Framework for Action - Participants -Organizers
Online coverage - NGO Consultation
Latest news - Follow-up to the Dakar Forum
The opinion of EFA partners - Grassroots stories
The EFA 2000 Assessment - The findings
The regional meetings - Evaluation
Press releases - Press kit
Photo corner - Media contacts
The findings > Thematic Studies> Textbooks and Learning Materials > Part 3
  Country EFA reports
  Regional Frameworks for Action

The provision of learning materials has been affected, like so many other sectors of society, by the political and economic changes of the decade since Jomtien. Worldwide, the principal trends were towards decentralization of selection and procurement, economic liberalization with a greater role for the private sector, and cost recovery to achieve systemic sustainability. The first two trends marched hand in hand, although the balance between them could vary: the state might still dominate in a nominally privatized economy, and commercial publishers might be active under centralized state procurement. The third trend, the drive for sustainability, was both a response to budgetary shortfalls in a period of rising demands and rising costs and a recognition that learning materials are a recurrent expense that cannot practically be financed by continual infusions of external loans and grants. Change was not always welcomed or rapid. As the authors of a World Bank review (1995, 13) observed:

In all countries entrenched ways of operating and vested interests will make change difficult. Education is intensely political: it affects the majority of citizens, involves all levels of government, almost always makes up the single largest component of public spending in developing countries, and involves public subsidies that are biased in favour of the elite. Prevailing systems of education spending and management often protect the interests of teachers' unions, university students, elites, and the central government rather than those of parents, communities, and the poor.


The trend towards decentralization in education had two motivations. One was the democratic belief that all stakeholders should be involved in a function so important to the community - that the wishes of teachers, parents, students, parent-teacher groups, and NGOs should be heard, and that some of these groups might be involved in decision-making. In its ultimate expression, responsibility would lodge in the schools, which would be able to decide for themselves how to use substantial portions of their allotted resources, sometimes with the advice of ad hoc committees of teachers, parents, and/or students (International Commission on Education for the 21st Century 1996).

The second motivation in many countries was the growing recognition that the existing system didn't work, and one reason was the cost and inertia of swollen bureaucracies. In the Philippines, for example, a Congressional Commission on Education complained that the Department of Education was "highly centralized ... and its complex procedures in asset management have resulted in loss and waste." It recommended that many responsibilities, including the procurement and distribution of textbooks and other essential materials, be devolved to the country's 126 divisional offices, bypassing entirely the 14 regional offices that lay between the divisions and Manila (Congressional Commission 1991, 39). Brickhill and Priestley (1993) found that "as a general principle, the further the decision-making process is from the actual user of the material, the less efficiently the funds are used."

For one reason or the other, governments decentralized the procurement of textbooks and/or their selection if there was a choice. The Russian Federation delegated responsibility for educational funding to the country's oblasts (provinces) in the early 1990s and gave teachers freedom of choice in textbooks. In the economically advanced post-communist countries of Eastern Europe as well, schools or local boards were given their own budgets and allowed to choose their own textbooks. Tanzania announced a policy in 1991 that would lead to selection at the district level and eventually in the schools. Mexico was to have completed decentralization by the end of 1997, at which time each of the 32 provinces would be in control of its schools, teachers, and budgets. Colombia devolved selection to the districts and schools, as did Kenya, Uganda, Zimbabwe, and some provinces of South Africa. By the end of the decade, India had almost completed devolving the publication of textbooks, teachers' guides, and work books to the state governments, which were expected to produce books at a more affordable price than the private sector had been able to achieve (Pinter 1999, 56; Stridsman 1999, 86; Noriega 1999, 127; ADEA/UNESCO survey 1999).

Not all governments by any means followed the trend or followed it all the way. In Singapore, for example, textbooks were published commercially but the writing of them remained with the Curriculum Development Institute. In Mexico, the government devolved administrative responsibility but not choice; it was reluctant to abandon the idea of "a single textbook, written and produced by an official body, because it has long been the means of conveying the government's desired message to the population and ... the record of national history" (Noriega 1999, 127). In Bangladesh, the National Curriculum and Textbook Board (NCTB) continued to develop the curriculum, write the textbooks, and produce the finished books, which it sold at a profit to the Directorate of Primary Education. The NCTB opposed any change to the system and, despite proposals by a consortium of funding agencies, the government also showed little active interest in reform (Sida 1996, 46-7).

Decentralization is seen to have several advantages. It encourages the involvement of parents and others in the community. It may give teachers more responsibility and status and, especially in large schools, develop their sense of collegiality. Most important, decisions can be taken in the light of local conditions and to meet local needs. They are no longer made by bureaucrats in a central office, divorced from the realities of the underfunded, often isolated, classroom. Local decision-making is particularly important for rural communities and for minority groups whose concerns, needs, and beliefs have often been ignored in educational books published in, and with examples from, large urban centres.

The transition is not without difficulties, however. Financial decentralization raises concern about money lost through incompetence, misappropriation, or corruption. Administratively, there may be confusion if the division of responsibilities is not clearly spelled out. There are often delays in the transfer of funds from the central to subordinate offices. Personnel at all the levels that are given new responsibilities, down to the school, require training in financial accounting, management of supplies, selection of books, and other administrative tasks - a massive task. In a strongly hierarchical system, teachers and others are apt to be influenced in their decisions by superior officers, undermining the intent of decentralization. If procurement as well as selection of textbooks is decentralized, retail outlets may not be easily accessible or may be late in obtaining the books for school use.


At the same time as they were moving responsibilities downwards, national governments were withdrawing from the business of providing textbooks and other learning materials. Government departments that wrote and produced textbooks as monopolies were turned into parastatal organizations, and monopoly publishing parastatals were sold, in part or all, to private owners. Commercial publishers were allowed to compete for textbook adoptions, against one another and, not infrequently, against a parastatal. State-owned printing plants were turned into parastatals and expected to run on strictly commercial lines; parastatal printers were privatized. The government often retained distribution of school books but, especially in the more liberalized economies, this function too might be given to the publishers. Liberalization was implemented in many patterns, but always in the expectation that commercially oriented enterprises would be more efficient than government and more responsive to educational needs, especially if there were competition in a multi-textbook system. Donors (Sida 1996, iii) and governments had come to recognize that state monopolies rarely met the need for learning materials, and that large-scale subsidies did nothing to encourage re-investment in publishing development.

In some countries, including Zimbabwe and Jamaica, learning materials had long been supplied by commercial publishers. In others that had had communist governments, commercial production and competition were new concepts to an entire generation. The transition could be abrupt. In the Philippines, a group of commercial textbook publishers and agents, who had been supplying the 5% of the student population in private schools, lobbied the Congress successfully in 1995 to achieve privatization of supply to the much larger and more scattered public system. The state was to get out of textbook writing, production, and distribution entirely within three years. The time allowed was inadequate, particularly when implementation was delayed by government inaction and arguments between the public and private sectors, and numerous abuses of the new system in its first year were reported (Chua 1999). Other countries approached the transition more cautiously, and many were still in the early stages at the decade's end.

Tanzania was one that took a deliberate approach. In 1991 it adopted a policy of transforming the state-managed provision of textbooks into a completely commercial system with minimal government involvement and decentralized selection. A Pilot Project for Publishing (PPP) was established with external support to advise and assist publishers and ministries on an orderly transition. The change was not without its difficulties, but by 1997 commercial publishers had produced 21 titles (1.75 million books) and another 28 titles were in progress. The first manuscripts were written in the Tanzania Institute of Education, but in 1996 the Institute was restricted to curriculum-related activities. The Ministry of Education continued to be responsible for procurement and distribution. The duration of the transition period has not been defined and will depend on how fast the actors can cope with new responsibilities, and some commercial publishers and booksellers feel the government should move faster in devolving spending power so as to allow full competition (Bgoya et al. 1997, 26-7; Stridsman 1999, 85-6; Sida 1996, Nyambura 1999, 13).

The beneficial effects of liberalization were to be seen in several countries. In Kenya, which already had a well established publishing industry with significant local ownership, schools were allowed to select the books they would use from up to six titles authorized for each subject/level by the Textbook Approvals Board. The new policies had resulted by 1997 in from three to five competing series in each of five primary-level subjects, and comparable choice in subjects at the secondary level except in physics, chemistry, and biology, for which books were mainly imported from the United Kingdom (ADEA/UNESCO survey 1999, Chakava 1997, 44).

In other countries progress was slow - outstandingly so in Ghana, where the government had entered into an agreement (heralded as a "unique, first-of-its-kind effort in all Africa") with five local publishers to co-publish 30 selected primary textbooks in 1984 but did not implement it until 1990. It was only three years later that the national publishers' association met for the first time with the Minister of Education to ask for more involvement. Since then the syllabi, once restricted, have become more publicly available, and in 1998 Ghana announced a more open

Table 7. Models of textbook provision in sub-Saharan Africa (not available)

textbook production after consultation with funding agencies, publishers, printers, booksellers, writers, and teachers (Crabbe 1999, 31-4; Nyambura 1998, 7). In Mongolia, the transition has been hampered by attitudes. The country has experienced and skilled people working in book production, but the business aspects of publishing are not fully understood, and publishers and printers are still primarily looking for funding and not for business opportunities (ADEA/UNESCO survey 1999).

Of 17 countries of sub-Saharan Africa surveyed in 1997 (Table 7), eight apparently continued to use the single textbook system in the publishing of primary materials. One, however, allowed a choice between state-published and privately published titles in some subjects, and three were in transition towards open markets and liberalized textbook publishing. Three countries offered a choice of parallel titles in every subject. Five countries gave limited opportunities for alternative titles in some subjects but not all. There was greater choice at the secondary level. Only Ethiopia and Mozambique apparently restricted schools to single textbooks; other countries had either parallel or, the majority, multiple choice (Bgoya et al. 1997, 38).

