil, they are not necessarily assist schools and teachers to improve pedagogy. able to use them regularly in school or take them home. This alters teaching methods, the ability to Limited but so far consistent evidence indicates assign homework, and the regularity with which that institutional contexts and incentives can pupils read or do problem sets. For these reasons, make a difference in the productivity of teaching improving the availability of textbooks will pir- (Savedoff, 1998; Hanushek and Jorgenson, 1996, marily benefit low-income sut dents. and Navarro and De La Cruz, 1998). Account- ability mechanisms, empowering teachers, dirce- In addition, misinterpretations of modern peda- tors and communities, can influence teacher at- gogical theories have led some teachers to argue tendance, teacher concern for seeking relevant that books should no longer be used in the clas- training— as opposed to formalistic certifica- room. Finally, the choice of textbooks has been tion and generally improve performance. nI - traditionally highly centralized in most systems centives based on performance are being tried, given economies of scale in the production of mostly as group rather than individual incentives, printed material, but a cost has been paid in in Chile and Uruguay. And monitoring systems terms of the local relevance of content, its adapt- with teeth are being implemented to enforce fro- ability to changed curricula or school conditions, mal accountability systems (such as school atten- which points to the need to examine this issue dance). carefully before designing systems and proce- dures for the selection of textbooks. Teacher certification requirements also need to change. Teachers, especially at the secondary Policy-makers have long argued that school ma- level, should increasingly be drawn from univre- terials, in addition to books, have a major impact sity graduates in disciplines other than teaching, on the quality of education. Schools that have who can then be trained in the specific art of paper, scissors, crayons, and science equipment teaching through one-year (or less) teaching cerit- available, as well as a variety of children's books fication courses and intensive practice teaching other than textbooks, provide a much richer envi- and in-service training. Some countries have al- ronment for learning than the bare classrooms ready created flexible conditions for profession- found in most of the region. Learning materials als to join the teaching profession, as a practical 16 enrich the learning experience of pupils that do In spite of their promise, the large-scale cost- not have access to them at home. effectiveness of computers and related Internet technologies on primary and secondary educa- Research shows clearly that spending on mateir- tion has not yet been proven.3 Evaluations of als has a much higher payoff than increasing small-scale experiments have shown that com- teacher salaries or reducing student-teacher ra- puters can have an impact on classroom learning tios, yet the political economy of education budg- (Levin, Glass, and Meister, 1986). Drill and ets is such that funding for materials is cut first practice programs, tutorial programs, a host of when educational spending is reduced. Special commercially available learning games, LOGO, care should be exercised to ensure that financing and computer simulations and animation used to for materials and textbooks is provided in the explain scientific principles, and even word proc- context of sustainable institutional and budgetary essing, with its built-in spell check and thesaurus arrangements in each country. programs, are all effective in improving student learning. The advent of Internet provides another TECHNOLOGY IN EDUCATION: popular form of data gathering for research and A NEW SOLUTION? access to a wealth of information. A recent IDB report (Castro, 1998) summarizes But the costs of computers, especially in Latin the current situation of technology in education. America, are still excessively high even with Briefly, the region’s record in the use of radio positive results. The current costs for equipment, and television for distance education and training computer specialists, maintenance personnel, and is impressive. Mexico has been operating its educational software in Spanish and Portuguese Telesecundaria for many years, with many mli- range from $40 per student for just two hours per lions of graduates. Brazil’s Globo Network, and week of computer interaction, to hundreds of its predecessor, Telecurso 2000, which offers a dollars per year. Furthermore if computers are to secondary equivalency program for young adults, be integrated into classroom teaching, then teach- has educated millions of the country’s poor, who ers must acquire more sophisticated teaching otherwise would not have access to secondary skills. Experience has shown that technology in education. CENAMEC in Venezuela reaches the classrooms works as a learning tool only if thousands of students through a highly-effective teachers are thoroughly trained to use the tech- primary-level radio instruction program. Evalua- nology and they find that it actually complements tions of interactive radio in Nicaragua, Venezuela their work or saves them time. No wonder, then, and Bolivia, as well as earlier evaluations of that in almost all schools that have computers in Telesecundaria, have demonstrated that such Latin America and the Caribbean (and in the programs are highly cost-effective. United States), the main result is to familiarize students with the technology itself (which is what Based on this as well as worldwide experience most parents expect) rather than to reform the and research, the case for investing in distance teaching process. learning by radio and television is strong. It is possible to extend education to populations that The challenge is to use computers and the Inter- are costly to reach by conventional schooling as net as a tool to build higher order cognitive skills, well as to train groups in specific skills, including more inquiry and project-focused modes of op- in-service teacher training, where a well-qualified eration and more collaborative working styles, instructor in a central location can teach large and to create smart learners. Past experience in- numbers of students or workers in specialized dicates that this is a difficult and time-consuming tasks. Reasonably high quality education and path, and one that has more to do with school training can be extended to populations who oht- reform than with technology as such. The region erwise would receive weaker quality education or none at all. 3 In contrast, information technology is already revolutionizing higher education. 17 should, therefore, invest in a wide variety of pilot ministry to monitor the actual functioning of the programs that seek to use information technology schools. Central government bureaucrats have to raise the quality of learning in schools, espe- rarely known whether classes are actually taking cially at the secondary level. Within five years place or what students are learning. School ni - these technologies may well become cost- spectors, allegedly the monitors of educational effective in the Latin American context. Hence standards, usually check the most formalistic the need to start early in the development of con- dimensions of instructional practices. In short, trolled experiments that use new software and the central ministry (and its inspectorate) have new approaches to teacher training. been better at repressing change and innovation than at operating the system for steady improve- In all cases it is fundamental not to start with ment. the technology but rather with educational ob- jectives and problems, and then to seek the most The result has been that most public schools in cost effective integrated teaching/learning sys- Latin America do not operate as coherent instiut - tem, including a variety of technologies, which tions with a sense of identity and commitment. need not be the most advanced, to solve those Bureaucratic controls stifle local initiatives. The problems. To be successful inside the school, all school principal, whom most research has shown technology-based reforms require strong support is the fundamental source of school leadership, from the top, acceptance and understanding by has had little authority, prestige or support. teachers, integration into the overall system of There has been little interaction with the commu- instruction, and phased introduction— to over- nity or with parents. come bureaucratic inertia and resistance. It may also be appropriate to experiment with asking Throughout the 1980s and 1990s, in the face of private vendors to provide technology-based these issues and problems, many Latin American learning either inside or outside the school ste- countries (e.g., Argentina, Chile, Colombia, El ting. Salvador, Mexico, Nicaragua, Peru, Venezuela) sought to reduce the central government’s role in SCHOOL AND SYSTEM education by decentralizing educational decision- MANAGEMENT REFORMS: CHANGING making (management) and, to a lesser extent, THE ORGANIZATION OF EDUCATION educational finance. Wherever the initial conid- tions of the educational system are highly cen- Historically, Latin American educational sys- tralized, it is clear that decentralization could be tems have been plagued by excessive centraliza- a tool for education reform? by liberating local tion and lack of school autonomy. It is conven- initiatives, channeling additional funds originated ient to treat the two issues separately since they locally, engaging new stakeholders or disarming are related but not quite the same. Decentraliza- interest groups born out of the excesses of cen- tion means moving decisions to lower levels of tralism. This may have a value in and of itself. In government (from central to provincial, from addition, decentralization of education is often provincial to municipal, etc.). Autonomy refers part of a larger process of political and admin- specifically to schools, although language may be istrative decentralization, in- loosely used in both cases. Traditionally, education has been highly con- trolled from central ministries. Key decisions about who teaches, how and when teachers are rewarded, what they teach, and school manage- ment objectives have never been in the hands of local decision-makers who are most aware of student and community needs. This has been coupled with the lack of capacity in the central 18 cluding increased participation, citizen involve- generally appropriate policy. The objective be- ment and responsive local authorities. Decen- hind school autonomy is to increase innovation tralization can also include direct support to pri- and responsiveness to pupil needs. Given more vate educational institutions and systems, as in local and school-based financial decision-making the case of the publicly-funded Fe y Alegría at the school site, parents will increase participa- schools in several countries and voucher systems tion and school teachers and administrators will in Chile and Colombia. increase quality, both by improving teaching and by using resources more efficiently. But decentralization is not a magic bullet. At its worst, decentralization is simply a means to de- To succeed, schools must collect and use data to volve unsolved educational problems to state or improve their management. Parents mus tcontrol provincial entities. In some cases, central gov- teacher time-on-task; local management must be ernments have decentralized primary education to accountable for teacher costs; incentives must be local levels without also transferring the corer- provided by the center for improved achievement sponding resources. In addition, lower level ad- levels; teachers must be selected according to ministrations are often less capable of imposing merit and with the participation of the commu- meritocratic criteria and are an easier prey to nity; etc. In at least two cases (the EDUCO pro- spoils politics in the hands of local and more tar- gram in El Salvador and the Brazilian state of ditional political leaderships. In most cases, Minas Gerais), research has shown that school lower levels are not fully prepared to receive the attendance and student achievement improved greater responsibilities that they acquire with significantly as a result of these reforms. decentralization and need time to adjust. It may also happen that the overall size of the bureauc- Parental and community participation is also a racy expands, without a corresponding increase key concomitant of school management reforms. in efficiency. In addition, to be effective, decne- A child's learning can be characterized as a proc- tralization of management requires the re- ess of "joint production" involving education in centralization and strengthening of evaluation the family and school. Especially during the and the normative and policy-setting roles of the years of primary schooling, parents who monitor center. their children's education closely are likely to motivate them to do well, as well as to assure With centralized management, "control" of what that their children attend school regularly. Par- is actually being done, or not, matters. Hence the ents collectively can also apply pressure on importance of supervisors and inspectors. With school principals and teachers to exert more ef- decentralized management, the results are what fort (Hannaway and Carnoy, 1993). Community matters. Transparency and equity figure promi- monitoring can also help to reduce absenteeism nently, as do data on enrollment and learning, per by putting community pressure on offending student costs, evaluation, information systems, teachers. El Salvador's EDUCO program has input and participation from stakeholders, and been particularly effective in these approaches. mass campaigns. Supervisory systems need to be reformulated as technical assistance to help The prerequisites to success of school autonomy schools achieve agreed upon goals. and increased parent involvement are both finan- cial and managerial. In the first place, educa- In short, decentralization works as long as it is tional spending will need to remain stable or even part of a broader and coherent reform package increase, so that communities and schools have containing the necessary provisions for prepar- real resources to allocate. In the second place, ing both the center and the local levels for the leaving schools to their own devices has been new functions. By contrast, considering the shown to bring no real benefits. Therefore central usual situation in the region, in which schools or state ministries of education need to become are powerless to manage their everyday matters, lean, strong and competent agencies with a new encouraging school autonomy tends to be a set of roles such as setting goals, promoting eq- 19 uity, monitoring progress, evaluating educational Costa Rica and Mexico enrolled over half of all 4 results, providing incentives for promoting inno- and 5 year-olds in preschools in 1995, compared vations and rewarding improved performance. to 20 percent or less in El Salvador, Ecuador and Central authorities will need to ensure that poor Honduras. There is a wide variation in the pub- localities and low-income schools have adequate lic/private mix of preschool provision. More mi - financial, technical and human resources to un- portantly, preschooling provision is highly dertake educational improvements by creating skewed. Throughout the region, poor, rural and mechanisms that favor equity in the distribution indigenous children have inadequate access to of resources across states and municipalities. preschooling. In many cases their enrollment ar- They will also need to