At the end of 1998, however, relatively entrepreneurial Cameroon moved away from multiple choice. It signed an agreement granting a newly privatized company, CEPER, S.A., the exclusive right to conceive, produce, and distribute all primary and nursery school textbooks in the country. It also exempted CEPER from all taxes on machines, spare parts, paper, and ink. CEPER is the successor to the state-owned Centre d'edition et de production pour l'enseignment et recherche which had specialized in educational materials but shared the market with commercial companies, including the well-established Editions CLE (Tshimanga 1999, 1-3). For the most part, the state kept some functions and turned others over to the private sector. In Zambia, manuscripts prepared under the Curriculum Development Institute were tendered to the private sector; the finished books were still being bought centrally in 1996 but there were plans to delegate that responsibility to school management boards (Munamwimbu 1996, 68-9). In Lebanon also, a state institution wrote the textbooks, which were then produced and distributed by private publishers (ADEA/UNESCO survey 1999). Several Eastern European countries have retained distribution as a governmental function, even though books may be produced in a free market. Occasionally the government has reclaimed responsibilities: Jamaica, which had previously allowed individual secondary schools to order books from their supplier of choice, is moving towards centralization of procurement in order to reduce costs.

In small countries, and in those with languages spoken by only a small population, a continuing state presence was inevitable. Commercial publishers cannot be expected to mount a sales force in countries as small as Niue, which has two schools, or Tokelau, which has three. In those Pacific island countries, central procurement remains the only feasible option. Mauritania encouraged private-sector involvement in publishing but the market was so unattractively small, as a result of population size and poverty, that the state was obliged to write, print, and distribute textbooks; even an attempt to involve commercial booksellers in distribution failed (British Council 1992, 11). In other small nations, state involvement was by choice. Guyana, with a population of roughly 800,000 persons and English as a language of instruction, might have used some textbooks produced commercially elsewhere but instead secured external support to write and produce its own primary- and secondary-level textbooks in a government centre.

Where state-owned or parastatal organizations had been forced give up monopoly positions, they frequently continued to dominate the market. As Brickhill has noted (1996, 21), it is difficult to break up a monopoly and establish a level playing field when books are needed immediately and not in two or three years or more. The former monopolists benefited moreover from the conservatism of procurement officers and teachers who hesitated to stray from the former official textbooks. Some parastatals, while autonomous and commercial, still received government funding or hidden benefits such as the use of state-subsidized printing plants. The new publishers did not always prosper. Following liberalization in Kazakhstan, 150 private publishers registered for business in competition with 14 state-owned publishers, but soon after only 12 had survived and only five or six were effectively functioning (Askerud 1996, 100).


In a time of rising enrolments and financial austerity, no developing country was allocating more than 1% of its primary education budget to textbooks, even though an increase of only 1 or 2% in most countries would have ensured a textbook:pupil ratio of 1:1 (Askerud 1997, 18). Public recurrent expenditures in education were swallowed by teachers' salaries. In a survey of sub-Saharan Africa (ADEA 1999, 31-2), only a handful of countries reported what proportion of recurrent educational expenditure was spent on learning materials in 1995 or 1996, but those statistics are gloomy. In primary education, only Mali, at 14.9%, was shown as devoting more than one-tenth of annual recurrent educational expenditures to learning materials; from there the proportion dropped to 8.4% in Comoros, 5.1% in Gambia,1.7% in Seychelles, and so on down to 0.2% in Uganda. The reported statistics for secondary education were somewhat higher: from 28.3% to Mali and 8.5% in Togo down to 2.4% in Gambia and 1.4% in Mauritania.

Ministries of education were apt to find their legislative masters allocating only a portion of the budget that had been requested to meet stated objectives of free or subsidized textbook provision. In Indonesia, the books needed to achieve a textbook-to-student ratio of 1:1 in grades 1 to 9 would have cost an estimated US$380 million in 1995, when the economy was still buoyant; the amount actually spent amounted to US$36 million budgeted by the government and about US$43 million paid by parents - a shortfall of almost 80% (ADEA/UNESCO survey 1999). One respondent with widespread experience observed that the shortage of textbooks is not so much a book famine as a money famine: there are many good quality textbooks available in most subjects in many countries but no funds to buy them, and classrooms are generally short of everything but students. Governments admittedly must make difficult choices. Another respondent, this time from the Philippines, recalled that the chair of a congressional committee asked publishers appearing before it: "How can we even think of books when people are hungry and dying?"

In countries that were moving from state or parastatal publishing to private-sector publishing of textbooks, there could be uncomfortable surprises. In principle private enterprise is more efficient than government and should be able to produce textbooks more cheaply. But privately owned publishers must recover from sales all their overhead costs - most significantly, wages, premises, office expenses, manuscript development, marketing, and interest charges on loans - as well as the costs of production and a reasonable return to shareholders. Ministries that were engaged in publishing normally included only production costs and perhaps some wages in calculating the cost of school books, but they did not always account for those expenses in full and sometimes enjoyed the privilege of state-subsidized printing or tax-exempt paper. Parastatals did not always pay full market costs either. In asking for budgetary allocations, ministries and parastatals used their own calculations to project the annual cost to the government of learning materials. Those that sold books also often simply charged only the production cost. In contrast, commercial publishers who enjoyed none of the hidden subsidies usually multiplied the cost of production (paper, printing, and binding) by three or more to set a retail price that would give them a profit and allow a markup for booksellers. The markup for booksellers could be waived for central procurement by the government, but the price per copy was still likely to be more than double what the Ministry had been accustomed to allowing in its budget estimates. The fiscal allocation would then have to be doubled or more to maintain the same level of provision. If the legislature rejected that option, the number of books that could be bought by the state would be reduced by half or more. Savings could be realized only if the transition to the private sector were accompanied by the appropriate reduction in the civil service establishment, including any senior staff who were writing textbooks or managing publication.

Despite financial stringency, some countries continued to provide textbooks without charge, even if in reduced numbers. Sri Lanka, Nepal, Bhutan, and Bangladesh, for example, all provided free textbooks to primary schools, and respondents to the ADEA/UNESCO survey from those countries reported no book shortages. In countries where an international funding agency was subsidizing textbooks for specific areas (for example, the poorest provinces) or specific groups (girls or indigenous minorities), free distribution was sometimes maintained generally because it was felt inappropriate, if not illegal, to provide textbooks to the rest of the country under less advantageous conditions. It seems likely that free provision will gradually grow less common. Bhutan is gradually withdrawing support for textbooks and stationery from the economically better-off areas because of rising costs. The Cambodian government is considering selling books for grade 10 (basic education ends with grade 9) and may legitimize the sale of core textbooks in lower grades to undermine the black market. Bulgaria suddenly accelerated the introduction of total cost recovery from parents after a World Bank-funded program ended; what had previously been regarded as politically unacceptable and unaffordable suddenly became policy from economic necessity (ADEA/UNESCO survey 1999). In the short and medium term, however, governments continue to bear most of the burden of supplying books.

Where textbooks are produced commercially and sold on the open market, there should, of course, be full cost recovery with enough profit to encourage re-investment in further manuscript development and revision. If necessary, the government may subsidize the publishers.

Where books are not sold, there has been growing attention to at least partial cost recovery through textbook rental. Book banks of this type were established at all levels of education, from primary to tertiary.

There is some evidence that parents prize education so highly as a long-term investment in the family that household demand for education is relatively price-inelastic. That is, enrolment does not drop in proportion to increases in private costs (World Bank 1995, 106). On the other hand, the elimination of school fees in Uganda led to an increase in enrolment of more than 2 million students (Watkins 1999). In Samoa, parents grumble at paying school fees of three to five tala per term, less than the cost of one packet of cigarettes there, and schools, which have difficulty collecting that amount, hesitate to charge more (UNESCO/Danida case studies: Samoa). Unquestionably, when textbooks are added to the direct costs of education - which may also include exercise books, pencils, uniforms, and transportation, even when the schooling itself is free - children from the poorest families are going to be discouraged from attending. In many countries, girls are more likely to be affected than boys, for a variety of reasons: a cultural belief that girls do not need as much schooling as boys because they will spend their lives in domestic tasks, dependence on girls to help with household chores, the extra cost of clothing for girls.

To compensate, various demand-side support programs have been devised to encourage attendance by girls, including free textbooks and stipends that can be paid to parents to compensate them for direct costs and for the loss of their daughters' time. Such projects have been undertaken, for example, in Bangladesh, Egypt, Guatemala, Mali, Morocco, Peru, Tunisia, and Yemen. In the Pakistani state of Balochistan, funding for the purchase of textbooks has been part of a rural program based on community support for schools. In Armenia, where schools operate book rental schemes with a four-year amortization, funds were provided for subsidies to the neediest 10% of students, identified by the school community. A similar scheme has been under way in Guinea. There the book rental fee of approximately US$1.40 per year can be waived if, in the opinion of the school management committee, a student cannot afford to pay it. Funding agencies have also targeted their subsidies to education by directing loans to the most disadvantaged areas of the recipient country. World Bank projects of this kind include support of primary-level textbook development and publishing in six provinces and autonomous regions of China and in ten states of Mexico, both involving the development and publishing of textbooks and other learning materials in indigenous languages that had previously been underserved (World Bank 1995, 115; Patrinos and Ariasingam 1997, 21-3, 26-8).

In private schools, textbooks have always been bought by students or families at market rates. The number of private schools varies from country to country and probably reflects to some degree the perception of the public system and the quality of its education. They sometimes play a prominent role. In the Dominican Republic, they have served about 20% of all primary students and 30% of lower-secondary students. In Colombia, private secondary schools outnumbered public ones almost two to one in the 87 municipalities covered by a World Bank project. Private schools normally were located in urban centres and catered to the elite or at least well-to-do. In the second half of the 1990s, the World Bank and other funding agencies began financing projects to increase access to the private system. One method was to issue vouchers for tuition and incidental expenses to needy students. Another was direct funding of private schools serving poor children, as in the Dominican Republic, or serving primary-age girls, as in the Northwest Frontier Province of Pakistan (Patrinos and Ariasingam 1997, 29-35).

Table 8. Projected increases in enrolment worldwide (in millions) not available
Availability of school books

Despite public statements and funders' support, and advances in some countries, it does not appear that textbooks are much more available globally ten years after the Jomtien Declaration than they were before. Quantitative evidence is scarce, but qualitative reports suggest that, if anything, the situation in many countries has got worse. School enrolments have risen, a result of the demographic curve and national attempts to achieve universal primary education. The number of students attending school worldwide increased by some 50 million between 1990 and 1995, and some 60 million additional children were expected to enter the primary school age group during the next five years (Jolly 1996). The projected enrolment increases spread across the three educational levels (Table 8) and were expected to be sharpest in sub-Saharan Africa, the Arab States, and Southern Asia. At the same time, funding for education has declined as a result of economic slowdown, the cost of servicing foreign debt, devaluation, policies of structural adjustment, and competing social needs. Following the collapse of communism, educational systems in Eastern Europe and Western Asia have fallen into disarray. In countries with large populations and a history of external textbook support, such as the Philippines and Bangladesh, the international banks are restricting educational loans to the most disadvantaged areas, leaving the majority of the school population to the vagaries of national legislative funding.

In a survey that covered 26 countries of Africa, all but nine of 103 respondents considered that the book famine in their country was serious, and 87% said there was a serious shortage of textbooks (ADEA/UNESCO survey 1999). Respondents to the same questionnaire elsewhere reported serious shortages of textbooks in some grades in Cambodia, for the new curriculum in Mongolia, and more generally in Afghanistan, Indonesia, the Philippines, Lebanon, and Pakistan.

There was not total unanimity, however. The respondents from Lebanon and Pakistan differed among themselves, half saying there was in fact no shortage of textbooks. Such differences of opinion underline the difficulty of judging availability when reliable data are scarce. Oliveira (cited in Askerud 1997, 35) said that

of the 93 countries responding to UNESCO inquiries on the topic, roughly 60% of them (about 57 countries) report that they have a sufficient supply of books; however, several of this group, including Paraguay, Burkina Faso, and Colombia, in fact suffer from a major scarcity of textbooks. This finding not only illustrates that data are not reliable; it also shows that different countries have different opinions about what amount of textbook coverage is sufficient. Moreover, the numbers given above usually only refer to delivery of textbooks to warehouses, not to what is actually available in schoolhouses in dense urban periphery neighbourhoods or remote rural villages. According to these same figures, twelve countries report that less than 30% of their schools are supplied with textbooks. In addition, there is extensive evidence that the data regarding textbook use supplied by these and by many other countries are unreliable.

Data from other sources ... are hardly more reliable, since ultimately, they came from the same informants who supplied UNESCO with their education statistics: the Ministries of Education.

Askerud concluded that during the last 20 years the ratios of books per pupil had fallen in most developing countries and that for the majority of the world's students, access to basic tools of learning is so limited as to constitute a major crisis.

A pilot survey of schooling conditions in 14 of the least developed countries of Africa and Asia, conducted by UNESCO and UNICEF (EFA Forum Secretariat 1996c), found that half the pupils had no textbooks.

The survey paints a dismal picture of how classrooms are equipped, making it difficult to imagine how learning can take place. Rural schools generally fare worse than urban ones for learning materials. Final grades tend to be better equipped than the early ones, although conditions are far from adequate. The study found that half of the children have no textbooks. In none of the countries surveyed does every classroom have a usable chalkboard. Very few countries have a world map. No country has all classrooms equipped with a teacher chair and table. Not all children have something with which to write. Ten out of the 14 countries had virtually no supplementary readers, and most countries have no classroom library. These findings give a picture of the gap between official requirements and the actual situation. Most countries have a ministerial norm that there should be one chalkboard, a teacher's chair and a table in each classroom, but these norms were rarely met by more than a small number of schools.

There were pockets of advance, or at least stasis. In Cambodia, for example, the situation improved as a result of external support for the production and free distribution of textbooks for grades 1 to 9 in the four core subjects: Khmer language, mathematics, science, and social studies. Books for grades 10-12 were also being developed and produced in a five-year project. Textbook:pupil ratios were gradually increasing to the target of 1:1 throughout the basic grades (ADEA/UNESCO survey 1999). In Mozambique, too, it was possible for every primary school pupil to have his or her own books as a result of a World Bank loan and subsidies from the Netherlands and Sweden. It was uncertain, however, whether that situation would continue (Sida 1996). In an isolated barangay of the central Philippines, 7 km from the nearest town, reached by an unpaved narrow road over hills, visitors to a two-room multi-grade school found an adequate supply of books for the different grades, although they were old and worn (Miguel and Barsaga 1997, 124). In Egypt, the Ministry of Education gave students one textbook and one activity book each year to keep, reprinting each year to meet new demand (UNESCO/Danida case studies: Egypt).

Kazakhstan almost tripled its annual expenditure on textbooks between 1996 and 1999. As a result, 83% of children in grade 1 and 70% of children in grade 2 are reported to have been supplied with a new generation of textbooks that follow post-communist child-centred methods of education. Samplings suggest comparable provision at some other grades. In some areas rural schools are better supplied than urban ones. Textbooks are provided on payment of an annual rental fee from grades 1 to 9, except for the grade 1 alphabet book, which is free. The rental fees average about 3% of a monthly family budget, and are waived for needy families. Students in grade 10 and 11 must buy their books. There is a plan to sell textbooks in the future through the retail market, but needy families would still obtain them without charge (UNESCO/Danida case studies: Kazakhstan).

A survey of six sub-Saharan African countries produced what appear to be favourable relations between target and actual textbook:pupil ratios (Table 9) but the authors (Bgoya et al. 1997, 42) describe the data as dubious, because there were rarely any statistics or systematic recording of the situation in schools; they suggest that the most reliable number is that from Zimbabwe, where all textbooks are published commercially and the Ministry has no vested interest in estimating distribution. At mid-decade, spending on education in most sub-Saharan countries had regressed and the shortage of textbooks was said to have reached crisis proportions (Altbach 1995, 179). In Zambia, the schools had been without sufficient new books for 17 years (Munamwimbu 1996, 68). In Ghana, books that had been purchased for the government for free ...

Table 9. No. of subjects and textbook to-pupil ratios in some sub-Saharan African countries (not available)

...distribution were found in the marketplace (Crabbe 1996, 39), but losses through theft were not unique to that country. In many, textbooks were stolen from government stores or taken from the printers. In Ethiopia, textbooks sometimes reached remote schools before urban schools - an unexpected reversal of the norm - because shipments for the urban schools had been diverted to the black market (Ambatchew 1999).

UNESCO/Danida case studies in three African countries also found a continuing shortage of textbooks. In Sénégal, the Ministry of Education decided in 1992 to stop the free provision of books and now makes textbooks available to children at comparatively low cost; efforts are made to keep the price lower than commercially produced books. The textbook:pupil ratio in the schools surveyed ranged from 1:2 to 1:10 for the French reader and from 1:5 to 1:10 for mathematics. Textbooks in science, history, and geography were only negligibly available. In general, girls had fewer books than boys (UNESCO/Danida case studies: Sénégal). In Tanzania, book shortages were reported in all subjects in the schools surveyed. Books were given free, lent by the school, or bought from shops that might be 50 km or more from the school. Schools received books at varying times, most often after the beginning of the school year. In two urban schools, a textbook:pupil ratio of 1:1 was reported in mathematics, science, and English, but in another school there were no grade 1 textbooks. In Grade 4, the ratios ranged from 1:1 to 1:20. Some schools had no books in the national languages. Books were most scarce in the rural schools (UNESCO/Danida case studies: Tanzania). Guinea introduced a book rental scheme in 1997, after an unsatisfactory experiment with private-sector provision. Rental schemes are administered by local five-member committees. The distribution of books is not always timely and some of the books are inappropriate for the pupils concerned. Many parents, moreover, are reluctant to rent textbooks for their children because they do not want to incur the possible cost of replacing any that are lost or damaged. Of textbooks that are available in the four core subjects, only a fraction were found to be actually in use. In one school, for example, only 39% of the available textbooks in French were rented, 44% in science, 41% in mathematics, and 26% in civic education (UNESCO/Danida case studies: Guinea).

In Latin America, educational spending emphasized higher education and quality education for a few at the expense of primary education, and especially primary education for the urban poor and rural population (Chapman et al. 1997, 151). In the early years of the decade, textbook coverage was reported to average 32% in Chile, Colombia, Bolivia, Ecuador, Panama, Peru, and Venezuela. About 76% of students in grades 1 to 3 had one primary or language/art book; about 30% had a mathematics book; and less than 10% had a third book. For grades 4 and 5 the numbers dropped by 20%. Columbia, Ecuador, and Guatemala presented an even bleaker picture (cited in Askerud 1997, 35). Even middle-income countries such as Brazil, Uruguay, and Venezuela had failed to maintain a consistent supply of textbooks. By the end of the decade, however, textbook distribution had improved in some countries at least. In Chile it was universal at both primary and secondary levels as a result of World Bank assistance, successful international competitive bidding, and negotiations with publishers in Spain and Mexico as well as Chile. Students had low-cost textbooks, 200 to 300 pages in length, with colour illustrations and good paper, at an average cost of US$3.

In a country as large as India, as might be expected, conditions varied greatly, as exemplified by three UNESCO/Danida case studies. In a survey of grades 1 and 5 in nine primary schools in Delhi, the capital, 100% of the students were reported as having textbooks, although books were in short supply in one of the schools in grade 4. Books were supplied free of charge in six schools in lower economic districts, and to 70% of the poorer students in a seventh school. Other children bought their books. Teachers were not provided with textbooks, however, and normally borrowed a copy from a student during classes (UNESCO/Danida case studies: India 1). In the state of Orissa, in contrast, learning materials were inadequate in all nine of the schools surveyed. Books were supplied free to pupils from Scheduled Castes and Scheduled Tribes, but in quantities well below enrolment. General students bought books on the open market, mostly from teachers who obtained them from outgoing students. Only two of the schools had science kits, and in most schools library books were locked up except during school inspections. Only one of the schools had a lavatory or urinal for children, and only one had even part-year access to safe drinking water (UNESCO/Danida case studies: India 3). In the state of Maharastra, the State Textbook Bureau provides textbooks, but in practice much of the demand is met by private publishers. Supplies are adequate in urban schools and in schools that are considered average to good. In schools that are considered weak, and in schools in rural areas, there are substantially fewer books. The State policy is to supply books without charge to children in the most disadvantaged groups, but books are provided only to selected classes each year and often arrive more than one month after the beginning of the school year. Illiterate parents oftenare sold books that are unsuitable for their children's age group (UNESCO/Danida case studies: India 2).