be ready to intervene in tios are less than half those of more privileged cases of local nepotism or capturing of schools groups. by local power groups. Finally these reforms must be linked with systematic efforts to raise Education policymakers often wonder whether it expectations and to measure the results of school is better to invest in early childhood development and pupil performance. or in improving the quality of primary education, and whether it is the public sector that should be EARLY CHILDHOOD DEVELOPMENT: investing more in preschools. Notice, however, HELPING CHILDREN COPE that sound investment in preschools and, where FROM THE START practicable, in associated health and nutrition services, is a means to improve primary educa- Early childhood development and education is tion. This is consistent with the mandates of the increasingly viewed as a high-yield public in- IDB'S Eighth General Increase of Resources. As vestment, if properly conducted. Investment in noted above, preschool education has a high pay- early childhood development lowers the cost of off in terms of helping low-income children be primary schooling of low-income children who more successful, remain in primary school and, normally come unprepared and physically disad- ultimately, do better in the labor market. Invest- vantaged to learn in first grade. By raising the ment in preschooling is, therefore, a critical eel- readiness to learn of the low-income pupils, ment of improvements in the quality of primary learning expectations in schools also rise, por- education and can effectively lower the cost per ducing significant improvements in overall unit of output of the first years of primary edu- teaching and student performance. Targeted early cation. By carefully targeting its efforts, the ni - childhood development also reduces inequity, creased investment in preschooling can be kept to improves health and may even have a long-term a minimum and the payoff can be particularly impact on increased employability and reduced high.4 delinquency. The IDB is preparing a comprehen- sive discussion of issues and strategies for sup- Most higher-income children across the region port of early childhood development M( orán already receive some form of preschool eduac- 1998; see also Myers, 1997). It should be em- tion. Therefore, the main objective of public pol- phasized that learning begins at birth. Initially, icy should be to target additional public re- childhood nutrition and health issues are paar- sources to low-income, at-risk pupils in urban mount. As a child approaches school age, educa- slums and in rural areas. tional and socialization objectives become ni- creasingly important. This paper mainly dsi - cusses the latter, especially the one or two years (preschooling) preceding entrance into primary school. 4 For example, providing a full year of preschooling to the poorest 25 percent of children at the same cost Latin American and Caribbean countries vary as that of primary school could increase total costs of greatly in the proportion of preschool age chli- a six-year primary system by no more than 4 percent. dren in preschools. For example, Argentina, 20 In many countries, a significant fraction of en- and community centers through radio broadcasts rollment, especially among the poor, is in mulit- to caretakers' homes. ple informal, private, unregulated and unac- counted-for day-care centers operating only a few Preschooling should not become an extension of months per year with untrained personnel. Hence, the formal primary system, resulting in exces- when low-income children do getp reschooling, it sively high costs and erroneous pedagogical ap- is often informal, of very low quality and may proaches, following the easy path of building, even have a negative impact on their social and hiring and inaugurating schools, with a high po- intellectual development. Public support, theer- litical payoff but little improvement in thee duca- fore, should also include training and subsidies to bility of children. In fact, nontargeted, formalized improve the quality of nonpublic services so that preschooling may well detract from the painful that they focus more on development than simply efforts to improve the quality of primary educa- on caretaking. Child care programs for poor tion. In short, governments should limit coverage families in the region, such as Wawa Wasi in to whatever size group the budget can afford Peru and Hogares de Cuidado in Venezuela, without compromising core standards of quality, might be cost-effectively upgraded to include beginning with targeted children at high risk of pedagogical components, as an alternative to cer- failure in the present education system. In par- ating higher-cost preschools linked with the pub- ticular, they should strengthen the educational lic system. Bolivia (PIDI program) has developed content of low-cost private and NGO-provided a highly innovative program for improving the programs. As additional funds become available quality of preschooling provided in private homes public support can be extended to less at-risk children. 21 The Challenge of Secondary Education STRUCTURE: obsolete. It neither offers a sound academic edu- THE MANY ROLES OF SECONDARY cation that is compatible with the needs of a EDUCATION AND THE INCREASED modern society nor does it handle well the task of HETEROGENEITY OF STUDENTS preparing some students to enter the labor mar- ket. The number of subjects taught is excessive, Until recently, secondary education, has been making it difficult to delve into them and develop the "forgotten" level of education in the region, skills. a result of a lack of political visibility and the weaknesses of the constituencies for public sec- In addition to these fundamental structural issues, ondary education. Higher education always had regional and country data, research and observa- strong interest groups behind it and political ac- tions show that enrollment ratios in secondary tivism at this level has been conspicuous. Since education, in terms of access and equity, are the 1980s, the multilateral banks have targeted much lower than in the region’s chief competitors primary education. As a result, reform and im- (see Table 4); repetition at earlier levels has re- provement are under way and primary compel - sulted in the enrollment of large numbers of tion rates are rapidly increasing, leading to en- overage children and young adults; and the poor, rollment pressures at higher levels. Unforut - especially those living in rural areas, are grossly nately, just when the need to expand and reform underrepresented in secondary education. In secondary education is greatest, funding is inade- terms of costs and efficiency, expenditures per quate and the analytical work is still lacking. student in the region, while varying greatly from Since little has been written up to now on the one country to another, average $400 per student, subject, this paper explicitly covers secondary which is lower than in the region's competitors. education. Two background papers on the subject At the same time, with few exceptions, student- have also been produced. teacher ratios are not high, averaging 16:1 in the region as a whole. In terms of quality, learning in Throughout the region, secondary schools face secondary school is inadequate by international dilemmas posed by their multiple and complex standards; teachers' knowledge of the subjects roles; namely, the task of preparing students for they teach is inadequate, as are their pedagogical higher education, forming solid citizens and skills and, often, their motivation; curriculum is offering what is necessary for those who com- encyclopedic and out of date; learning materials plete their schooling at this level and enter the are scarce and inadequate; schools rarely have a labor market. The problems are complicated by sense of mission and identity; and school dirce- the increasing numbers of students entering sce- tors enjoy little authority and recognition .Many ondary education with far different social back- of these problems are similar to those of primary grounds from those who previously participated education. in a relatively elitist system, as well as by dec- ades of neglect. The result is that, in most coun- tries of the region, the traditional structure, cur- riculum and approach of secondary education is 22 6 Table 4 Historical Evolution of IDB Lending in Education Total Total % Lending Total IDB % Education IDB in Education Lending in Education / Year Primary Secondary Higher Vo-Tec Multi-Level Lending Lending / Total IDB Social Sectors Social (US$ mill.) (US$ Lending (US$ mill.) Sectors mill.) 1965 8.0 8.0 175.2 4.6 11.4 70.2 1966 29.1 29.1 396.0 7.3 139.6 20.8 1967 61.9 4.0 65.9 481.8 13.7 124.8 52.8 1968 9.4 9.4 402.7 2.3 56.3 16.7 1969 9.5 16.9 26.4 644.5 4.1 99.7 26.5 1970 6.7 4.5 11.2 626.1 1.8 84.7 13.2 1971 56.4 10.9 67.3 628.0 10.7 127 53.0 1972 13.4 15.3 28.7 772.7 3.7 133.9 21.4 1973 32.5 25.3 57.8 870.4 6.6 152.5 37.9 1974 9.0 10.0 19.0 1102.1 1.7 129.3 14.7 1975 50.0 20.8 70.8 1360.2 5.2 214.8 33.0 1976 12.8 6.6 19.4 1462.9 1.3 238.8 8.1 1977 9.6 14.6 20.4 44.6 1809.6 2.5 349 12.8 1978 54.9 30.0 15.5 100.4 1823.3 5.5 243.4 41.2 1979 13.9 13.3 27.2 2177.2 1.2 249.4 10.9 1980 32.5 21.0 53.5 2114.6 2.5 337 15.9 1981 7.5 7.2 14.7 2437.8 0.6 282.9 5.2 1982 28.0 147.8 25.4 201.2 2688.9 7.5 767.1 26.2 1983 71.0 2.0 78.5 15.5 167.0 3000.6 5.6 486.2 34.3 1984 25.4 25.4 3500.7 0.7 605.5 4.2 1985 19.9 37.5 34.6 92.0 2985.4 3.1 413.7 22.2 1986 55.8 12.0 50.4 118.2 2974.8 4.0 930.1 12.7 1987 12.1 63.2 75.3 2286.6 3.3 637.2 11.8 1988 0.0 1611.3 0.0 484.1 0.0 1989 90.4 4.7 20.0 115.1 2552.9 4.5 637.3 18.1 1990 14.4 14.4 3803.3 0.4 1680.7 0.9 1991 57.3 57.3 5330.3 1.1 2971.4 1.9 1992 39.6 56.0 40.0 135.6 5994.6 2.3 3295 4.1 1993 177.8 43.0 220.8 5963.2 3.7 2599.1 8.5 1994 473.0 175.1 321.0 969.1 5231.5 18.5 3681.4 26.3 1995 69.6 37.3 106.9 7223.3 1.5 4994.9 2.1 1996 140.0 3.3 100.0 243.3 6740.6 3.6 5137.3 4.7 1997 6.6 421.0 82.5 103.0 613.0 6024.0 10.2 3414.72 18.0 1998 144.1 144.1 2233.7 6.5 1772.5 8.1 TOTAL 1335.1 561.0 895.5 583.8 576.8 3952.1 89430.8 4.4 37482.72 10.5 Source: IDB STAIRS Database. * Investments in pre-primary education are included in these figures. Although no precise figures are presented in the relevant loan documents, Bank investments in pre-primary education exceed $26 million. In addition, as of late 1996, Bank investments in project-related training n(oneducation loans) were in excess of $2 billion. 23 Fortunately, many countries, especially the more cal and vocational programs and move them advanced ones, have begun to pay attention to somewhere else. secondary education. Chile, Argentina, Colombia and Uruguay have recently undertaken major • Technical courses, by and large, are being efforts to expand and improve secondary educa- pushed to the post-secondary level. This ap- tion. El Salvador, Costa Rica, Dominican Re- proach resolves the chronic problem of sec- public, Mexico, among others, have sought to ondary schools that are torn between pre- expand and improve lower secondary education paring students for higher education and pro- (grades 6 to 9). Brazil is planning to make secon- viding occupational training (which often dary education improvement a national priority renders them unable to perform correctly ei- over the next four years. The time is, therefore, ther function). ripe to review issues and establish priorities. • Several large countries are opting for techni- One issue that is hotly debated is the relationship cal tracks that are less an occupational between academic, technical/vocational, scine- training than an academic education with tific/mathematical, liberal arts and gen- emphasis on some broad family of occupa- eral/business education. Around the world there tions. These are the polimodales of Argen- is an almost infinite variety of approaches to this tina and Mexico, which are not very different issue. The United States and Canada have single from the French baccalaureats. comprehensive secondary schools which offer academic programs with varying degrees of • The few successful secondary technical tracking and demands, as well as voca- schools are closely tied to industry. They tional/technical programs. France splits post are expensive, offer high quality education primary schools into different modalities— and training and tend to specialize in one ni- mainly math/science, arts and humanities, dustrial area (electronics, dairy, wines, ce- commercial and technical and also has separate ramics and precision mechanics). A few of vocational schools. Traditionally, Latin America the best schools in this category offer a vari- has opted for the academic/vocational dichotomy; ety of services to the industrial sectors they and several countries have also adopted the cater to, sometimes including research and French model of splitting academic schools. development. Germany sends two thirds of each secondary school cohort through its highly praised • In all cases, there is the need to make the apprenticeship system. If this perplexing variety teaching of theory more applied, more con- of models were not enough, several industrialized crete, more focused on solving problems, countries are in the process of rethinking these rather than being simply the memorization approaches and models. of facts and theories. One of the new tenden- cies is to make academic education more Examining worldwide and regional issues and practical and concrete which is not the same trends, we can identify the following important as vocational. Academic and vocational edu- new approaches with regard to structure and cation need to strengthen the bridge between schooling objectives: theory and practice because academic educa- tion needs practical applications and voca- • There is an increasing separation of voca- tional education needs training in stronger tional preparation from the academic theoretical foundations. streams, the major exception being in the ar- eas of business and office training that have Beyond these approaches, there remains the a natural affinity with academic studies. complex dilemma of the choice between offre- Academic schools need to shed their techin- 24 ing the same syllabus to all or tracking students areas and urban slums (including teacher incen- and offering easier or more applied programs to tives); financial and other incentives for moit- some students and academically more demanding vated teachers and principals to work in under- ones to others. There is as yet no simple and privileged areas; distance education for hard-to- clear approach to this difficult question. reach groups (rural and young adults); and, in some cases, direct financial support to lower- MEETING QUANTITATIVE NEEDS income students to cover the cost of foregone AND QUALITATIVE CHALLENGES wages (in the form of, say, textbook subsidies). Most countries are targeting their school con- In addition to these structural questions, secon- struction efforts, and a few (Mexico, El Salvador dary education faces equally difficult issues of and Brazil) are using distance education for hard- quantity and quality. The region’s current 55 to-reach populations. But very few are providing percent enrollment ratio will need to be in- monetary and other incentives for qualified creased significantly in order to raise the over- teachers to work in the urban slums and rural all educational attainment level of the general areas. population to compete on international markets and reduce inequity in the region. For the lower- Qualitative improvement in secondary education income countries, the focus will be on expanding is similar to that of primary education, but there access to a complete lower-secondary education, are