In Central Asia, the shortages ranged from significant to critical. Supply was impeded by lack of funding, bureaucratic inertia, inexperience in operating under the post-communist economy, reliance on imports for raw materials, and outdated printing equipment. Totally new textbooks were required in the light of revisions of the curriculum to suit new political and social realities and the displacement of Russian as the official language of instruction. In Central and Eastern Europe the situation varied from country to country. In some, new books were scarce and Russian books from before 1990 were still in use. Estimates of the current textbook shortage in Russia itself begin at 100 million copies and go up from there. In 1996, a working document (EFA Forum Secretariat 1996b) prepared for the Mid-Decade Meeting on Declaration for All could report that

The majority of developing nations have responded to the challenge of the Jomtien Conference by taking various actions to: evaluate their existing education systems; design policies to improve basic education; strengthen the capacities of people and institutions to manage education; create a more supportive policy environment; broaden partnerships and diversify the resource base; and explore ways to use traditional and modern channels of information and communication to provide basic education.

But at the end of the decade, a survey by Oxfam International (Watkins 1999) was less positive in its finding:

The limited progress towards basic education achieved during the 1990s points to a wide discrepancy between international declarations of intent and substantive planning for change. In the five years after Jomtien, over 100 "education for all" strategies were drawn up by governments, often in cooperation with donors. At the same time, the World Summit for Children, the Social Summit, and the Beijing Conference on Women all generated their own national planning exercises. Some success has been achieved. For instance, the national plans of action drawn up after the World Summit for Children have resulted in major advances in the areas of child and maternal health. In education, by contrast, national planning has generated limited progress. All too often, it has been characterised by vague statements of intent, empty promises, and a failure to mobilise resources.

A senior official of the World Bank remarked more succinctly that it was "particularly distressing to see how easy it is for governments to articulate a policy as an objective and not fulfil it" (Sosale 1999, 159).

Quality and use

The quality of textbooks continues to vary. It remains difficult in most countries to find good authors, and many books continue to be written by subject specialists or curriculum developers in the Ministry of Education or one of its agencies. Many such authors were still oblivious or indifferent to the experience and abilities of the intended users. A respondent to the ADEA/UNESCO survey described the government-produced textbooks in Pakistan as by and large pedagogically unsound, sometimes filled with mistakes, and usually very unattractive, in contrast to the books published by transnationals that are used in the private schools. Another said that student performance in that country in science and social studies is generally low because of the shortage of good books.

In systems with multiple choice of textbooks, there have been complaints, probably inevitable, about the process of evaluation and approval. In Kenya, the government was accused of favouring the former state publishing house (Bgoya et al. 1997, 35). In Romania, officials in the educational system are free to become authors of competing textbooks, leading to concerns about conflict of interest and influence over subordinates (Papadima 1999, 134-5). In the Philippines, publishers complained about inconsistency in appraisal: according to one, some books won conditional approval, even though they did not cover all the required curriculum, while others were rejected over an illustration or exercise (Chua 1999). During the rapid adoption of new curricula and textbooks after the change of government in South Africa, textbooks, teachers' guides, and work books were sometimes evaluated separately and a teachers' guide might be approved while its companion books were rejected.

At the end of the decade, physical quality of textbooks was still poor where printing equipment was outdated and in need of repair, as in Mexico (Noriega 1999, 128) and much of Central Asia, or where paper of good quality was scarce or expensive. Adhesive binding continued to give way and books fell apart. Printed appearance improved in many countries with the spread of computer-assisted composition (also called desktop publishing, or DTP) and offset lithographic presses. In replacing traditional forms of typesetting, DTP also speeded the development of new textbooks and the revision of existing ones, although in the hands of undertrained operators it also opened the way to poor typographic design and illustration.

Increasing attention was paid to gender balance, but stereotyping persisted. A science textbook in Pakistan had a page illustrating the functions of various parts of the body: the function of the head, a woman's, was to carry a water pot. A survey in Mozambique, Zambia, and Zimbabwe (Brickhill et al. 1996, i-iv, 9) found that women were under-represented in all the strategic points of textbook publishing and provision - that is, in writing, commissioning, and publishing manuscripts, school procurement, and bookshop stocking. In all three countries the authors found evidence in textbooks of gender stereotyping, especially in the portrayal of adults. In Zimbabwe where, unlike the other two countries, textbooks were produced entirely in the private sector, they found "a continuous thread" of "gender stereotyping and prejudice against women in virtually any roles outside the home." Efforts had been made in Zambia at role reversal, role balance, and assertiveness of female role models, but stereotyping still appeared in illustrations through order of appearance, role representation, and size, voice, and posture.

In the Amman Affirmation at the end of its Mid-Decade Meeting in June 1996, the International Consultative Forum on Education for All acknowledged that progress towards closing the gender gap in education had been excruciatingly slow. It also stressed the essential role of the mother tongue for initial instruction and in that respect, at least, some progress was being made. In Ethiopia, 11 national languages were used in instruction at the national level. In Bolivia, classrooms were allocated books in Quechua, Aymara, and Guarini as well as Spanish. In Mexico, an ambitious educational project in 10 of the poorest southern states included bilingual textbooks and learning materials for indigenous schools (Wassie 1996, 53; Sida 1996, 52; World Bank 1995, 149). Textbooks in South Africa are published in 11 official languages. There a large commercial house, Juta, launched an innovative series of learners' resource books, activity books, and teachers' resource books in five African languages and were developing them in four more, using stories, folk tales, and incidents common to the readers' experience. In the South Pacific, the difficulties of providing learning materials in local languages for small populations was being overcome through regional cooperation, as described in Part IV.

But the colonial languages have proved tenacious. In the Sahel countries of Burkina Faso, Benin, Niger, and Sénégal, the state distributed free textbooks in French, bought on a large scale through international competitive bidding, and the national languages were taught only at an experimental level, with no systematic production of materials in them. Indeed, in Burkina Faso there had been a step backwards; the national language was the sole language of instruction until 1985, when the government changed, and only since 1995 has it been gradually reintroduced (Jung 1999, 27-8). Publication in national languages can prove difficult, even controversial, if there is no agreement on orthography. In South Africa, where approval of textbooks is under provincial jurisdiction, a book in Zulu, for example, might be approved in one province and rejected in another because of its spelling.

European experiences and values also persisted. In first grade in Thailand, students learned that "Ama took an apple to school", even though the only apples were expensive imports. In Nepal, "Sushilla took a bus to the zoo" in fifth grade, although 60% of children living in the mountains had never seen a wheel. In Côte d'Ivoire, primary texts were full of French markets and "my time in Paris". Such curricula alienated the majority of children from their culture and communities (Bennett 1993, 42-3).

There was still no guarantee that teachers would use the books that were available, or use them effectively. Teachers did not always have an accompanying teachers' guide. Sometimes they did not know how to teach when every student had his or her own book; sometimes they were overly protective of the books; sometimes they felt the books provided were not appropriate or relevant to the needs of their pupils (Sida 1996, 17). In the increasing number of systems where parents were expected to buy textbooks, some families could not afford the expense, although some countries had programs to assist their poorest communities. In systems with free distribution, teachers might still be personally responsible for replacing any textbooks that were lost or damaged, and therefore kept them locked up. On the other hand, even when such a policy changed, as in the Philippines, many teachers apparently did not believe that they would not be held responsible for losses or simply were not told of the new directive. Where multiple choice was a new concept after generations of single textbooks, teachers proved unexpectedly resistant to the idea of changing their habits. In Romania, both teachers and parents interpreted the new alternative textbooks as auxiliary to the old "official" one, and some feared that alternative textbooks would create chaos in the classroom (Papadima 1999, 134).

Educational materials

In most schools, the textbook and chalkboard were still the primary educational aids. Teachers relied heavily on the textbook, especially in science and mathematics, subjects in which they themselves were apt to be weak. In some jurisdictions, teachers were given lesson plans that they were expected to follow, leaving little scope for creativity or innovation. In the eight countries surveyed in the UNESCO/Danida case studies, only schools in India and Jamaica reported the availability of electronic aids such as television, tape recorders, and computers. All other schools depended on charts, chalkboards, slates, mathematics sets, books, or aids made by the teachers themselves. India began a country-wide program in 1986 to provide schools with a list of materials, including maps, globes, science kits, and flannel boards, but not all the listed materials were received. Classes visited in India under the UNESCO/Danida case studies were typical of many in developing countries in that:

teachers read aloud to the class from the textbook

teachers asked a few good students to read aloud

teachers asked questions related to what was read aloud, but answered the questions themselves

teachers wrote material on the chalkboard for children to copy

children copied what was written on the blackboard

children did sums on their slates In one class, the teacher was unaware that the children could not find the correct page in their textbook.

In a sample of Jamaican schools, most materials used in the classroom were made by the teachers themselves. Most teachers were trained to use the materials and about 80% of those observed used them appropriately. About three-quarters of the teachers included teaching aids in their lesson plans, and in all but a very few classes the children understood the materials. The materials were more likely to be used if they could be stored in cupboards in the classroom. Most, however, were kept in the head teacher's office (UNESCO/Danida case studies: Jamaica).

The use of educational materials was part of pre-service teacher training in all eight countries in the UNESCO case-study survey, but only in Jamaica did teachers feel satisfied that the training was adequate. In Kazakhstan, teachers were receiving intensive training in the use of a new generation of textbooks from teams with expertise in personal and social psychology, educational theory, and curriculum and textbook practices. Authors played a crucial role in the process, visiting schools and demonstrating activities (UNESCO/Danida case studies: Kazakhstan). In another country, in contrast, there was no provision for training in the use of educational materials to teach a revised curriculum, the responsibility being left to school inspectors and the Institute of Education.

In some rural schools, students were encouraged to collect seeds, grain stalks, match sticks, and other items that could be used as teaching aids in mathematics. Children also helped to make aids such as word cards, picture cards, number plates, abacuses, clock faces, balls, and pots. Girls generally were more eager to engage in these activities than boys except in Jamaica, where the reverse was true, and in Tanzania, where there was no obvious difference between the genders. In other schools, teaching aids were either completely absent or inadequate in quality and quantity. In some cases they were simply decoration, usually placed above the students' eye level (UNESCO/Danida case studies).