differences. As in primary education, the fun- especially in rural areas. For higher income damental need is for increased learning, espe- countries, the focus will be on reaching theu n- cially of higher order skills in mathematics, derserved populations in lower-secondary eduac- communications and language. In addition, sec- tion and on the overall expansion of upper- ondary education in the region can no longer af- secondary education. As noted, increases in en- ford to overlook the increasing importance of rollment in each country will depend on socoi- developing abilities and skills other than aac- economic conditions and outlook, as well as on demic ones. Learning goals, therefore, should the rigor with which they identify cost-effective also include increased capacity for cooperative solutions. Most countries in the region have al- problem solving and teamwork; civic responis- ready embarked on secondary school expansion bility in a democratic society; creativity and ni- programs. novation; an understanding of the role of technol- ogy in society; environmental awareness; and Based on a simulation of projected increases in knowledge of foreign languages. Studies are secondary education, from 55 to 75 percent of needed to identify how best to achieve these df-i the school-age population, the region would ferent learning goals. The problem of learning in need to invest over US$10 billion in secondary secondary education is often not only a problem school construction over the next 10-12 years. of inadequate curriculum, but also of a lack of Given recent and expected economic growth in correspondence between the curriculum that is Latin America, this amount is not excessive. In actually being taught and the official curriculum, fact, if economic growth continues at the pace the textbooks used, and teacher training, espe- registered between 1990 and 1996 (3.2 percent cially in the more traditional academic areas of per annum) major increases in quality would also mathematics, science and language. If the prob- be affordable provided the right policies are fol- lem is a lack of correspondence, the solution is lowed. not another round of curriculum reform, but rather improved pedagogy and better teaching For purposes of improving equity, countries will materials. need to implement proactive policies targeting the poor and underserved. These policies could Attracting more highly-qualified teachers is even include targeted school construction in under- more important in secondary education, since served urban slum and rural areas; more recur- secondary school teachers have more labor mar- rent funding and assistance to schools in rural ket options available to them. The increased al- 25 bor market demand for computer skills will make d eveloped countries. As discussed below, this it particularly difficult to recruit mathematics needs to take place in a cost-effective manner. teachers. As a result, financial incentives will need to be provided to all teachers, but especially Perhaps more critically than in primary educa- those in scarcity areas such as mathematics. tion, to ensure a commitment to superior instruc- However, salary increases without increased por - tion and learning, public school management er- fessional responsibilities and accountability could form will need to focus on giving the school id- well be self-defeating. Teacher training instiut- rector authority and adequate remuneration, en- tions and faculties need to receive special attne- sure that he/she is qualified, and provide feed- tion since they are essential for each country’s back as well as rewards on the basis of perform- economic future. Training institutions for secon- ance. The Brazilian state of Paraná as well as dary school teachers need to pay special attention Argentina have initiated these types of secondary to increasing their graduates' knowledge of sub- school reforms. School directors could receive ject matter and ensuring that all new teachers significantly higher salaries provided they are have a practical knowledge of technology. selected on the basis of competency and are sub- ject to performance reviews. Particularly in sec- Improved instruction requires more time on ondary education, authorities will need to provide task. In most Latin American countries5 the financial and administrative incentives for teahc- length of the school day varies between 3 and 4.5 ers to teach full time in one school to ensure that hours. Overall, Latin American public schools teachers can act as part of a team rather than as offer between 500 and 800 hours of schooling itinerant providers of lessons (taxi-teachers). The per year compared to the 1,200 hours offered by physical facilities of secondary schools will er- schools in the industrialized countries (OAS, quire adequate space for teachers to prepare their 1998). In addition, in many countries between 10 lessons and interact with other teachers. Efforts and 40 days are regularly lost to strikes (IDB, should be made to end the isolation of teachers 1996). Inside the classroom much time is taken by opening up the classroom to observation by up with routine and administrative matters rather colleagues and advisers. Parental involvement, than active teaching. A number of countries while important, will not be a panacea, especially (Chile, the state of Paraná in Brazil, and Uru- in cases where parents only have a primary edu- guay) are actively seeking to increase student cation or students are already young adults. contact hours to approach the level of 5 Time in school appears to be adequate in the Eng- lish-speaking Caribbean. 26 Navigating and Managing the System: Tools to Achieve Reform Goals Declining birth rates and a decreased dependency will decrease to 1.0 by the year 2010 from the ratio in many countries will ease the pressure on 1990 figure of 1.3. Between 1995 and 2005 the public spending for expanding lower levels of school-age population is expected to increase by schooling and should allow governments to focus no more than 8 percent. Countries that were se- more on raising the quality of primary education. verely hit by the crisis of the eighties have not, by But increased financing needs to be provided in a the late nineties, recovered the per capita levels of cost-effective manner and must include incentives public spending reached twenty years ago, but for improved performance. At the same time, most of them are clearly making substantial ef- information systems must be devised to measure forts at reestablishing minimally adequate fund- the impact of the increased investment and to ing levels. permit adjustments in the financing mix. On average, public expenditure in the region is EDUCATION FINANCING 4.5 percent of GDP, about as much as could be AND EFFICIENCY: expected in education as a proportion of GDP. MORE RESOURCES, BETTER USED This average masks wide variations, with the English-speaking Caribbean countries well above Managing Educational Financing and many other countries in Central and South America well below international norms. Fur- In Latin America, financial constraints continue thermore, a significantly larger portion of these to hamper improvements in education thus shap- funds than in the Far East go to higher education, ing educational reform and limiting the room for and only 1.1 percent of GDP goes to primary unwise policy initiatives. In the early 1980s, just education (see Table 5). Deep distortions and as they should have begun transforming their problems persist in education finance in the er- industrial base and investing heavily in improving gion. A substantial increase in funding for the the educational infrastructure, most countries in education sector under the current rules for the region well besieged by a massive debt crisis. budget distribution and use faces a serious risk of Although educational expansion continued, it did being misallocated (see Graph 4). so at the cost of allowing educational quality to stagnate or decline even further. Yet, the er- A recent report that reviewed the IDB’s lending sumption of economic growth in the 1990s has experience for educational projects D( eutsch and not dampened pressures to keep public spending Verdisco, 1997) found little presence, in most down. Nonetheless, Latin America and the Car- projects, of system-wide assessments of educa- ibbean have a window of financial opportunity tion finance, calling into question the for investment over the next twenty years, as the sustainability of projects and the overall sound- demographic transition results in smaller num- ness of the policies encouraged. As an illustar- bers of school age children and an increased per- tion, most education systems assign increases in centage of the population in the labor force (see public financing to teachers' salaries, leaving Graph 3 and Birdsall, 1998). Dependency ratios other key inputsC textbooks, school supplies— 27 Table 5 Education Policy and Resources: Comparative Studies Public Expenditure Primary Pupil- Duration of Pri- Country on Education Expenditure per Student Teacher Ratio mary Education Primary % of Secondary % of Tertiary % of GNP Pupils per Teacher Years % of GNP GNP per Capita GNP per Capita per Capita LATIN AMERICA AND THE CARIBBEAN 1980 1995 1980 1994 1980 1995 1980 1995 1995 1995 Argentina 2.7 4.5 6.5 16.2 -- 12.0 10.4 17.0 -- 7 Bolivia 4.4 6.6 13.7 -- -- 18.0 -- 67.0 -- 8 Brazil 3.6 -- 8.7 -- 11.0 -- 0.1 -- 23 8 Chile 4.6 2.9 9.6 8.5 -- 9.0 -- 21.0 27 8 Colombia 1.9 3.5 5.2 10.5 -- 11.0 41.1 29.0 25 5 Costa Rica 7.8 4.5 13.1 10.6 -- 19.0 76.1 44.0 31 6 Cuba 7.2 -- 10.4 -- -- -- 28.5 -- 14 6 Dom. Rep. 2.2 1.9 3.1 2.9 -- 5.0 -- 5.0 35 8 Ecuador 5.6 3.4 5.6 3.9 -- 15.0 22.3 34.0 26 6 El Salvador 3.9 2.2 12.4 -- -- 5.0 103.5 8.0 28 9 Guatemala -- 1.7 4.9 6.2 -- 5.0 -- 33.0 34 6 Haiti 1.5 -- 5.9 -- -- -- 65.3 -- -- 6 Honduras 3.2 3.9 10.9 -- -- 22.0 72.1 59.0 35 6 Jamaica 7.0 8.2 14.0 14.7 -- 25.0 166.6 193.0 37 6 Mexico 4.7 5.3 4.3 7.8 -- 20.0 -- 61.0 29 6 Nicaragua 3.4 -- 7.8 13.1 -- -- 85.9 -- 38 6 Panama 4.8 5.2 12.0 11.7 -- 13.0 29.1 47.0 -- 6 Paraguay 1.5 2.9 -- 7.9 -- 11.0 -- 52.0 24 6 Peru 3.1 -- 7.2 -- -- -- 5.1 -- 28 6 Trinidad &Tobago 4.0 4.5 9.2 -- -- 17.0 55.1 77.0 25 7 Uruguay 2.3 2.8 9.3 8.3 -- 8.0 -- 28.4 20 6 Venezuela 4.4 5.2 3.0 -- -- -- 56.8 -- 23 9 EUROPEAN UNION Austria 5.6 5.5 16.1 18.8 -- 25.0 37.9 32.0 12 4 Belgium 6.1 5.7 17.8 -- -- 25.0 34.8 35.0 12 6 Denmark 6.9 8.3 38.4 -- -- -- -- 55.0 10 6 Finland 5.3 7.6 20.7 24.0 -- 30.0 27.8 46.0 -- 6 France 5.0 5.9 12.0 15.9 -- 26.0 21.8 24.0 19 5 Germany -- 4.7 -- -- -- -- -- 35.0 18 4 Ireland -- 6.3 11.5 14.9 -- 23.0 38.8 38.0 23 6 Italy -- 4.9 -- 19.9 -- 26.0 -- 23.0 11 5 Netherlands 7.6 5.3 13.8 -- -- 20.0 53.7 44.0 19 6 Portugal 3.8 5.4 13.5 17.2 -- 20.0 -- 25.0 12 6 Spain -- 5.0 -- 14.1 -- 21.0 -- 18.0 18 5 Sweden 9.0 8.0 43.0 45.2 -- -- 25.6 76.0 11 6 United Kingdom 5.6 5.5 16.0 -- -- 22.0 79.7 44.0 19 6 EAST ASIA AND PACIFIC AREA (Selected Countries) Australia 5.5 5.6 -- -- -- 29.6 30.0 -- 16 6 China 2.5 2.3 3.8 5.6 -- 14.0 -- 81.0 24 5 Japan 5.8 3.8 14.8 -- -- 19.0 21.1 16.0 18 6 Korea, Rep. 3.7 3.7 10.4 14.7 -- 12.0 7.1 6.0 32 6 Malaysia 6.0 5.3 12.0 10.9 -- 22.0 148.6 77.0 20 6 N. Zealand 5.8 6.7 15.0 16.9 -- 23.0 33.3 39.0 18 6 Thailand 3.4 4.2 8.8 -- -- 11.0 -- 25.0 20 6 NORTH AMERICA Canada 6.9 7.3 -- -- -- -- 27.9 36.0 16 6 United States 6.7 5.3 27.1 -- -- 24.0 48.3 23.0 16 6 Note: International data on education are compiled byU NESCO´s Division of Statistics in cooperation with national commissions for UNESCO and national statistical services. The data in the table were compiled usinga UNESCO electronic database corresponding to various tables in itsS tatistical Yearbook 1996. Further, data for 1995 are from UNESCO´s forthcoming World Education Report 1998. They are not yet available in time series but are in print in the World Bank’s World Development Indicators 1998. Definitions: Public expenditure on education is the percentage of GNP accounted for by public spending on public education plus subsidies to private education at the primary, secondary, and tertiary levelsE. xpenditure on teaching materials is the percentage of public spending on teaching materials (textbooks, books, and other scholastic supplies) to total public spending on primary or secondary education. Primary pupil-teacher ratio is the number of pupils enrolled in primary school divided by the number of primary school teachers (regardless of their teaching assni-g ment). Duration of primary education is the minimum number of grades (years) a child is expected to cover in primary schooling. 28 Graph 4 Issues of Efficiency: Average Years Required to Graduate from Primary School Nicaragua (6 years) Guatemala (6 years) El Salvador (9 years) Honduras (6 years) Peru (6 years) Minas Ge- rais Costa Rica (8 years) (6 years) Chile (5 years) Venezuela (6 years) Colombia (5 years) Uruguay (6 years) Panama (6 years) 6 7 8 9 10 11 12 chronically underfunded (Arcia, Alvarez and over time and by category and level. Measuring Scobie, 1997). Similarly, during periods of fi- total spending is made much more complicated in nancial difficulty, funding for school materials is countries where educational spending is con- the first budget item to be reduced. Lending by trolled in part or in whole by provinces, states, or international agencies for textbooks, maintenance municipalities. Without adequate data and care- or school supplies, for example, may have the ful analysis, such topics will remain the subject unintended result of reinforcing misguided and of casual policies and, often, misguided deci- unsustainable financing arrangements. sions. In some cases, a significant portion of spending Another critical reform, discussed earlier in this on education goes to bloated bureaucracies at paper, is to change the way teachers are paid. ministerial levels, excessive numbers of non- Often, initial salaries are too low and fail to at- teaching staff and janitorial personnel. In some tract the best potential candidates. Professional states of the Brazilian Northeast, less than one advancement is based exclusively on seniority fourth of all financial resources were spent in the and the retirement age is way below that for other classrooms and some schools had more non- occupations. teaching than teaching staff. The reform process should also take equity di- Critical to the reform of financial management rectly into consideration when making financial systems is better knowledge about expenditures. decisions. Education in rural and disadvantaged Many ministries and secretaries of education lack areas, which have little voice and votes, is sys- information about total spending on education tematically discriminated against. Forty percent 29 of teachers in primary education serving these lic, and Swope, 1998, on Fe y Alegría) that pri- groups are untrained, compared to almost none vately managed education is less costly than for students in the higher-income brackets. In public education on a per student basis and, even some countries, only ten percent of poor rural after controlling for income, produces higher children have access to preschooling, compared achieving students. This is apparently a result of to 45 percent of the urban nonpoor. The public strong school directors who pay increased atten- education system is increasingly abandoned for tion to results, higher student teacher ratios, and private schooling by upper- and middle-class ni - increased expenditures on learning materials. come students. Their potential for a relatively These conclusions are contradicted by recent er- stronger voice in political affairs contributes to search in Chile (Carnoy, 1998) that shows that as widening the gap in the quality of education er- private schools become publicly subsidized, they