A village teacher in Maharastra, India, provided an example of innovative use of low-cost teaching materials. Vaishali Pawar has conscientiously attended in-service training offered to rural teachers in her area by an NGO. Her classroom has low blackboards on which the pupils can practise handwriting or draw pictures. A few pictorial charts hang at the children's eye level. To supplement reading materials, Vaishali has drawn charts that are large replicas of pages in the grade 1 language book. Her drawing is not the best, but the charts have child-appeal and she has found that beginners who have problems with the size of the text in their primers are encouraged to read on the big pages. To teach mathematics in her multi-grade class, she uses graded work cards. While she teaches the grade 1 children, the grade 2 pupils use the cards individually or in pairs to practise newly learned skills. They thus become actively involved in learning and develop independent study habits (UNESCO/Danida case studies: India 2).

Some countries were already actively exploring the potential of computers for education early in the 1990s. By 1991, Argentina, Belize, Brazil, Chile, Colombia, Grenada, India, Korea, Malaysia, Mexico, Peru, the Philippines, Sénégal, Sri Lanka, Swaziland, Thailand, and Trinidad and Tobago had attempted to use computers in at least some schools. India, Mexico, and Tunisia were teaching computer literacy at the primary level, and Argentina, Brazil, Kenya, and Sénégal were using computers to develop critical thinking, creativity, and problem solving. Costa Rica promised to put computers in half of all primary schools (Lockheed and Verspoor 1991, 53; Jolly 1996).

Well designed computer programs had been proven effective: a test in the Philippines found that in 100 hours grade 1 students could learn to read and write English at a level achieved by only a fraction of children in conventional classes at the same school. However, research in North America had already concluded that computers were less cost-effective than other equally effective methods of teaching (Lockheed and Verspoor 1991, 51). In any event, countries that could not afford to provide textbooks were hard pressed to find the money for computers or simpler electronic equipment such as videorecorders. And at the end of the decade many - perhaps most - of the schools in Africa and South Asia still lacked reliable electric power to run computers, much less the telephone lines to connect with the Internet. Fewer than 10% of schools in Central Asia had computers. Even in Eastern Europe, where computers were somewhat more common, most were old and could not support CD-ROM or Internet applications (Pinter 1999, 58). In a positive contrast, Barbados gave national priority status to installing multimedia facilities in all its schools and was considering including software among the materials available under its well-established textbook rental program.

Supplementary reading and libraries

In general, supplementary reading materials continued to be scarcer than textbooks. As the decade came to a close, school libraries were said to have the lowest of priorities in educational spending. The majority of schools possessed no library. Where some semblance of a school library did exist, it was often no more than a few shelves of outdated and worn out material, inadequately staffed. A respondent in Bangladesh reported: "Most of the students do not read books other than prescribed books. The curriculum and learning process in primary education do not encourage students to read or consult books other than the textbooks. There are no libraries attached to the schools or near them" (ADEA/UNESCO survey 1999). A study in Samoa (UNESCO/Danida case studies: Samoa) found that in nine of the ten schools examined there were few books of any kind, whether instructional or recreational readers, textbooks, or teachers' guides. The books that were there tended to be library books donated from overseas (many of them outdated or inappropriate to local culture), instructional readers from New Zealand, and a small number of readers produced locally in the early 1990s. It appeared that the few available books were rarely used. Books were found on staff room shelves, in locked cupboards, and still in packing boxes. Many were kept in the principal's office, and access to them was not easy. Teachers relied instead on the blackboard, charts they had prepared themselves, and self-produced books. Children sat on woven mats and read together as the teacher pointed to each word. The rural schools had grid wire windows and on wet, windy days learning material on the walls was ruined.

In Southern Africa, better-than-average libraries were usually in secondary schools and heavily reliant on parental support. In Mali, a school library might consist of only 50 to 100 books, including textbooks, in a cupboard in the principal's office; most secondary schools had libraries but lacked suitable premises and budgets for them. In South Africa, the Department of Education and Training, which was responsible for the education of black Africans, did not officially acknowledge the role of school libraries until 1983. In 1997, the Education Foundation found that only 17% of schools in South Africa had libraries and they were concentrated in white and Indian and, to lesser extent coloured, schools (Radebe 1998, 43-4; Sidibé 1998, 23). In countries such as Mozambique and Cambodia, schools were destroyed and with them any libraries they had. In Ethiopia, librarians were held directly responsible for any loss of books and were accordingly reluctant to lend them (Ambatchew 1999).

School libraries and supplementary reading materials received growing attention during the decade, however. As the post-Jomtien era opened, the World Bank had at least 14 projects involving primary or secondary school library development and at least five more at the tertiary level (Read 1992, 322). Indonesia had 117,000 school libraries with a total collection of about 106 million books, an average of 900 per school (Taryadi 1995, 482). In the course of the decade small school libraries were centrally procured and distributed in Bolivia, Chile, and Peru (López 1999, 138). In Egypt, a survey of 14 urban and rural schools of varying standards found that students were encouraged to use the school library and learning materials, and most of the schools had a computer and several software programs (UNESCO/Danida case studies: Egypt). The Carnegie Foundation was about to seek opportunities for strengthening libraries in selected African countries (Priestley forthcoming).

Cambodia, whose educational system had been ravaged by civil war, was developing supplementary materials for the schools, often with support from NGOs and other international agencies. Up to 100 titles were involved, but some were simply copies of international children's books with Khmer translations pasted in by hand, and distribution tended to be concentrated in core schools in certain provinces. Reference books were almost non-existent. Two educational magazines aimed at children were fairly available in the schools and at bookstalls in Phnom Penh, however, and a few schools had small libraries, especially at the upper secondary level (ADEA/UNESCO survey 1999).

A study of African school libraries (Rosenberg 1998) reported three significant forms of initiative:

Mali began a "library in every school" campaign with initial support from the French Co-operation Mission. The pilot project involved ten schools, with twenty more to follow out of the nearly some 250 public schools that offered nine years of education. For each library, the local community was required to provide appropriate premises, 20 chairs, two reading tables, and two sets of shelves, and to nominate a teacher to act as librarian. The government would then train the librarian and provide 225 books. Local communities could buy additional books, possibly by raising money in the cities. The government coordinated the network of libraries (Sidibé 1998, 24-7).

The READ Trust in South Africa, founded in 1979 by a small group of volunteers, had grown to a significant trainer and supplier of materials. The Trust trained teachers in librarianship and in ways to encourage literacy and writing skills. It produced attractive, illustrated books, posters, and theme packs designed to dovetail with the curriculum. It put together and distributed classroom libraries of about 60 books - colourful and exciting fiction, non-fiction related to the curriculum, and a few reference books such as a dictionary and atlas -stored in a lockable wooden box that doubled as a display cabinet. The libraries could be replenished annually. Students were encouraged to use the books, first in silent reading lessons, then in any free time during school hours, and eventually to take home and share with the family. In reading and writing, children in READ-equipped classes outperformed control school counterparts by as much as 189%, and were ahead by 18 months in reading scores and two years in writing scores. An average of 15 boxes per school cost about 20,000 rand (US$4,700), far less than provision of a formal school library (Radebe 1998, 45-52).

Box-book libraries were distributed in Mozambique to reinforce understanding of Portuguese, the country's official language but for most people their second one. The project was funded by the Canadian Organization for Development through Education (CODE), but reported to the Cabo Delgado provincial director of the Ministry of Education and was managed by a local NGO, Progresso. Books bought in Maputo were sent to the provincial directorate, where they were packed in boxes produced locally and sent to district education officers, who forwarded them to the schools. The initial plan was to create libraries of 150 books (15 titles, each in 10 copies), but by the end 46 titles had been bought, a total of 166,663 volumes. In the first year, all books were imported from Brazil, but locally published books were bought starting in Year 2 and by Year 5 all new purchases were locally produced. Training was incorporated after a mid-term evaluation found that many teachers were poorly prepared to teach the early stages of reading and writing. Over five years, about 110,000 students were served in 550 schools at a total cost of US$303,117 - an average of US$527 per box and US$2.75 per pupil (Amaral 1998, 82-93).

In Thailand and Laos, UNICEF provided more than 400 book-box libraries each year to new graduates of teacher-training colleges and had them carry them to their schools throughout the countries (ADEA/UNESCO survey 1999).

In addition, community resource centres - designed to serve an entire community, although often located in schools and used mostly by students and teachers - were appearing in most African countries. The Ghana Book Trust, supported by CODE, bought about US$45,000 worth of books in the mid-1990s, all of them written and published by Ghanaians, and distributed them to community libraries throughout the country (Crabbe 1996, 41). Teachers' resource centres were also established throughout Africa, each of them intended to serve a number of schools with a collection of textbooks, reference materials, and classroom sets.

Jamaica was the only country in the UNESCO case-study survey to provide a substantial annual allocation for school libraries, and its Library Service is a model for the Caribbean region in its philosophy, structure, and systems. The Library Service's mandate includes ensuring that every school has a library. It distributes library books to schools and encourages inter-library loans from the main library to the schools. Schools are encouraged to accept gifts of books but the Library Service assesses whether they are appropriate and up to date. In order to improve school libraries, it runs annual training sessions for principals, teacher/librarians, and other staff members. Its effectiveness is constrained, however, by the fact that many schools do not have the room for a library and by the high mobility of professional librarians (UNESCO/Danida case studies: Jamaica).

Commercial publishers, civil society organizations (CSOs), and international agencies took an increasingly active part in the provision of supplementary reading material for children. (CSO is a broad term that includes non-governmental organizations, religious organizations, youth groups, trade unions, and other bodies.) In order to relieve the shortage of children's literature in Kenya, East African Educational Publishers inaugurated a Junior Readers Series in 1988 with 10 books. Ten years later there were 127 titles in the series, some acquired from publishers in other African countries, and with average sales of 3,000 copies the series was the fastest growing section of the publishing house (Chakava 1997,42-3; Chakava 1998, 4-5). CODE sponsored a Children's Book Project in Tanzania that assisted the publication of selected titles and bought 3,000 copies of each for distribution to rural libraries and primary school, with the understanding that the publisher would place at least 2,000 more copies in the open market. The buy-back ensured that publishers would recover their initial investment, with profit to be made from sales through bookshops (Lema 1997, 91-5). CODE ran a similar, but less ambitious, project in Guyana. In 1997 UNESCO and APNET launched a pan-African publishing initiative to produce popular books in science for young adults that would address Africa's energy crisis. Six books are being developed by teams from several countries and will appear in English, French, and Portuguese from local publishers in a series called Science for Africa/KAWI - "kawi" being Kiswahili for "energy". The books will describe alternative energy sources, such as solar energy and biogas, in a context of African experience and folklore and practical exercises (Wade 1998, 13). In Pakistan, the government and donor agencies brought international and local professionals together to produce Urdu fiction and non-fiction books for middle school students. More than 150 titles were produced and put in 5,000 schools of the Punjab (ADEA/UNESCO survey 1999).

These and other projects have had to deal with the scarcity of retail book outlets in many countries, and with the reluctance of most booksellers to stock many supplementary readers, which have a less certain sale than textbooks.

Donated books became an important addition to public, school, and university libraries. Book Aid International, a UK NGO, sent more half a million books each year to libraries and institutions, responding to profiles of interest submitted by the recipients. It also supported the Inter-African Book Support Scheme, in which selected books from African publishers were bought and sent to African libraries. That program, and a similar book distribution program operated by CODE, were found to work best when they were demand-driven - that is, when the books were sent to meet stated needs of the recipients and not as an offloading of unsaleable volumes (Priestley forthcoming).

Production of textbooks

Local production of learning materials probably increased during the 1990s, although it is difficult to determine by how much. National interest encouraged the development of local publishing in order to achieve self-reliance, nurture a cultural identity, create local employment, and use local printing capacity. On the other hand, the national economy could not always support a local industry, nor could the immediate demands for large numbers of textbooks always be met by national publishers or printers who lacked capital, capacity, and experience. At a time of structural adjustment, fiscal stringency, devaluation, and inflation in developing countries, foreign publishers had the advantages of sufficient capital, stable currencies, and low freight rates. Their branches in developing countries could attract better staff, buy paper and print with the leverage of a large customer, draw for adaptation on the textbooks published by the parent company and sister branches, and benefit from the profit on dictionaries, atlases, and other reference works published by the parent organization. For French publishers in particular, the African textbook market continued to be attractive. Its potential for growth was greater than that of industrialized countries because of demographic curves and efforts to increase net enrolments. Furthermore, the distribution of the world's French-speaking people was shifting away from Europe, towards 50% living in Africa. French publishers invested in African companies, as in Côte d'Ivoire, and the large French publisher Hachette had a subsidiary, EDICEF, specifically to publish textbooks for markets outside France (Moingeon 1999a, 103-4).

British transnationals continued to be active throughout the English-speaking world, although in some places, as in Jamaica, their attempts to establish branches failed (Randle 1999, 169) and in others they had sold their interests to local publishers, many of whom they had trained. Some British publishers adopted a policy of decentralization, in which autonomous branches operated as regional publishers. Macmillan Kenya, for example, had a local managing director and a fully local staff, published what was appropriate for Kenya without interference from a head office, ran its own in-house training programs, traded with other African countries regardless of whether they had a Macmillan office of their own, and retained all profits for new projects in Kenya (Johnstone 1999, 102; Muita 1999, 152-3). Macmillan had a more traditional operation in Swaziland, where it assisted in the writing of the textbooks used in the schools and enjoyed a monopoly over their publication (ADEA/UNESCO survey 1999). Between 1989 and 1994, British book exports to sub-Saharan Africa increased from just under £39 million to almost £57.5 million (Bgoya 1997, 27). At the end of the decade, however, a British consultant reported that publishers were growing less interested in new initiatives in developing countries because the return on investment was so uncertain (ADEA/UNESCO survey 1999).

Book exports were not all from North to South. India, a publishing superstate in its region, was exporting textbooks and other books for children to Southeast Asia and Africa (Singh 1995). Malawi met an urgent need for textbooks by inviting publishers from other members of the SADC region to tender titles for it to buy with funding from Danida (ADEA/UNESCO survey 1999). The LIMUSA Publishing Company of Mexico published secondary school books for use in Chile, textbooks in primary mathematics for El Salvador, in a variety of subjects for bilingual education in Guatemala, in social science for Nicaragua, and in mathematics and Spanish for Venezuela. Its international success was all the more remarkable in that it was achieved despite state domination of its own domestic textbook market (Noriega 1999, 130).

Of special importance was a growing tendency towards publishing partnerships between local and non-local firms. Such agreements assumed that textbooks would be sold commercially and envisaged a long-term relationship involving sharing of costs and revenue as well as the transfer of knowledge to the local partner. For example, a Canadian publisher, Irwin, cooperated with Ian Randle Publishers of Jamaica in the development of textbooks for the Caribbean. Another Canadian firm, Fraternité Matin, joined with Jamana in Mali to produce two grammar books. Both partnerships were assisted by a Canadian government initiative, International Publishing Partnership, which provided administrative, informational, and financial support. In Cameroon, the French publisher EDICEF worked with a local partner, Editions CLE, to develop textbooks and work books in mathematics and French after the country's externally supported monopoly publisher collapsed. The new books were cheaper than the old ones and were bought by almost 50% of the students in the first year they were available; subsequently the European Union subsidized copies for students in disadvantaged areas (Newton 1999a, 166; Randle 1999, 171; Konate 1999, 178; Davies 1997, 89; Loric 1999a, 182-4).

In what is still a relatively rare example of cooperation between publishers in neighbouring countries, Fountain Publishers of Uganda was able, with local authors, to adapt textbooks in science that had originated with East African Educational Publishers in Kenya. In general, however, regional cross-border trade in educational publishing was impeded by differences in language and syllabi, trade barriers, and difficulties in communication (Tumusiime 1999a, 123-4). International agencies played an active part in some joint ventures. UNESCO was involved in a UNDP-funded project to produce a new English course in China. Foreign publishers were invited to tender for the project in cooperation with the People's Education Press, the largest educational publisher in China. The project involved considerable counterpart training. The books continue to be sold, at very low prices, at a rate of more than 40 million copies per year, but the foreign publisher, who was paid a flat fee, barely broke even, making it doubtful whether the venture could be repeated (ADEA/UNESCO survey 1999).

Constraints in book provision

The constraints that public- and private-sector publishers faced at the end of the decade were little changed from those they faced in 1990, although the increasing role of the private sector and concern about cost recovery resulted in some reconfigurations. Inconsistencies or weaknesses in government were more significant, for example, to private publishers than to a public institution that might be part of the problem. The route towards liberalization had many roadblocks.

Government Government policy was often inconsistent - not only because of changes in leadership but because of conflicts between government departments. Publishers had to deal with several departments, few of them directly concerned with textbooks. The Ministry of Education normally had the leading role, but the Ministry of Industry might be concerned with protecting domestic printing and manufacture; the Ministry of Trade might be regulating the import and export of paper, machinery, and books; the Ministry of Finance possibly regulated the availability of foreign currency and imposed taxes and tariffs; the Ministry of Agriculture might have some say in the importation of raw materials; the Ministry of Culture might set language policies and regulate publishing and printing. Some countries, among ...

Table 10. Requirements vs. production and distribution in Tanzania, 1993-1995 (not available)

...them Ghana and Zambia, waived taxes on imported printing goods meant for book publishing, but in other countries local printers and publishers had to pay tax and duties on imported raw materials such as paper at the same time as competitive foreign-produced books were imported duty-free.

More fundamentally, few countries in the developing world had coherent policies for publishing development or the provision of learning materials. Askerud (1997, 19) found that lack of funding is often exaggerated as a cause for failure in providing basic learning materials, and that other problems are more fundamental. Textbooks did not get the attention they required. Few decision-makers or managers understood publishing or even the difference between publishing and printing, and few gave proper recognition to those in the civil service, or outside it, who did have publishing expertise. As a result, the design of programs was left to educators who had little awareness of the professional skills required to produce and distribute quality learning materials successfully, of possible institutional needs for reorganization, staffing, and budgeting, of the economics of educational publishing, or of the difficulties of marrying educational concerns with the concerns of commercial publishing. From ignorance of the totality and complexity of the book chain, textbooks were treated as simple commodities and planners concentrated only on manuscript development and production. Furthermore, some consultants who had been retained to advise on textbook projects had no interest in seeing local publishing industries develop, and ministries were often reluctant to surrender control of textbook provision even when the need was pointed out. Others (for example, Brickhill 1996, Brunswic and Hajjar 1991, Montagnes 1998) have also pointed to such problems.

In those countries that were in transition from public to private provision of learning materials, inconsistency and unprofessionalism were apt to be compounded by confusion. Few countries made five- or ten-year plans for the change. Some failed to spell out clearly the roles of the players, the timing of change, or the operational details of each phase. The capacity of the private sector was not always taken into account or the need to rebuild publishing skills that had atrophied under state monopoly. If local publishers could not respond to an abrupt transition, the way was opened to foreign ones. In Zambia, for example, the failure of the state monopoly and rapid liberalization led to a renewal of British transnational domination of the market (Brickhill 1996, 20;, Munamwimbu 1996, 67).

Educational statistics remained unreliable and usually outdated. Under such circumstances it was difficult for governments to project requirements or for publishers to determine print runs. Data available from Tanzania (Table 10) indicate the discrepancies in supply and demand that can occur as a result of inadequate information, in this case in a period of increasing private-sector publication and continuing centralized public procurement. Variations of this magnitude can be financially disastrous to commercial publishers. So can unexpected cuts to educational budgets by national legislatures. In the Philippines, for example, publishers who had lobbied for privatization were dismayed when they discovered how small the annual funding would be in comparison with the anticipated market.

Failures in communication within government further hindered efforts to improve the availability of learning materials. Policies might be communicated so imprecisely that recipients were uncertain what they were expected to do, and as a result did not implement new policies or did so incorrectly. Even clearly-worded policies did not reach the schools, or were not believed when they did arrive. In Mozambique, another kind of failure in communication appeared. That country had three separate book-box library projects between 1982 and 1996, but there was little exchange of information among them and lessons learned from one project were not transmitted to the next (Amaral 1998, 84).