ceived by the privileged few and the general lose their cost advantages, continue to select only population. In several countries, over 30 percent those students with expected higher academic of the education budget goes to support free achievement, and do not raise the achievement of higher education for the middle and upper poorer students. Also, while most studies attempt classes. to control for the socioeconomic level of students, they may inadequately measure the selection bi- Education reform should also be geared to ef- ases of private schools. Finally, institutions such fectively leveraging private resources for edu- as Fe y Alegría, a nonprofit organization affili- cation and to move from subsidizing supply to ated with the Catholic church which is over 90 demand, while paying due attention to equity percent publicly funded, must be differentiated objectives. Private spending on education is a from other private for profit or nonprofit schools. major component of total educational spending in At the very least, with adequate oversight and most Latin American countries. Even within the regulation, public support for and encouragement public system, private spending has been "hdi- of privately provided education can leverage den" in the form of family expenditures on uni- public funds. Appropriate incentive and account- forms, school supplies, books, transportation and ability systems for private schools must be in "voluntary" contributions to schools through par- place to increase the chances of achieving ni- ent organizations. In addition, children from rural creased equity, efficiency and diversity. areas and small villages who attend secondary school in larger towns often live in the home of In sum, be it on quality, equity or efficiency friends and relatives. There are many opportuni- grounds, education financing is now generally ties to encourage private support of public edu- recognized to be closely associated with patterns cation through family subsidization and nonprofit of resource use. If education finance is put right, management of schools by NGOs, provided eq- significant resources will be liberated to be spent uity considerations are kept in mind. in underfunded areas and, at the same time, the resources available will be better used. Private education, especially at the secondary level, can be strengthened and even supported, COST-EFFECTIVENESS IN PRIMARY thus relieving a part of the burden on public re- AND SECONDARY EDUCATION sources, provided public education of quality is also made available to the underprivileged. Chile The total amount of resources invested by the and Colombia have already implemented voucher IDB in education during three decades of educa- systems whereby poor students can attend private tional lending is less than the amount of er- schools. Efforts could be made to increase the sources lost in one single year due to high pir- access of private institutions to credit markets for mary school repetition rates (see Graph 4 for a school construction. Governments could also er- summary of average years to graduate from pri- strain from building public schools in middle- mary education). A major effort is already under and upper-income localities. There is some evi- way in a number of countries to reduce repeit- dence (Jiménez, 1995, on the Dominican Repub- tion, the most common response to the problem 30 being mandatory automatic promotion. While as the medium-cost high-impact interventions this can be considered, for the most part, a step discussed in this paper. forward when compared to the "culture of repeit- tion" prevalent in so many education systems in At the secondary level, expansion and quality the region, the decision to ban repetition must be improvements together are costly. It is important accompanied by wider organizational and peda- therefore to seek the most cost-effective mixes of gogic reforms. The establishment of automatic inputs to secondary education in order to achieve promotion between grades one and two by the both quantitative and qualitative goals. This issue state of Sn o Paulo in 1990 provides an example has not been adequately analyzed or explained. of the unintended consequences of not following this advice. The state government did not accom- Student-teacher ratios are, by far, the most im- pany this administrative reform with programs of portant element affecting unit costs. In nearly all teacher training and intensive support for slow Latin American countries it appears that student- learners. The result was a drop in repetition be- teacher ratios in secondary education are not high tween grades one and two but, eventually, an ni - (the regional average is 16:1); but there is a ten- crease in repetition after grade two, with only a dency to reduce student-teacher ratios almost very minor decrease in overall repetition rates. surreptitiouslyC without any explicit policy or More recently, the state has taken additional guidelines. International studies examining the steps to reduce future repetition. impact of class size on learning show ambiguous results. Korea, with among the highest scores on A recent study (Schiefelbein et al., 1998) asked a international tests, has had for many years a stu- group of experts to estimate the cost- dent class size of 40:1 in secondary schools. Re- effectiveness of various primary school intervne- ducing class size would make more sense if it tions in the region. One of the main conclusions were combined with intensive teacher training to of this exercise was that educational decisions change pedagogical approaches and to support throughout Latin America are often made without the teaching of higher order skills. taking cost-effectiveness into account. Big ex- pensive interventions are often undertaken where Student-teacher ratios in secondary education are less costly ones could have a larger impact. Re- based on a combination of student class size, gional educational decisionmakers need to justify number of student-hours per week and number of their decisions and the trade-offs more explicitly. teaching-hours per week. For example, if sut - The study suggested that Latin American coun- dents attend classes 30 hours per week, teachers tries need to seriously question some very expne- teach 20 hours a week, and class size is 25:1, sive interventions which have a small impact on then the student teacher ratio is 16:1. In addition, learning. Examples of such interventions are average school size, incomplete and/or limited school feeding programs; wide-scale, use of infrastructure, restrictions with regard to subject computers instead of using them in pilot projects areas teachers are permitted to teach, and policies first; lowered student-teacher ratios without cor- on minimum and maximum hours of teaching in responding changes in pedagogy, and raising single or multiple schools can also have an mi - overall teachers salaries without complementary pact on student teacher ratios. inputs and redesigned incentives. The study idne- tified a number of almost costless but effective While in nearly all Latin American countries interventions which could be considered6, as well (but not in the English-speaking Caribbean) it is important to increase the hours students spend actively learning in secondary school, a vast 6 For example, assign best teachers to first grade, and expensive construction program (doubling reduce classroom teacher turnover during the school year, enforce regulations on the official length of the the number of physical facilities to eliminate school year, undertake mass media campaigns to double shifting), as some countries have re- encourage parents to read to children, undertake cently undertaken, is not the only option. Almost sample testing of children and distribution of results. cost-less alternatives include increasing the 31 length of the school year, ending the habit of long Fortunately, many countries in the region are strikes and giving students more homework, making major efforts to improve statistical sys- which has been shown in many studies to have a tems, which is an essential component of better positive impact on student achievement. In short, project design and improved decision-making and single shift with full facilities is an option but policy dialogue. The advent of the computer and other more affordable approaches are available. the Internet has meant that information can now be collected and analyzed rapidly and made Policies which reduce repetition and dropout available in a timely and user-friendly manner. rates will lead to major savings on recurrent Six Latin American countries are participating in costs. Also, as noted previously, distance educa- the OECD/UNESCO indicators project, which tion has been shown to be a cost-effective means establishes a new and reliable set of critical indi- of extending coverage to the underprivileged and cators of inputs, outputs, processes, manage- underserved (especially in rural areas, but also ment, and costs (see OECD, 1998). Brazil and for adults) at lower costs than traditional secon- Chile, in particular, have taken the lead in devel- dary education. Finally, as noted above, support oping education indicators and disseminating of private education can yield public savings, them to stakeholders. provided equity considerations are kept in mind. Some of the statistics which ought to be collected USING INFORMATION to inform policy decisions include number of TO MAKE DECISIONS textbooks in the classroom per student, library books, actual number of school days and hours, Indicators and Student Flows and whether students attended preschool. At the secondary level, systems must be established to In spite of some progress in a few countries, eliminate the double counting of teachers working severe technical (lack of key data, inconsisten- in two or more schools. Systematic measure- cies in definitions) and institutional (low utili- ments of the cost of quality enhancing intervne- zation, little comparability among countries and tions must be undertaken. It is also possible now sometimes among provinces or states in the to systematically evaluate classroom pedagogy same nation) problems persist (McMeekin, (through structured observation of representative 1998). In particular, the incorrect measurement school samples) or even the status of relation- of student flows results in overestimated dropout ships between schools and the community. rates and underestimated repetition. Over ten years ago, Schiefelbein called attention to this Performance Evaluation and Standards problem in a number of Latin American coun- tries. Fletcher and Costa Ribeiro showed that the Until recently, most countries in the region (xe- dropout rate between grades one and two in Bar- cept the English-speaking Caribbean) had little zil was less than 2 percent, compared to "official idea of the extent to which their students were estimates" of 50 percent. Student flows tend to mastering the national curriculum. They had vary greatly by educational cycle (primary, sec- "standards" but did not have the means to meas- ondary), area (urban, semi-urban and rural), ure them or meet them. Nearly every country has gender and region. Until recently, policy deic- now set up a student assessment system which sions were not based on an analysis of these surveys a sample or an entire population of stu- variations. Better statistics, properly used, can dents at a given level of schooling in a nation, have a major impact on policy. For example, in state, province, or municipality. In addition to many countries, the findings on repetition have national tests, a few countries have participated led to a redirection of policies toward primary in international comparative tests. There is ni - education and a focus on understanding and re- creasing awareness that international compari- ducing repetition. sons, far from being simply "bad news," provide a benchmark for setting national learning goals, especially in developing problem-solving skills 32 and higher order thinking as well as basic skills. ing by identifying "effective" schools that score With this in mind, fifteen Latin American coun- higher than expected given the socioeconomic tries joined together in 1995 to develop a regional background of their students, or by measuring the fourth grade test in mathematics and language, improvements in a school's scores over succes- through the UNESCO/OREALC Programa de sive periods. Medición de la Calidad de la Educación, which receives partial support from the IDB. Initial re- To ensure that assessments are a truly a tool for sults of this study are now available (UNESCO, improving learning it will be important to foster 1998). competent testing agencies outside government; train and adequately remunerate experts in cur- Public authorities are establishing a consensus riculum testing, sample survey methodology and to set educational goals and determine whether analytical techniques; participate in regional and children, institutions and school systems are international efforts; explore measurement of meeting them. The establishment of assessments other areas of learning such as civic education; is a critical step toward taking educational goals and as discussed below, carry out applied re- and standards seriously. Throughout the region, search to identify causes of high or low achieve- teachers and their unions have up to now been ment. In addition, the smaller and poorer coun- passive bystanders or even opponents of asses- tries, which lack technical expertise and er- ment. But it is in their best interest to adopt a sources, will need to aim for only a small number more professional approach to their work, up- of tests (e.g., only in math and language) in a few grade classroom procedures and participate fully grades, and should consider forming consortia in all assessment programs. Nonetheless, there (composed, say, of Central American countries), are many shortcomings in the assessment move- riding piggyback on international programs and ment in the region, including inadequate dissemi- taking advantage of technical assistance from nation and usage, technical and methodological neighboring countries with more experience. mistakes and misconceptions and, possibly, negative backlash effects of tests on teaching. Applied Research: Understanding What Works Properly utilized, assessments can influence many elements of the education system, including As of 1998, only a few evaluations of successful national education policy; education reform por- education experiments had been widely dissemi- grams; education curriculum; decisions by par- nated in Latin America. These included radio ents, students, and teachers; local and regional mathematics in Nicaragua (Jamison et al., 1981); education policy; and school-level pedagogy and educational TV in El Salvador (Hornik, 1973); programs. Assessments can help countries bring the Northeast Education Program in Brazil (Har- into closer alignment the intended curriculum, the bison and Hanushek, 1992); Escuela Nueva in actual curriculum, textbooks, teacher knowledge, Colombia (McEwan, 1995; Psacharopoulos et classroom pedagogy and learning. The resulting al., 1995; Rojas and Castillo, 1988); P-900 in adjustments will make it possible to set higher Chile (Gutman, 1993); Escuelas Fe y Alegría but attainable learning standards, improve (Swope et al., 1998); EDUCO in El Salvador teacher training and pedagogy, and ensure that (Ministerio de Educación, 1996); and accelerated the revised curriculum is followed in practice. primary schools in Brazil (Oliveira, 1998). The Assessments can also be used to channel addi- small amount of good research that has been tional resources to the region's neediest schools, adequately disseminated has had extraordinary particularly those in rural areas and in urban impact. slums. For example, Chile's "900 schools pro- gram," is explicitly directed at improving the per- While there may well be scores of research stud- formance of the worst schools. To provide useful ies and good practice cases with reasonably good information to schools, parents and teachers will methodology, very little is used for policy pur- require determining the "value added" of scholo- poses. In addition, much research in the region 33 consists of irrelevant, poorly controlled studies of small-scale experiments with little chance of be- While it would be presumptuous in this docu- ing replicated. To a significant extent this situa- ment to identify a specific research agenda, tion is changing. More policy-based research three clear priorities come to mind. In the first throughout the region is forthcoming, especially place, the national assessments undertaken in as a result of the increased number of national nearly every Latin American country offer a rich assessments of learning which provide clear tar- source of data to measure what works in terms gets to measure the impact of interventions. of increased learning. In the second place, there is a need for applied research on the impact of In order to strengthen applied research, graduate instructional technology on learning. Thirdly, centers of excellence in teaching and research the IDB (as well as the World Bank) has initi- (including regional centers of excellence in tets- ated studies on a wide variety of issues related ing and measurement) must be strengthened. The to classroom pedagogy, teacher attitudes and same holds true for economics, sociology and effectiveness, and teacher careers and working political science departments that today perform conditions. Additional important subjects for most of the quality research on educational policy study include measuring the cost-effectiveness of issues in Latin America and the Caribbean. a wide variety of primary and secondary school interventions; seeking to better understand the It will also be fundamental to create consensus relationship between critical skills needed in the among all stakeholders on the value and im- current and future labor market and the learning portance of research and on an agenda for re- provided by schools; and examining the effce- search, which will vary from country to country. tiveness of new schooling objectives such as civic Researchers, policymakers and the informed responsibility and participation, environmental public must agree on the goals of research and be awareness, good health practices and violence willing to disclose the results, even if ambiguous prevention. "Meta-analysis" of research already or negative. Recent collaborative research efforts completed but poorly disseminated has to be con- in El Salvador, Nicaragua (Reimers and McGinn, sidered another priority. 1997) and Northeast Brazil (Ministry of Educa- tion, 1997) are good models to follow. 34 The Art of Implementing Education Reform ONE PRIORITY, the primary level. In many cases, most children DIFFERENT INITIAL CONDITIONS enter school, but a high percentage repeats early grades and usually drops out after three or four Developing a strategy to move primary and sec- years. Teacher and student absenteeism is high, ondary education forward in each country re- effectively reducing the amount of time that chli- quires deciding which problems need solving dren spend in school during any year. In addition, most urgently and to what degree other issues children and teachers have limited, if any, access can be dealt with simultaneously. In each coun- to books and other school materials, and the try, priority should be given to those reforms quality of teaching is generally poor, limited by that are appropriate to the main problems faced the quality of teacher education and few opporut- in each country. These problems and their solu- nities for in-service training. Many rural schools tions will vary with the economic resources of have a single teacher or lack the last year or two the country and its level of educational devel- of a six-grade system. Most teachers are not opment. For example, the "education index," a trained to teach in multi-grade situations even composite of an adult literacy index and a com- though there are alternatives, such as the Escuela bined index of primary, secondary and higher Nueva and interactive radio, which can greatly education enrollment ratios, has been developed improve the quality of education, even in isolated by the UNDP as part of its larger Human Devel- areas. It is very important to recover the true opment Index (see UNDP 1997). Based on 1994 value of rural schools as a first step in the proc- data, countries with "high" educational develop- ess of turning them into good schools. Instead of ment are Chile, Argentina, Uruguay, Peru, Guy- dismissing rural schools as second rate, appro- ana, Costa Rica and Trinidad and Tobago. The priate recognition has to be made of their pecuil- countries that fall in the "medium" range of edu- arities and adequate support must be provided in cational development are Brazil, Mexico, Co- order to make them as effective as urban schools, lombia, Panama, Venezuela, Ecuador, Paraguay an objective that has already been achieved in and Jamaica. The "low" educational development several cases in the region. nations are Bolivia, Nicaragua, Honduras, Gua- temala, El Salvador, Dominican Republic, Haiti In countries such as Argentina, Chile, Costa and Belize. In addition, there are large disparities Rica, Cuba, Jamaica, Mexico, Trinidad and To- in achievement within countries, especially in bago and Uruguay, 85 percent or more of the large nations. In Brazil, for example, average age cohort completes fifth grade. For these years of schooling range from 4.1 in the Norht - countries and in the larger urban areas of low- east to 6.2 in the Southeast. income countries, the main problem is no longer access or primary school completion, but the In a few lower-income countries, and in the learning process itself. Even in these countries, poorer regions of others, access to full-length problems such as hard to get books, the inability primary schools or to lower secondary schools is of sending books home with the students, teacher still limited. For these countries and regions, ni- and student absenteeism, poor physical facilities creasing initial enrollment is still an issue. Even and bad management are still a problem, par- more of a priority, in these as well as somewhat ticularly in poorer provinces and states. better-off countries, is improving conditions in primary education so that all children complete 35 The focus of secondary education in the lower- the way schooling is delivered while simultaen- income countries will be on expanding access to ously building the capacity to implement the a complete lower-secondary education, espe- changes. cially in rural areas. In higher-income coun- tries, the focus will be on reaching the remain- In the best of all worlds, once reforms are under ing underserved populations in lower-secondary way, they create their own dynamic and serve to education and on expanding upper-secondary weaken traditional obstacles. This is the case education. Quality must be improved in all when reforms unleash talent and innovation that countries subject to their financial means. The was previously constrained by institutional ob- level of increases in secondary enrollment in each stacles or policies that stifled local initiatives. It country will depend on socioeconomic conditions is also the case when new structures are devel- and outlooks, and the rigor with which countries oped that work to increase effort. For example, identify cost-effective solutions to meet both devolving control of educational finances and quantitative and qualitative targets. supervision to the communities may greatly mi - prove teacher and pupil attendance and function Key shared challenges remain for all countries much better than traditional teacher supervision in the region. In practice, this means that mechanisms. Evaluative examinations that show lower-income countries and "late starters," such how schools are doing over time may stimulate as Brazil, hope they can meet coverage and teachers and principals to try much harder to im- quality goals simultaneously through innovative prove pupil performance. But how these well- funding mechanisms, incentive systems and conceived reforms fare in the real world at the short-term training in "best practices." Policy local level depends largely on several factors, as reform must balance immediate quantitative pir- noted below: orities with actions that anticipate the problems that may arise in the next stage. The challenge is • Reform requires leadership and consensus. to acknowledge the most urgent problems and, Considering that the countries of Latin simultaneously, explore means to avoid the difif- America tend to be weak and unable to en- culties experienced by other countries that have force policies which depend on thousands of already achieved "functioning schools." scattered actors (teachers and administra- tors), leaders must foster consensus and build DYNAMICS OF THE REFORM PROCESS coalitions among bureaucrats and busines - men, unions and public opinion, principals Reforming education in any country is a difficult and teachers, as well as parents and students, process. National educational systems have a since educational reforms cannot be drafted particular organizational history, with vested ni - in central or provincial ministries and de- terests built around existing organizational creed from the top. The reform process has structures. Teaching methods and content have to be transparent and engage teachers, par- long histories. Examination systems and the cur- ents, business groups, labor unions and edu- ricula they drive have become part of the nation's cational administrators throughout.7 culture. The educational system in many coun- tries, states and municipalities is the single lagr- est employer and single largest source of income for local professionals. Such structures are ex- tremely difficult to change because they are as 7 The PREAL-Inter-American Dialogue project also much a way of life as a way of making a living. has shown that leadership, policy analysis and stra- tegic direction for the reform process can be por - In Latin America there is little tradition ofa well- vided by locally-based NGOs with some foreign as-sistance. The work being done in Guatemala and by educated, relatively efficient, merit-based local Chilean ex-president Aylwin in Ecuador, which is level administration with internalized profse- centered around policy dialogue as a primary tool, sional norms. Educational reforms must change could be applied to similar exercises. 36 Box 2 CARICOM Countries Many of the observations, conclusions and recommendations in this paper apply to the countries of the Egn- lish-speaking Caribbean (CARICOM). However, these countries have a very different tradition from the other countries in the region and, to some extent, a different set of problems.F or example, the quality of public schools in the CARICOM countries tends to be higher than that of the private schools; schooling begins at age five compared to age seven in most Latin American countries; CARICOM countries have a long tradition of high stakes evaluation of learning and achievement at the end of each cycle of education; the number of hours of schooling is generally higher than in Latin American countries; and textbooks are readily available. Ind a- dition, there is at least one piece of evidence that the CARICOM countries do better than Latin American countries on international examinations: in the 1989 IEA international reading examination Trinidad and Tobago scored significantly higher than Venezuela. On the other hand, absenteeism is a major problem in CARICOM countries and achievement levels for girls are highert han those of boys. Secondary enrollment percentages in CARICOM countries are higher, but percentages in higher education are lower than in the rest of the region. Secondary education is highly differentiated among academic, technical and other types of schools. Overall, CARICOM countries invest more in education as a percentage of GDP than their countre- parts in Latin America. Even with these differences these countries face issues of low achievement, especially among boys. CARICOM countries are seeking to emphasize communications and practical mathematics skills, reducex e- cessive secondary school diversification, expand differentiated post-secondary training and encourage private expenditures in education. These countries are seeking to diversify and expand their economies to reduce high unemployment that also leads to widespread emigration, often of the most talented graduates. • Social marketing is the means to obtain the applied policy research with reform proc- collaboration of civil society and, in par- esses, and consistent and reliable information ticular, parents. Social marketing has been about the course of reform. Consensus-based used extensively in healthb ut, with a few ex- research can reveal whether reform objec- ceptions, totally inadequately in education. tives are being realized and how strategies Contrary to popular belief, social marketing should change.8 is not intuitive. It is not about making more speeches or getting more media exposure. • Checks, balances and incentives must en- Social marketing has become a highly com- courage actors and stakeholders to behave plex and specialized field that involves ac- in ways that strengthen the reform process. tivities from informed policy dialogue among The starting point of such incentives is to ni - the members of the elite, to the key process clude stakeholders in the decision processes, that starts by asking what will resonate with thus developing ownership of reform by key the target clientele of the reforms and with actors. The incentives also include instiut - key stakeholders. It is a professional field of tional analysis and identification of the vair- expertise and, therefore, education reform ables that must change in order to change ni - has to allocate funds for well-designed and stitutional behavior. What is the probability executed social marketing activities. that there will be serious political resistance to the reforms from various • Reforms require transparency, continuity and feedback permitting mid-course correc- tions. Reforms should not be identified with a particular government, which is in power 8 The interaction of applied policy research with for only a short time, but with a commitment participatory approaches can be very fruitful (e.g., by all the political stakeholders. Transpar- Reimers and McGinn’s work on HIID experiences, ency derives from the honest interaction of 1997). 