Few government officers appreciated the time it takes to write, edit, test and print textbooks. As a result many contracts allowed unrealistically short times from signing to delivery. The resulting textbooks were apt to suffer in quality or would have to be brought in from outside the country.

Political instability, authoritarian governments, and war weakened or destroyed both educational and publishing facilities in the most unfortunate countries.

Finance As devaluation and inflation drove up the cost of imported raw materials, the commercial school book publishers who were being encouraged by the new policies of liberalization and decentralization continued to have great difficulty in finding capital and credit. Investors found publishing unattractive because of low returns and high risk. Banks were reluctant to extend loans because sales revenue in publishing was notoriously unpredictable and few publishers had sufficient collateral or assets. Concessional credit from the World Bank and other international and regional finance institutions, available for agriculture and some other industries, was not extended to book publishing. If loans could be negotiated, interest rates throughout most of the developing world were high - commonly, from 23% to 40% per year. One bank in Ghana did offer loans to publishers but none took up the opportunity because the annual interest rate was 45%. If publishers were able to borrow money at lower rates abroad, they faced the further possible hazard of devaluation. Some in Tanzania went bankrupt after devaluation of the shilling, when the loans they had negotiated in foreign currency had to be repaid. On a positive note, Nigeria established an Education Trust Fund, analogous to its long-established agricultural development banks, to provide soft loans not only to institutions but to publishers, printers, artists, and others involved in producing learning materials (Crabbe 1996, 32; Bgoya 1997, 21; ADEA/UNESCO survey 1999).

Payment for textbooks was nearly always slow. Publishers might wait six months or a year for their money, making it even more difficult for them to produce books for the new school year. In Nigeria, the state and federal governments owed publishers considerable amounts of money for books delivered, in some cases, in the previous decade (O'Connor 1999, 120).

Governments occasionally discriminated against local publishers, in effect depreciating what capital the publishers had. In Romania, international publishers were invited to bid on textbook provision and guaranteed payment in hard currency. Local publishers, in contrast, were paid in local currency in the amount established at the time of contract, regardless of any decline in value of the currency by the time books were delivered and payment was finally made. One publisher calculated that his loss from currency instability amounted to 60% in the case of a civics textbook. Joint ventures between local and foreign publishers were discouraged because the foreign partner would receive its share only in local currency. The Ministry of Education further hampered liberalization by establishing unrealistic times for printing and delivery and for teachers to assess competing books (Papadima 1999, 135; Moingeon 1999b, 148; ADEA/UNESCO survey 1999).

The capital required to bid on major projects restricted the ability of local companies to participate in major projects funded by the international banks. In Indonesia, private companies were allowed to publish books in a number of fields formerly reserved for the government-owned Balai Pustaka. In practice, only large firms were able to compete. The government issued an invitation in 1999 for bids for textbooks and teachers' guides at the junior secondary level in biology, physics, and English. Publishers had to pay 1.5 million rupiahs (about $US220) per subject to obtain the bidding documents. They had further to demonstrate their financial soundness and resources and prove that they had been operating for at least four years with an important part of their business being the publishing and/or distribution of educational books. This was a year after the collapse of the Indonesian economy, and only the biggest of publishers were able to qualify (ADEA/UNESCO survey 1999).

In a similar instance two or three years earlier, a West African distributor found it impossible to participate in a World Bank project that involved four textbooks and teachers' guides for the following reasons. It had to find CFA 50,000 (about US$100) to acquire the tender documents, a bank guarantee of 2.5% of the value of the tender, and a staff housing tax clearance equal to 1.5% of the salary of each member of the staff. If successful, it would have had to provide CFA 9.8 million (about US$20,000), about 5% of the value of the contract, in guarantees. The company would also have had to demonstrate that it had been financially healthy for some years and had assets worth at least four times the value of the contract sum. The government would advance 30% of the contract price, but only subject to a bank's guarantee. The requirements in this case, coming from both the government and the Bank, effectively doubled the guarantees normally required by the World Bank (Bgoya et al. 1997, 86-7).

Market size Not every country was large enough to support a competitive publishing industry. This was evidently true of small island nations, such as those of the Pacific and Caribbean. It was equally true of many other countries. Of the 37 in sub-Saharan Africa, only two or three were thought to have a domestic market large enough to support a book industry, even in primary-level textbooks (British Council 1992, 21). In Central and Eastern Europe, the Soros Foundation considered a country in which 7 million people spoke the same language to be a large publishing market, but only 10 of the 22 former communist countries in which the foundation was active met that criterion. In the other countries print runs and talent pools were smaller and the cost per copy higher (Pinter 1999, 52). Even the relatively successful book industries of Kenya and Zimbabwe together could draw on a market that was only about one-twentieth of a small European country such as Norway, Denmark, or the Netherlands - countries that had a single language and a much better infrastructure (Brickhill 1996, 24).

Local markets were further reduced by photocopying and pirated editions. In most countries, families had limited disposable income for books, including textbooks.

Markets could be expanded by export where there was a common language, as in Latin America. Elsewhere, the barriers mentioned previously worked to restrict publishers to their own countries.

Distribution It was no easier to reach isolated rural schools at the end of the decade than at its beginning. In Tanzania, for example, official figures in 1994 indicated that more than 16 million textbooks had been delivered to district storerooms over the previous five years. Had all reached the schools, that should have resulted in a textbook:pupil ratio of 1:3 or 1:4. A trial survey that year found that the true ratio was closer to 1:9, with extreme variations between schools and districts. Many books didn't reach the schools, and some that did were kept in school storerooms "due to uncertainty over supply". Some books produced by the state were being bought by parents in the market (Brickhill 1996, 17-18).

Liberalization simply increased the problem if the state was not undertaking responsibility for distribution. Bookstores were scarce outside the main centres, and in many countries even well-established religious or family retail bookstore chains were diversifying, with books as only a minor item. In Latin America, it was said, no publishing house had a distribution network capable of delivering books to every school in every corner of the country, and in many instances books bought by the government simply remained in government warehouses (de Bedout 1999. 43). Kenya sought to resolve the problem by consolidating school orders for textbooks at the district level and then ordering the books from local booksellers, thereby strengthening the retail sector of the book trade and reducing the discrepancies that occurred under central procurement (ADEA/UNESCO survey 1999).

Production Paper remained the principal expense. Some countries, such as Mozambique, had to depend on foreign printing for lack of domestic capacity. In general, the constraints previously noted prevailed.

Human resources The lack of professionally trained staff continued to hamper state and private publishers in most countries. Large publishing companies did a certain amount of training in-house, however, as did transnationals. In general, the greatest need was in publishing management and marketing, although more training was also needed in textbook development, editing, design, and evaluation. A study undertaken by Sida towards the end of the decade (Christensen et al. 1999) drew particular attention to the shortage of well-trained staff at the middle level.

Training received new attention in the 1990s and a number of initiatives were taken. The most ambitious was the African Publishing Institute, a training wing of the African Publishers' Network (APNET), which was established in 1992 with the help of several funding agencies, both governmental and NGO. The Institute held occasional workshops in its early years. In 1999 it was scheduled to run 22 national and regional training workshops throughout the continent, on a variety of topics, and in English, French, or Portuguese depending on the country. Workshops are held in cooperation with national publishing associations, using local trainers in accord with a comprehensive syllabus in book publishing and management (Fasemore 1998).

The Asian Cultural Center for UNESCO, supported by the Japanese Ministry of Education and the Japan Book Publishers Association, organized annual courses in various aspects of publishing and book promotion for publishing staff in Asia and Oceania, as it had since 1967. El Centro Regional para el Fomento del Libro en Latina y el Caribe (CERLALC), an international organization created under an agreement between the government of Colombia and UNESCO, offered training courses in book-related topics and published training handbooks for book professionals. UNESCO sponsored training workshops in other countries and, through the International Institute for Educational Planning, supported the development and publication of a series of training modules in textbook planning and publishing in Spanish, French, English, and Arabic, available in print and for a time as an interactive course on the Internet.

A sampling of other initiatives - and it is only a sampling - illustrates the interest of governments and CSOs in publishing training (Priestley forthcoming, Montagnes 1997).

The UK Department for International Development (formerly the Overseas Development Administration) sponsored a number of programs in the Commonwealth and the Arab States. The British Council offered training courses in China, India, Zimbabwe, Tanzania, Indonesia, and Ukraine. The French Ministry of Cooperation organized training in Africa and France for librarians, publishers, booksellers, authors, and illustrators. The Deutsche Stiftung für Internationale Entwicklung (DSE) organized short and medium-term training courses and workshops in the development of materials, including several concerned with the writing, editing, and publishing of textbooks in national languages in francophone West Africa. Sida and the Finnish Ministry of Foreign Affairs supported training in Mozambique, Tanzania, and Zambia.

Among NGOs, the Afro-Asian Book Council organized workshops on author development in Nigeria, Pakistan, Kenya, Ethiopia, and Malaysia. Two German NGOs offered training internationally, the Deutsche Gesellschaft für Technische Zusammenarbeit (GTZ) and the Ausstellungs und Messe GmbH. The latter turned its training attention from Asia and Africa in the 1990s and ran workshops, with support from the German government, in Central and Eastern Europe. Le Centre africain de formation à l'édition et à la diffusion (CAFED), with headquarters in Tunis, offered initial, advanced, and refresher training courses and internships in France for managerial and technical staff. Coopération par l'éducation et la culture (CEC), a Belgian CSO, ran workshops for authors and illustrators of textbooks and children's books in Niger, Zaire, and Tunisia. CODE ran training programs in children's literature, illustration, and design in East Africa, and in editing and book design in Guyana. A Norwegian agency, International Movement towards Educational Change (IMTEC), organized a series of workshops on evaluation, editing, illustration, and financial controls for textbook publishers of Pakistan. In the 1990s, it organized exchange visits of educational publishers of the Baltic countries and Western Europe. The International Network for the Availability of Scientific Publications (INASP) organized workshops on journal production and management for editors and publishers in Southeast Asia and Africa, and commissioned a reference handbook on book marketing and promotion. PUBWATCH, an NGO based in New York, conducted professional workshops, seminars, and training schools throughout Central and Eastern Europe and the former Soviet Union, published a quarterly newsletter describing assistance programs, and sponsored the translation and co-publication of Western guidebooks on book publishing. The Open Society Institute's Centre for Publishing Development in Budapest created a Web site with material that can be used for training in electronic publishing. The Publishing Training Centre (formerly Book House Training Centre) in London ran courses in Latin America, Central and Southeast Asia, and Central and East Europe and developed distance learning packages in proofreading, editing, and copyright.