37 groups? Do institutions exist that are willing At the same time, governments and local admni- to implement the reform? What is the incen- istrators have often failed to recognized the difif- tive structure that will make reforms self- cult situations in which teachers work. A major reinforcing and durable? requirement of reforms in most countries of the region must be overcoming a long history of con- • Key actors must have the capacity to imple- flictive relationships between the Ministry of ment reforms. Many reforms assume that Education and teachers' unions, which has built teaching and learning will improve by simply up a thick layer of mutual mistrust. Teachers decentralizing control over decision-making, need to feel that they are valued by administra- creating a much better curriculum, sending tors. Minas Gerais and Chile have already had books to schools, or making in-service trani - success in getting union support for reform by ing available to teachers. This is rarely the improving teacher’s working conditions and pay case. Reforms need to improve the capacity as part of a focus on school improvement. of teachers and administrators at all levels to deliver educational services more effectively. Until recently, international lending agencies paid Capacity building includes developing infor- surprisingly little attention to institutional capac- mation systems, dissemination mechanisms ity and problems of implementation. They have and evaluation systems, and training key ac- now learned that the process of developing the tors to use them to improve teaching and reform and the human capital available to carry it learning. out are absolutely crucial to the reform project, just as the success of a business depends on the A specific area in need of renewed attempts at individuals running it as much as the attractive- policy research and dialogue is the role of ness of the products it produces. Agencies such teacher's unions, a key player in the dynamics of as the IDB can help in building consensus for the educational reform. Conventional wisdom has it reform and can provide or finance technical as- that unions oppose educational reform, even if sistance for strengthening the implementation they may have not always been entirely effective capacity of educational systems. They also have (Murillo, 1997). Indeed, teachers' unions in Latin the responsibility of playing an active role in dsi- American countries have generally done poorly in seminating good practices emerging in the region getting the public to identify with their position, and the world, and their own policy and research in part because teachers' unions have rarely work, thus contributing to a well-informed policy fought for reforms that increase the educational formulation process. performance of children or make schools more effective. 38 Looking to the Future: Developing an IDB Action Plan LEARNING FROM THE PAST: ensure adequate and sustainable physical mainet- THE IDB’S EDUCATION LENDING nance programs. Since 1965, the year its first education loan was Bank lending through 1976 focused on higher extended, the Bank has committed on average education, followed by vocational/technical close to 5 percent of its yearly lending to the edu- training and science and technology (see Table cation sector. Between 1994 and 1997, lending to 4). In 1976, the first loan in support of primary education increased to 9.5 percent of total lend- education was made to Colombia. Since then, the ing. To a large extent, this increase coincided IDB has rapidly expanded its lending to primary with the IDB's new development strategy as ar- education. In fact, from 1978 through 1993, over ticulated in the "Report on the Eighth General 80 percent of Bank lending for education was Increase in Resources" of 1994, which speciif- directed toward primary education, as it became cally provided for increased support to education increasingly clear that the region was far behind (Comas, 1995). its competitors in the provision of an adequate basic education. Over time, and particularly since 1994, the na- ture of lending for education changed as the Support for secondary education, as well as tech- IDB began to more fully understand the nature nical and vocational education, has increased of the education process. The Bank has increas- since 1994. In 1997, secondary and techni- ingly taken a systemic reform-based viewpoint. cal/vocational education accounted for more than Infrastructure support has decreased and more 80 percent of new IDB education loans. Lending support has gone to renovation and individual for higher education as well as for science and school expansion rather than new construction. technology went down after 1989. However, fol- An increasing amount has gone to equipment and lowing the development of strategies in these two materials. Within this budget item, the emphasis areas, an upsurge can be expected in the next few has shifted from laboratory equipment for science years. and technology, to textbooks and, most recently, to mass media and information technology. There IDB LENDING PRIORITIES has also been a major growth of support for in- AND STRATEGY stitutional development, improvement of infro- mation systems and training of teachers and ad- The IDB is committed to the development of ministrators. Curriculum reform is now regularly education in the region. The Bank is expected to linked with components such as textbook por- provide support to all levels of educa- duction and teacher training activities. Quality tionC primary, secondary, and higher education improvement rather than expansion per se has and training, as well as the related areas of early become the focus of recent IDB lending, even childhood development and science and technlo- though this should not be taken as an indication ogy. Given current needs, lending to secondary of a lack of interest in supporting the construc- education is expected to be high in quanti- tion and rehabilitation of educational infrastruc- ture wherever justified, as well as on helping to 39 tative terms, if not in the numbers of operations poverty, or institutional weaknesses cannot exe- approved. cute development projects that are too complex or require upsetting powerful vested interests. Regardless of the size of its lending program in The staff of the Bank will evaluate each situation the sector, the main objective of support for pir- and design the best and most realistic project, mary and secondary education will be to improve considering the great potential for change that is quality and equity. The IDB will also support created by the funds mobilized in a loan, but also increased access to schooling where necessary, the difficulties of implementation. In some cases, especially at the secondary level. it will be necessary to realize that the moment is not ripe for an IDB loan or that objectives, espe- Despite the IDB's intention to support the re- cially those related to reform, need to be modest. forms described herein, it cannot impose eduac- tion reform on countries. The locus of reform has To the extent relevant in each specific context, to be inside the country. The political will, the IDB operations in primary and secondary educa- leadership and the right institutional climate have tion will be guided by the observations and rec- to be present in order for reform to begin and ommendations found in this paper, as discussed thrive. The role of multilateral banks is to be on below. the alert and ready to step in when conditions are ripe. Projects will be designed to respond to the The IDB will support reviving the school as an needs of each country, rather than follow pre-set active sphere of management, innovation and models and standard approaches. The IDB will social responsibility through increased auton- understand the moment of the country, its insit- omy, intense community participation and, tutions and its needs, as well as the variety of wherever relevant, decentralization and local economic and educational conditions within the government involvement. This will be seen as region. part of an integral effort aimed at establishing a new role for central authorities in terms of over- In all of its efforts in primary and secondary sight, provision of incentives and raising expce- education, the IDB will seek to ensure the qual- tations regarding school effort and pupil per- ity, implementability and sustainability of its formance. projects. The "Strategy on Supporting Reform in the Delivery of Social Services" details the nature In view of its importance, the IDB will signifi- of these approaches. In particular, the IDB will cantly increase its financial support of pre- and encourage client, beneficiary and stakeholder in-service training of teachers. This could in- ownership; undertake sectoral and economic clude, inter alia, supporting innovative programs analysis to gain an understanding of the issues and the extension of those programs judged cost- and options in education reform and their real- effective, adequately equipping teacher training tionship to the larger socioeconomic environment; institutions, and establishing incentive programs analyze institutional capacity and sustainability; to attract higher quality teachers to the education and devise simple but robust monitoring and system. The IDB will also support strengthening evaluation instruments. During the process of the role and performance of teachers through ni- designing projects the IDB will test new ideas on novative approaches to recruiting, training, sup- a small-scale (pilot project) basis; use a process porting and rewarding effective teachers and rather than a blueprint approach for education making them true partners in reform efforts. The reform projects; and utilize a wider range of IDB will support effective pre-service training as lending instruments. well as in-service training which is innovative in design and organically linked to sustainable Some countries can tackle very ambitious reform pedagogic and management innovations. The projects because they have stakeholder commti- IDB will support analytical work and programs ment, strong leadership and institutional strength. to improve teacher incentives and accountability Other countries, due to political turmoil, acute at the school level. 40 in isolated areas. At all levels, the Bank will caer- The IDB will support a prudent but intense ap- fully watch the gender balance. It will suggest plication of the potential of technology to ex- compensatory policies both in poor rural areas pand coverage and improve quality. Efforts will where girls fall behind and in more advanced ra- be made to replicate and adapt the already suc- eas where boys' enrollment is falling behind that cessful use of radio and television to improve of girls (particularly at higher levels of educa- quality and increase access, as well as to identify tion). new objectives for mass media programs, and evaluate these programs. The IDB will also sup- The IDB will support better assessment, statis- port a wide variety of development and pilot por- tics, applied research and feedback to grams in the use of computers and the Internet, stakeholders as well as pilot projects, experi- especially in secondary education, focussing on mentation and innovation. This will include en- developing software and training programs, in hanced awareness of bottlenecks in student flows; anticipation of the time when costs will be low monitoring performance on a comparative basis, enough to permit widespread adaptation. not only students in a particular classroom, but all students and schools in a country, or even in a The IDB will focus on sustainable financing of set of countries; and policy-oriented research on the most cost-effective mixes of inputs and proc- education issues, as an important ingredient in esses to raise student achievement and retention strengthening the capacity of all concerned par- in school, and the use of financing mechanisms ties to gather and process the necessary data for to provide incentives for cost-effective behaviors the purpose of informed policy dialogue. among the key education players. The IDB will emphasize support for attempts to improve the The IDB will support primary and secondary productivity of educational spending, rather than education in the context of a concern for mus- simply contributing with additional resources per tering widespread social support for reforms se, even though such additions will, no doubt, be from traditional and nontraditional granted when justified by the particular situation stakeholders alike: students, parents, teacher of a country. In particular, IDB lending for edu- and educational authorities, but also NGOs, cational materials will insist on mechanisms as- businesses, the media and others. Deliberate suring their regular delivery to classrooms and efforts at informed policy dialogue nurtured by their daily use, as well as the development of top level educational research and dissemination budgeting processes within the ministries of eud- of regional and international good practices will cation that sustain steady flows of resources for accompany education reforms at the design, mi - their purchase and distribution in the future. plementation and evaluation stages. The IDB will infuse its support to primary and While the IDB is pleased to respond to the in- secondary education with a clear and sustained creasing requests for support of secondary edu- focus on equity. The objective will be to provide cation, it does not regard primary education as school places of good quality to lower-income a transient preoccupation but as a long- students. At the primary level, this will include a standing commitment. While some countries wide variety of actions: targeting school con- have been able to increase significantly the pre- struction in urban slums and rural areas, sup- formance of their primary schools, the issue of porting compensatory programs aimed at socially primary education is far from resolved and for a disadvantaged and indigenous children, targeting considerable time there will be a need to pursue preschooling, incentive programs for teachers to its further development. The emphasis will be on work in underserved areas, etc. Equity enhancing providing quality education to all children, and actions in secondary education could include di- on ensuring that all children finish six and evne- rect subsidies to poorer students, targeted school tually nine years of basic education with func- construction, teacher incentives, and mass media tional skills and knowledge. While large-scale programs aimed at older students and those living capital investments in primary school construc- 41 tion may no longer be necessary in most coun- of increasing time on task, and use of private tries, major investments are still called for in sources of funding provided equity needs are ad- elements such as preschooling, learning materials dressed. Strategies to attract higher quality and training. The IDB will, therefore, proactively teachers to the profession and to reform teacher review progress in primary education and seek training are of particular importance in secondary follow-on loans to the current set under impel- education. mentation. Given the difficulties of appropriating more pub- With regard to secondary education, the IDB is lic funds for education, there are good reasons to prepared to fund the capital costs associated leverage public funds through support for private with expanding access. While the IDB will pro- education. Subsidies may permit the private sec- vide capital funds for construction of physical tor to cater to needy clienteles, possibly with facilities and provision of equipment, teacher more efficiency and lower unit costs than the training and many other expensive items, unless public sector. Complex regulatory issues will countries are willing to commit additional er- eventually surface as the private sector expands sources to fund current expenditures, the mi - at all levels, as is already the case regarding pri- provements will not materialize and the IDB vate higher education and private providers of loans will not pay off. For the lower-income preschool education. These issues are far from countries, the IDB's focus will be on expanding clear at this point and increasing attention to access to lower-secondary education; for higher- them is to be expected both on the research and income countries, the focus will be on reaching policy fronts. This increased attention will result underserved populations in lower-secondary edu- from public pressures in favor of quality control cation and on overall expansion of upper- and improved information about the options open secondary education. for the students and families. Hence, the IDB will consider funding some fixed costs of private edu- The IDB’s support for increased access to sec- cation, provided equity considerations are met ondary education will always be accompanied and regulatory issues are duly sorted out; this by attention to the design of secondary school may require some changes in its financial tools. models which are relevant for the 21st century, to quality improvement, increased equity and IDB LEADERSHIP IN THE REGION efficiency, and better school management. Rede- fining the model of secondary education could As education reform becomes a priority of gov- include establishing a common curriculum core ernments and society alike, and as innovation emphasizing higher order learning objectives, proliferates, the IDB is uniquely positioned to new approaches to technical and vocational edu- play a role in collecting, systematizing and dsi- cation, new curricula in areas such as technology seminating good practices in the field. In this awareness, and developing outreach programs for role, the Bank can make widely available the sot- young adults seeking secondary school equiva- ries of success and failure for the benefit of each lency. new attempt at tackling the challenges of mi - proving learning, equitable access and educa- Attention to the most cost-effective mix of inputs tional attainment throughout the region. In par- to achieve secondary education goals will include ticular, many new solutions to long-standing edu- policy measures related to student teacher ratios, cation problems are cropping up throughout the inexpensive school construction and optimum region. These efforts can serve as models for school sizes, cost-effective ways other countries facing similar issues. There is much to learn from others, but each situation also requires analyzing what will work in that par- ticular time and place. 42 To play a regional leadership role in primary and sight and supervision of teachers, the potential secondary education requires identifying emegr- role of teachers' unions in improving quality, and ing issues and seeking an understanding and a teachers' salaries. More work will be needed to consensus on them. It is also necessary to interact define best practices, to disseminate the results to with other development agencies and with the policymakers and leaders, and to turn these into countries themselves, and find funding sources. bankable programs. In particular, jointly as well as on its own, the IDB will support applied research, on-site and The IDB supports the UNESCO/OREALC La- virtual conferences, study tours and training pro- boratorio Latinoamericano de Medición de la grams. Calidad de la Educación, which has enabled Latin American countries to share their expeir- The IDB has already identified a number of cru- ences and to give a regional test of reading and cial emerging issues in primary and secondary mathematics achievement to third and fourth education. Work is under way on some of these graders. In the future, it will be important for issues, while the IDB is seeking partnerships and Latin American countries to continue to share funding for others. Four emerging issues that are information and experience on this very complex of particular importance are those related to topic to ensure that assessment and other meas- technology in education, teachers, assessment urement instruments truly serve to improve edu- and statistics, and private education. cation. A specific future need is undertaking ap- plied research to exploit the rich sources of data With regard to technology, the region needs to from country and region-wide assessment efforts, undertake pilot and development projects to as well as promoting regional centers of excel- meet long-term educational needs, with a strong lence in assessment and applied educational er- focus on teacher training and software devel- search. Another fundamental activity is to de- opment. With this in mind, the Bank has idenit- velop comparative education statistics, using a fied math and science education at the secondary "report card approach," to track progress in pri- level as a critical area for improvement. The IDB mary and secondary education which can build has initiated planning work on a regional pilot on existing and future national and international project to improve learning in mathematics and statistical systems. Of particular importance will science in "average" public secondary schools, be to track progress in reducing repetition. using a "multi-channel" approach to learning (e.g., computer, Internet, CDs, radio, television, In some cases, partial subsidies permit the pri- print, face-to-face teacher training etc.). The aim vate sector to cater to needy clienteles with more is to begin with about 50 schools in each of four efficiency and lower costs than the public sector. countries to identify specific learning needs in Hence, the IDB will consider funding some costs science and mathematics, identify the most ap- of all levels of private education, provided equity propriate learning materials worldwide, establish is taken into account. Analytical work is needed the best combination of channels for learning, to identify and evaluate the different ways that and begin in-service teacher training. It is ex- the public sector can support private education to pected that the pilot and development project extend the reach of public funds, increase com- would be financed through a combination of petition and ensure equity, as well as to develop technical cooperation and project funds. The er- new financing tools for the IDB. sult would be to position the region to take ad- vantage of the expected rapid decrease in the cost Additional important subjects for study include of computing power and connectivity. measuring the cost-effectiveness of a wide variety of primary and secondary school interventions; The IDB has undertaken a series of studies ex- seeking to better understand the relationship be- amining ways of attracting better teachers into tween critical skills needed in the current and education, case studies of successful pre- and in- future labor market and the learning provided by service training programs, teacher careers, over- schools; and examining schooling objectives such 43 as civic responsibility and participation, envi- which are very important in some areas. For all ronmental awareness, good health practices, and these reasons, the IDB will reconsider the set of violence prevention, which have rarely been financial instruments that can be deployed to studied in the past. "Meta-analysis" of research support education. One possibility under consdi- already completed but poorly disseminated has to eration is the development of a simplified proec- be considered another priority. dure for financing pilot projects, research and participation in international programs. Other Given the needs and the limited resources avali- possibilities include financing private education able, the Bank will continue to cooperate with through support for student loan schemes (usu- other key players in the field, especially the ally but not exclusively for higher education), World Bank, UNESCO, ECLAC, UNICEF and lines of credit for private school construction and USAID, by identifying common issues and er- voucher systems. search agendas, sharing information, jointly sup- porting conferences and analytical work, and According to the Plan of Action endorsed by the capitalizing on the relative strengths of each of Presidents gathered at the Summit of the Ameir- these organizations. The IDB will make special cas in Santiago de Chile in April, 1998, "… the efforts to link with and begin a dialogue with IDB is encouraged to work with member coun- NGOs working in education, including groups tries to substantially increase the share of new such as the Inter-American Dialogue, the Center lending for primary and secondary education, for Research in Development and Education by more than doubling the quantity over the next (CIDE), and think tanks and advocacy groups at three years, compared to the previous three the country level, as well as the growing number years." The Bank is committed to this objective, of private institutions in education and in pub- although it will be reached only to the extent that lishing. the Bank's borrowers are also committed to this level of external financing, if the IDB has ade- Education loans have become complex operations quate staffing and more streamlined procedures, with far more vulnerable components than infar- and if ongoing projects are implemented efif- structure loans. There is an increasing perception ciently and effectively. The IDB will also have to inside and outside the Bank that present financial take a more proactive attitude in the search for tools are inadequate for dealing with the new meaningful operations. To further the Bank's generation of education loans (the same could be commitment, Management will shortly present an said for social sector loans in general). The action plan for the next several years that out- preparation cycle is too long. There is little flexi- lines steps to achieve the lending goal and other bility to begin some components earlier. There objectives stated in the strategy. are no short procedures to make small loans, 44 References Amadeo, Eduardo; Jose Marcio Camargo; Antonio Emilio S. Marques; Candido Gomes. 1994. Fiscal Crisis and Asymmetries in the Education System in Brazil. InC oping with Crisis: Austerity, Adjustment, and Human Resources, eds. J. Samoff. Paris: ILO-UNESCO Arcia, Gustavo, Carola Alvarez and Tanya Scobie. 1998. El financiamiento de la educación y la reforma educativa: Un marco para la sustentabilidad. In Financiamiento de la educación en América Latina. Santiago, Chile: PREAL-UNESCO Birdsall, Nancy. 1998. Education, the People’s Asset. Washington, D.C.: Inter-American Development Bank. Bock, Kathrin; Timmerman, Dieter. 1995. Education, Youth, and the Labor Market in Germany. Universitat Bielefeld, Fakultat fur Padagogik. Mimeo. Braslavsky, Cecilia. 1995. Transformaciones en curso en el sistema educativo argentino. InE ducación, equidad y competitividad económica en las Américas: Un proyecto del Diálogo Interamericano, Vol. II, eds. J. Puryear and J.J. Brunner. Washington, D.C.: Organization of American States. pp. 7-50. Carnoy, Martin and Castro, Claudio de Moura. 1997. ¿Qué rumbo debe tomar el mejoramiento de educación en America Latina? In La reforma educativa en América Latina, eds. C. de Moura Castro and M. Carnoy. Washington, D.C.: Inter-American Development Bank. Carnoy, Martin and Levin, Henry. 1975. Evaluation of Educational Media: Some Issues.I nstructional Science, vol. 4: 385-406. Carnoy, Martin and McEwan, Patrick J. 1997. Is Private Education More Cost-Effective than Public? The Case of Chile. Stanford University School of Education. Mimeo. Carnoy, Martin. 1996a. Joint Production of Education. In International Encyclopedia of the Economics of Education, Second Edition, ed. M. Carnoy. Cambridge, UK: Elsevier Science, pp. 297-302. Carnoy, Martin; Torres, Carlos Alberto. 1994. Educational Change and Structural Adjustment: A Case Study of Costa Rica. In Coping with Crisis: Austerity, Adjustment and Human Resources, ed. Joel Samoff. Paris: ILO/UNESCO. Cole, Michael and Cole, Sheila R. 1996. The Development of Children. W.H. Freeman & Co. Comas, Xavier. 1995. Strategic and Policy Issues for a Bank Education Strategy. Washington, D.C.: Inter-American Development Bank. Comité Técnico Asesor del diálogo nacional sobre la modernización del la educación chilena. 1994. Los desafíos del la educación chilena frente al Siglo XXI. Santiago de Chile: Editorial Universitaria. Cox, Cristian. 1997. La reforma de la educación chilena: Contexto, contenidos, implementación. Washington, D.C.: Inter-American Dialogue, Program to Promote Educational Reform in Latin America and the Caribbean (PREAL). 45 Crouch, Luis. 1995. Financing Secondary Expansion in Latin America: An Estimation of Magnitudes Required, and Private and Decentralized Options. North Carolina: Research Triangle Institute. Cuban, Larry. 1986. Teachers and Machines: The Classroom Use of Technology since 1920. New York: Teachers College Press. Darling-Hammond, Linda. 1997. The Right to Learn. San Francisco: Jossey-Bass. Deutsch, Ruthane and Verdisco, Aimee. 1997. Lessons to be Learned from the IDB Portfolio of Primary and Secondary Education Loans in Execution. Washington, D.C.: Inter-American Development Bank. Mimeo. Dreeben, Robert. 1968. On What is Learned in School. Massachusetts: Addison-Wesley Publishing Company. ECLAC. 1995. Calidad y equidad de la educación media en Chile. Santiago. Elmore, Richard, et al.1998. How Schools Construct Accountability. Cambridge, MA: Consortium for Policy Research in Education, Harvard Graduate School of Education. Mimeo. Elmore, Richard. 1997. Investment in Teacher Learning. Philadelphia, PA: Consortium for Policy Research in Education. Espínola, Viola. 1993. The Educational Reform of the Military Regime in Chile: The System's Response to Competition, Choice and Market Relations. Ph.D. dissertation. University of Wales. Filmer, D. and Lant Pritchett. 1998. Educational Attainment Profiles of the Poor (and Rich): DHS Evi- dence from Around the Globe. Mimeo. Gardner, Howard. 1993. Frames of Mind: The Theory of Multiple Intelligences. Cambridge, Mass.: Har- vard University Press. Gershberg, Alec I. 1997. Decentralization and Recentralization: Lessons from the Social Sectors in Mexico and Nicaragua. Working Paper. Washington, D.C.: Inter-American Development Bank. Mimeo. Golladay, Frederick, et al. 1996. A Human Capital Strategy for Competing in World Markets. In Towards the Twenty First Century: A Long-Term Development Strategy for the Middle East and North Af- rica. World Bank. Government of Brazil, Ministry of Education. 1997.A valiaHn o do Ensino Medio e Acesso ao Ensino Supe- rior. Brasilia. Government of Brazil, Ministry of Education. 1998. Diretrizes Nacionais Para a OrganizaHn o Curricular de Ensino Medio. Brasilia. Hannaway, Jane and Carnoy, Martin. 1993. Decentralization and School Improvement. San Francisco: Jossey-Bass. Hanushek, Eric A. 1986. The Economics of Schooling: Production and Efficiency in Public Schools. Journal of Economic Literature, Vol. 24. No. 3. Pp. 1141-1177. Hanushek, Eric A. and Jorgenson, Dale W. 1996. Improving America's Schools: The Role of Incentives. National Research Council. Washington, D.C.: National Academy Press. 46 Harbison, Ralph and Hanushek, Eric A. 1992. Educational Performance of the Poor: Lessons from Rural Northeast Brazil. The World Bank. United States of America: Oxford University Press. Inter-American Development Bank. 1994. Report on the Eighth General Increase in the Resources of the Inter-American Development Bank (AB-1704). Washington, D.C.: Inter-American Development Bank. ____________. 1996. Economic and Social Progress in Latin America: Making Social Services Work. Washington, D.C.: Inter-American Development Bank. ____________, Office of Evaluation. 1996. Listening to Stakeholders of Primary Schools. Washington, D.C.: Inter-American Development Bank. Inter-American Dialogue. 