Formal education for publishers was also available at universities in a number of countries, although many of the courses were more academic than practical.

Donors and funding agencies

At least 150 agencies - international and regional banks, governments, and CSOs - had an interest in book development during the 1990s. The breadth of their interest in the development and provision of learning materials is evident in the foregoing pages. Although sustainability was a long-term goal, it was clear, as costs rose and national economies languished, that dependence on external aid would continue for many years. Five trends were significant.

Funding Six years after Jomtiem, the International Forum for Education for All reported that since 1990 there had been an overall increase in governments' financial commitments and disbursements to basic education (EFA Forum Secretariat 1996a). But the decade in general was one of financial belt-tightening in most countries, and as Northern governments reduced their budgets for overseas development, the textbook/learning materials components of educational programs inevitably were affected. Even so well-established an agency as the Asia/Pacific Cultural Centre for UNESCO announced in 1997 that it would have reduce its activities as a result of the downturn in the Japanese economy. Some bilateral development agencies announced that they would concentrate their activities on a few countries and/or core curriculum subjects (Priestley forthcoming).

Non-governmental organizations - local and non-local - had been well represented at Jomtien and were increasingly active, often in innovative ways, in supporting book provision. Many of the non-local ones operated with financial assistance from their governments. CODE, for example, received substantial funding from the Canadian International Development Agency, although the agency itself played little direct part in book provision. CSOs too were affected by global recession, suffering either in government funding or in their fundraising efforts with individuals and corporations.

Of particular importance in this context were local CSOs associated with the book trade. Among the most important were associations of writers, printers, publishers, booksellers, and librarians, who could respond to local conditions and seek common solutions to problems.

Geographic concentration The decade opened with funder attention still focused on Africa, the continent most in need of educational assistance. As the educational legacy of communist governments collapsed, attention - from large-scale projects financed by the World and Asian Development Banks to complementary initiatives by some NGOs, including the Soros Foundation and PUBWATCH - turned increasingly to countries of East and Central Europe and Central and West Asia. However, the development and provision of textbooks and learning materials, particularly in indigenous languages, continued to be supported by loans and grants in other regions of the world.

Donor coordination From the 1960s to the 1980s, there was little coordination among donors. Some supported curriculum design, others production, others distribution. During the 1990s there was increasing attention to inter-agency discussion leading to a more coherent approach and an efficient, integrated system for book provision. Towards the end of the decade there was increasing interest in sector-wide support. In this approach, led by Sida, funding agencies meet together with government agencies to work out a plan under which as many parts of the book sector as possible receive the help they need (Priestley forthcoming).

The Working Group on Books and Learning Materials of the Association for the Development of Education in Africa (ADEA) provides a forum on book-sector issues for African ministries of education, development agencies, and CSOs. It encourages the sharing of ideas and experiences among funding agencies; finances research into key areas of book policy; publishes the results of that research and books of practical publishing advice in its series, Perspectives on African Book Development; works with the UNESCO/Danida Initiative on Basic Learning Materials by encouraging collaboration among developing countries in specialized areas of national book development; helps national governments bring together the public and private sectors in book provision; and works with other ADEA working groups to promote education (Treffgarne 1999, 22).

The Bellagio Publishing Network is an informal association of donors, African publishers, NGOs, and individuals committed to strengthening indigenous publishing and book development in Africa. The participants meet from time to time to discuss common issues. The Network also publishes a newsletter and a book series that have become an important source of information about publishing development. The group evolved from an international conference on publishing and development organized by the Obor Foundation and held at the Rockefeller Center in Bellagio, Italy, in 1991.

One of the first indirect results of that conference was the formation in 1992 of the African Publishing Network (APNET), which brings together national publishing associations from all sub-Saharan Africa, anglophone, francophone, and lusophone. It has been assisted by several government agencies and NGOs in its work in networking, training, and trade promotion.

The Asia-Pacific Cooperative Programme for Reading Promotion and Book Development (APPREB) is a UNESCO-sponsored and supported network of regional institutions and organizations which promote books and reading in Asia and the Pacific region. It was launched in December 1991 in Kuala Lumpur, at UNESCO's initiative, in an attempt to redress the publishing imbalance in the region - one in which many countries have made considerable advances in book development while others remained at an early stage.

The need for further coordination remains, however, and conflicts have occasionally arisen between funding agencies. At mid-decade in the Northwest Frontier Province of Pakistan, for example, half a dozen agencies were funding separate projects in curriculum reform and textbook provision. In Cambodia in 1993, an Educational Donors Group had 35 members, but about 15 more operated independently. In Guyana, two funding agencies supported textbook production with DTP technology, with the result that in one room the small nation was learning how to operate separate technologies for primary and secondary level books, one IBM-based, the other Macintosh-based. In Mozambique, the Dutch and Swedish governments were supporting Caixa Escolar (the School Fund), under which consumer subsidies were channeled from the Ministry of Education to district offices for the purchase of textbooks by the schools. The World Bank indicated an interest in the Caixa Escolar scheme in 1993, but eventually funded a parallel system under which books were printed outside the country and distributed free of charge. The Bank's project counteracted the Caixa Escolar system and the newly liberalized Mozambican school policy (Read 1995, 208; Sida 1996, 38).

A more fundamental division between funders was in their approach to liberalization. One camp wanted to free production as quickly as possible. The other sought a phased development of capacity (Brickhill 1996, 19). The first was intended to meet immediate needs and often led to involvement, if not dominance, of the market by transnational companies with deep pockets and long established lists that could be adapted quickly to national needs. The other was intended to provide for an orderly transition and nurture indigenous publishing.

Demand-side financing In the last half of the decade, agencies increasingly rejected the concept of providing printing paper and books as commodities to meet short-term needs for textbooks and learning materials. More attention was paid to developing a viable system of provision, with concern for all links in the book chain. Sida, for example, decided to change its funding for textbooks, in countries which had begun decentralizing procurement, from the direct support of production processes to support for the development of competence and the local (district or school level) purchase of materials (Sida 1996, 23). The World Bank funded a number of educational projects involving demand-side mechanisms such as stipends, vouchers, and targeted and subsidies to disadvantaged and minority groups (Patrinos and Ariasingam 1997, 18-20).

World Bank policies The World Bank was by far the single largest external source of finance for education. Between 1963 and 1997, it was involved in 560 projects in 112 countries, and at the end of that period was still involved in 202 education projects in 88 countries. Its projects ranged from very small countries such as São Tomé, with 100,000 people, to China. Between 1980 and 1995, its total volume of lending for education tripled, and education's share in overall Bank lending doubled. Between 1994 and 1997 it lent US$550 million for textbooks alone, the greater part for books at the primary level. Bank funding in 1995 accounted for about one quarter of all aid to education. Even though that massive assistance represented only about one-half of one per cent of the total spending in developing countries on education, the Bank's policies were of extraordinary significance. The fact that regional banks marched more or less in step with the World Bank added to the latter's importance (Nwankwo 1999, 139; McGregor 1994, 63; Sosale 1999, 1; World Bank 1995, 14-15).

The World Bank's bank policies were not without critics, in particular its insistence in opening textbooks to international competitive bidding in any contracts above a set threshold. Although the policy was questioned most often in discussions about Africa, in part because of the vigorous advocacy of APNET, that is not the only continent on which the issue has arisen. In the Philippines, international competitive bidding came into open conflict with national law, which required textbooks used in the public system to be published by companies that were totally locally owned. The Bank has a general policy of encouraging awards through bidding to the "lowest evaluated bid", a phrase that implies concern for quality, and in specific projects has shown some flexibility in implementing competitive bidding - for example, by agreeing to weight bids with price as only a minor factor - but its officers have not been consistent. Regional banks have been somewhat more flexible in requiring international competitive bidding. Despite, or because of, the criticisms, local publishers in both Nigeria and Mali have won competitions for textbooks under World Bank projects against foreign-owned firms.

In 1997, the Bank organized a seminar at which stakeholders from different part of the book chain and from different parts of the world gathered to share their experiences. The proceedings of that seminar (Sosale 1999), frequently quoted in this survey, comprise the most significant published examination of external funding to publishing development since the proceedings of the Bellagio conference at the beginning of the decade (Altbach 1992).

The seminar itself was part of a policy re-evaluation by the Bank. On the final day, one senior staff member wondered aloud whether the Bank should reconsider its insistence on the "extraordinarily complex" rules of international competitive bidding for textbooks and look at appropriate national or local competitive bidding or other sorts of supply-side or demand-side financing.

A new Bank policy on textbook provision, released for discussion in draft in 1999, indicates a greater interest in articulating educational publishing with general publishing on the one hand and with educational outcomes on the other. It recognizes the importance of textbooks to a viable publishing industry and the importance of a viable publishing industry to the sustainable provision of learning materials - relationships that will be discussed further in Part V. The Bank's priorities, as set out in the document, are: (a) support to good classroom teaching and learning practice, including a coherent program for the provision of teachers' guides and teacher training; (b) adherence to legal and other measures for the protection of copyright and other intellectual property rights; (c) articulation of agreed roles of the public and private sectors in the development, production, and equitable distribution of textbooks and reading materials; (d) maintenance of transparent and competitive processes in the selection or purchase of books for educational use or for contracting publishing or printing services; and (e) commitment to longer-term financing of book development and provision.

Concerns remain among publishers, nevertheless. The new policy does not address procedural details that were raised at a meeting of West African publishers in 1998. They include high deposits for bidding, inadequate information, complex bidding documents, cumbersome procedures, low advance payments to the winning publisher, and delivery periods so short that indigenous publishers cannot hope to meet them (Anonymous 1999).

Return to contents
[ Discussion Forum | Contact | Site map | Search this site | top ]