1998. The Future at Stake, Report of the Task Force on Education, Equity, and Economic Competitiveness in Latin America and the Caribbean. Washington, D.C. Jamison, D., B. Searle, Heyneman, S.P. and Galda, K. 1981. Improving Elementary Mathematics Educa- tion in Nicaragua: An Experimental Study of the Impact of Textbooks and Radio on Achievement. Discussion Paper No. 81-5. Washington, D.C.: The World Bank. Jimenez, Emmanuel, and Lockheed, Marlaine E. 1995. Public and Private Secondary Education in Devel- oping Countries, A Comparative Study. World Bank Discussion Paper #309.W ashington, D.C. Kestrom, Joyce M. 1998. The Untapped Power of Music: Its Role in the Curriculum and its Effect on Aac- demic Achievement.N ASSP Bulletin, v.82, n.4597, pp.34-43, April. Levin, Henry, Glass, G. and Meister, G. 1987. Cost-Effectiveness Analysis of Computer-Assisted Instruction. Evaluation Review, vol. 11, no. 1 (February): 50-72. Lockheed, Marlaine; Verspoor, Adriaan et al. 1991 .Improving the Quality of Primary Education. New York, Oxford University Press for the World Bank. MacEwan, Patrick. 1995. Primary School Reform for Rural Development: An Evaluation of Colombian New Schools. Washington, D.C.: Inter-American Development Bank. McMeekin, R.W. 1998. Education Statistics in Latin America and the Caribbean. Technical Study. Washington, D.C.: Inter-American Development Bank. Ministerio de Educación. 1997. Evaluación del Programa EDUCO: Impacto de la participación de los padres de familia. Ministerio de Educación, República de El Salvador-Banco Mundial. San Salvador: Mimeo. Morán, Ricardo. 1998. ECCD Strategy: An Outline of Messages. Leveling the Field for the Children of Poverty: Early Child Care and Development and the IDB. Inter-American Development Bank. Washington, D.C.: Mimeo. Murillo, Victoria. 1996. Latin American Unions and the Reform of Social Service Delivery Systems: Institutional Constraints and Policy Choice. Working Paper. Washington, D.C.: Inter-American Development Bank. 47 Murnane, Richard and Levy, Frank. 1996. Teaching the New Basic Skills. New York: Free Press. Murnane, Richard J., Singer, Judith D., Willet, et al. 1991. Who will Teach? Policies that Matter. Cambridge, Massachusetts: Harvard University Press. Myers, Robert G. 1992a. The Twelve Who Survive. London: Rutledge. _______. 1992b. Investing in Early Childhood Development Programs: Toward a Definition of a World Bank Strategy. Washington, D.C.: World Bank, Latin America and the Caribbean Regional Office, Technical Department, Human Resources Division. _______. 1996. Preschool Education in Latin America: State of Practice. Washington, D.C.: Inter- American Dialogue, Program to Promote Educational Reform in Latin America and the Caribbean (PREAL), July. Navarro, Juan Carlos and De La Cruz, Rafael. 1998. Escuelas federales, estatales y sin fines de lucro en Venezuela. In La organización marca la diferencia: Educación y salud en América Latina, ed. W. Savedoff. Washington, D.C.: Inter-American Development Bank. Navarro, Juan Carlos. 1998. Descentralización, gasto y política social. In Descentralización en perspectiva, Rafael De La Cruz. Caracas: IESA. Oliveira, Joao Batista Araujo. 1998. A Pedagogia do Sucesso. Sao Paulo: Instituto Ayrton Senna. OREALC. 1994. Medición de la Calidad de la Educación. Resultados. Santiago de Chile: UNESCO- OREALC. Prawda, Juan. 1993. Educational Decentralization in Latin America: Lessons Learned. International Journal of Educational Development, 13, no, 3: 253-64. Psacharopoulos, George, Rojas, Carlos and Velez, Eduardo. 1993. Achievement Evaluation of Colombia Escuela Nueva: Is Multi-grade the Answer? Comparative Education Review. 37. Pp.263-276. Rama, Germán. 1992. ¿Aprenden los estudiantes en el Ciclo Básico de Educación Media? Montevideo: ECLAC. Reimers, Fernando and McGinn, Noel. 1997. Informed Dialogue: Using Research to Shape Education Policy Around the World. United States of America: Praeger. Rounds Parry, Taryn. 1994. The Impact of Decentralization and Competition on the Quality of Education: An Assessment of Educational Reforms in Chile. University of Georgia. Unpublished paper. Savedoff, William (ed.). 1998. La organización marca la diferencia: Educación y salud en América Latina. Washington, D.C.: Inter-American Development Bank. Savedoff, William and Piras, Claudia. 1998. How Much do Teachers Earn? Working Paper. Inter- American Development Bank. Schiefelbein, Ernesto. 1996. School-related Economic Incentives in Latin America: ReducingD rop-out and Repetition and Combating Child Labor. Innocenti Occasional Papers. Child Right Series No.12. Santiago: CIDE. 48 ________. 1998. Education in the Americas: Quality and Equity in the Globalization Process. Washington, D.C.: Organization of American States. Schiefelbein, Ernesto, Wolff, Laurence and Schiefelbein, Paulina. 1998. Cost Effectiveness of Education Policies in Latin America: A Survey of Expert Opinion. Washington D.C.: Inter-American Development Bank. Shulman, Lee. 1997. Enseñanza, formación de maestros y reforma educativa. In La reforma educativa en Améica Latina, eds. C. de Moura Castro and M. Carnoy. Washington, D.C.: Inter-American De- velopment Bank. Swope, John, Latorre, Latorre and Celedón, Francisca. 1998. Fe y Alegría. Un sistema eficiente de edu- cación primaria en América Latina. Documento No.1. Santiago: CIDE. UNESCO. 1995. World Education Report, 1995. Paris: UNESCO. ________. 1996. Nicaragua´s School Autonomy Reform: A First Look. Washington, D.C.: Mimeo. ________. 1996. Statistical Yearbook, 1996. ________. 1997. Statistical Yearbook, 1997. ________. 1997. World Education Report, 1997. Paris: UNESCO ________. 1998. World Education Report, 1998. Paris: UNESCO. Forthcoming. Vexler, Idel, et. al. 1997. La educación secundaria de adolescentes en el Perú. Lima: Foro Educativo. Winkler, Donald and Rounds, Taryn. 1996. Municipal and Private Sector Response to Decentralization and School Choice. Economics of Education Review, vol. 15, no. 2: 189-204. Wolff, Laurence, Schiefelbein, Ernesto and Valenzuela, Jorge. 199.4 Improving the Quality of Primary Education in Latin america and the Caribbean: Towards the 21st Century. World Bank Discussion Paper No. 257. Washington, D.C.: The World Bank. Wolff, Laurence. 1998. Educational Assessments in Latin America: Current Progress and Future Challenges. Occasional Paper Series. Washington, D.C.: Inter-American Dialogue. World Bank. 1998. World Development Indicators. Washington, D.C.: The World Bank ________. 1994. El Salvador Community Education Strategy: Decentralized School Management. Country Department II, Human Resources Operations Division, Latin America and the Caribbean Regional Office, December 8. World Economic Forum. 1997. The Global Competitiveness Report, 1997. World Economic Report. 49 50 Annexes 51 52 Background Papers for the Strategy Cost Effectiveness of Education Policies in Latin America: a Survey of Expert Opinion, by E. Schiefel- bein, L. Wolff, and P. Schiefelbein. Sustainable Development Department Technical Papers Series, EDU-109, December 1998. (A survey creating an index of cost-effectiveness of primary education policies in the region.) Education Reform in Latin America and the Caribbean, by J. C. Navarro, M. Carnoy and C. de Moura Castro. Unpublished. (A review of issues related to management, financing, and monitoring of primary and secondary education.) The Evolution of IDB Lending in Education, by Aimee Verdisco. Unpublished. (A review of the level, content, and objectives of IDB lending in education.) The School Cycle: Preschool and Primary Education, by J. C. Navarro, M. Carnoy and C. de Moura Castro. Unpublished. (A review of issues related to quality and inputs to pre-schooling and primary education.) Secondary Education in Latin America and the Caribbean: The Challenge of Growth and Reform, by L. Wolff and C. de Moura Castro. Sustainable Development Department Technical Papers Series, EDU-111, January 2000. (A review of access, equity, quality, and efficiency in secondary eduac- tion.) Secondary Schools and the Transition to Work in Latin America and the Caribbean, by C. de Moura Castro and M. Carnoy. Sustainable Development Department Technical Papers Series, EDU-112, February 2000. (A review of curriculum, organization and models of secondary education as they relate to the labor market.) 53 Annex Box 1: What is Quality in Education? Educational quality has many, sometimes conflicting definitions. Quality is best defined as the extent to which children learn the basic skills and knowledge necessary to function in a modern society and utilize these skills in their life. Since this is difficult to measure, there are numerous proxies for this definition: 1. Output Quality: The most fundamental definition of quality is thato f the extent to which children attain the knowledge and skills which society wishes to impart to them. These are not simply academic skills (e.g., Howard Gardner has identified seven such dimensions: linguistic, logical-mathematicalb, odily-kinesthetic, spatial, musi- cal, inter-personal, and intra-personal). In the 21st century the technological revolution will require new worker skills. The new worker will need to be a problem solver, flexible, computer literate, a communicator, and able to work in teams. A high quality elementary and secondary school system iso ne which has a good match between what children learn and what is needed to function in a modern society. Output quality can be measured in thea -l bor market through tests of workers’ knowledge and measurements of worker productivity. This is rarely donee b- cause of complexity and costs. In schools, output quality can be measured by achievement tests. Given current cru- riculum goals throughout the world designed to educate this kind of citizen, low quality can, in part, be defined as “rote learning or learning in isolation” and high quality would be the learning of higher order skills. Implications for Latin America: LAC education systems are not educating for the citizen-worker of the 21st century. Nor is LAC providing adequate mathematics and reading skills. IEA studies show that LAC conu- tries are far below the developed world and also below the Far East and the Middle East. On international tests, LAC scores only above Africa. The English-speaking Caribbean may be doing slightlyb etter. 2. Quality as Value Added: Quality may be defined in relation to “value-added.” In this definition, a high quality school or school system is one which increases the learning of students relative to their status at the beginning of their schooling period. This suggests that a school whose clientele is children of illiterate peasants could be of “higher quality” than an urban middle class school even if its absolute results, in terms of test scores or retention, were relatively low. Quality as value added can be measured statistically and through time series data. A number of countries are explicitly recognizing this element, especially by providing additional resources to at risk schoo, ls recognizing schools doing better than expected (e.g., France) and providing rewards to schools which improve their scores. Implications for Latin America:T here is not enough awareness of this issue in LAC, both among polit- cal leaders and the general public. Chile has recognizedt hat low performing schools need special help and is por- viding incentives for improved performance. 3. Quality as High Standards: Quality may also be defined as “high examination and promotion standards.” This definition is useful if it is accompanied by a commitment that all children can achieve these standards, but is counter-productive if adequate resources are not provided and most children end up failing. In the United States, chaotic expectations of learning lead to no standards, but states are now seeking to define and enforce standards, especially at the secondary level. Implications for Latin America: There is evidence that teachers are using “high standards” to fail chli- dren in first grade when they are unable to read by the end of the first year. But this approach does not take into account different learning styles or children’s impoverished backgrounds, and is counterproductive. At higher grades, while the national curriculum can be clear and modern in its objectives, teachers pay little attention to it and learning is inadequate. The “standards” movement at (for example) the eighth grade level requires realistic curriculum expectations and a commitment providing adequate physical resources and school processes so that all, or nearly all, children can achieve at the desired level. Brazil is moving in this direction. 4. Quality of School Inputs: Quality is often defined in relation to physical and other inputs, such as school buildings, textbooks, computers and number and educational levels of teachers. A high quality school would be one with good physical facilities, adequate equipment and educational materials, and well-trained teachers. It is assumed that input quality will lead to better learning achievement and retention, and there is a fair amount ofe -r search seeking to measure this relationship. Implications for Latin America: School inputs are inadequate in rural schools (e.g., expenditures per student, teacher qualifications) and often in urban slum schools. An extreme case is Northeast Brazil. A review of research has identified textbooks, teacher subject knowledge, time on task, school snacks and others as input ee-l ments that have an impact on achievement, but has not endorsed lower student-teacher ratios as effective, contrary to popular belief. Teacher qualifications, as measured by number of years of formal training, have also been found to have a weak impact, if any, on student learning, probably due to the fact that years of training is a poor proxy for quality teaching. 54 5. Quality in School Processes: Quality may also be considered in relation to schooling processes. In this case quality refers to good school management, well articulated, modern classroom pedagogical practices, flexibility to revise or change processes as needed, and dedicated teachers working together as a team to achieve specific goals. Good school processes are often the “missing element” necessary to ensure that school inputs lead to increased learning and retention. They can be measured mainly through systematic observation (qualitative research). There is increasing interest in this element of schooling. Implications for Latin America: The “typical” school has a director who has been centrally selected on the basis of bureaucratic criteria; teachers often have two or more jobs, rarely remain in the school after teaching, work in isolation, and are inadequately supervised (“civil service” mentality). But there are many efforts in the region to change school processes (e.g.,E scuela Nueva in Colombia, EDUCO in El Salvador, community involve- ment in schooling in Minas Gerais and in secondary schools in Nicaragua, teacher learning circles in Uruguay, etc.) 6. Proxies for Output Quality: levels of literacy and schooling in the general population, and completion and repetition rates: Levels of schooling can be taken as proxy for learning in the general population. This assumes rough equivalency of (for example) secondary school levels across countriesW. ithin the school system, in the ab- sence of well structured, internationally comparable achievement tests, and considering that it is difficult to test children who drop out of school, the number of children who complete a particular level of schooling (e.g., mco- pletion rates) is often taken as a proxy for learning achievement. High repetition rates are also often considered a proxy for low achievement since, clearly, something is amiss when upward of 50 percent of children fail the year. Implications for Latin America: Levels of schooling in the general population are significantly lower than East Asian countries and the difference is growing. Current completion rates are low, compared to compie-t tors. Repetition, especially in basic education, is the highest in the world, leading to inefficient use of funds. Re-p tition is grossly underreported in official statistics. Only modest progress in reducing repetition. 55