Report No. 12366-BR Brazil Higher Education Reform October 6, 1993 Population and Human Resources Operations Division Country Department I Latin America and the Caribbean Regional Office FOR OFFICIAL USE ONLY MICROGRAPHICS Report No: 12366 BR Type: SEC C,~ ~ ;. Document of the World Bank This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed withot World Baik authoriz-ation .; : FI-SCAL YEA January I - December 31 ABBREVIATIONS AND ACRONYMS CAPES MEC's Agency for Training of High Level Personnel CEF Federal Savings Bank CFE Federal Council of Education CFES Federal Council of Higher Education CNPq National Council for Scientific and Technological Development FINEP Agency for Financing of Studies and Projects FNDCT National Scientific and Technological Development Fund FTE Full-time Equivalent IPEA Institute of Applied Economic Research MEC Ministry of Education NUPES Higher Education Studies Center of the University of Sao Paulo PADCT Support Program for Scientific and Technological Development PCE Education Credit Program PUC-RJ Catholic University of Rio de Janeiro PUC-SP Catholic University of Sao Paulo SENESU National Secretary for Higher Education UCLA University of California at Los Angeles UNICAMP State University of Campinas UNIFOR University of Fortaleza UJSP State University of Sao Paulo UFMG Federal University of Minas Gerais UFSC Federal University of Santa Catarina FOR OMCAL USE ONLY TABLE OF CONTENTS SUMUMARY ANDRECOMMENDATIONS ................... i I. 1NTRODUCION ....................I................... I The Need for Higher Education Reform in BrzU .................1 The Objective of This Report ............................. 2 I. DESCRIPION OF THE HIGHER EDUCATION SYSTEM OF BRAZIL. 4 Structure and History. ....................... .4 Equity and Extenml Efficiency .8 Graduate Education and Research . Private Higher Education .12 m. ISSUES IN PIGHER EDUCATION .. 18 Costs of Federally Funded Higher Education ............... ... 18 Governance and Management of Federal Institutions ........... ... 23 FimncingofHigher Education .......................... . 25 Relationships between the Federal Government and Private HigherEducation .............................. 30 IV. PRINCIPLES FOR IGHER EDUCATIONREFORM .33 The Economics of Higher Education and Implications for Refonn .33 Other Elements in the Reform of Higher Education .36 V. PROPOSED REFORMS IN HIGHER EDUCATION .... 38 IHigher Education Financing ............................. 38 Reforming the Student Loan Program ....................... 40 Governance, Management, and Oversight of Higher Education ... .... 43 Other Reform Measures ............................... 48 VI. FINANCIAL AND OTHER IMPLICATIONS OF REFORM .51 LIST OF TEXT TABLES AND GRAPHS Chart 1 Brazil - Undergraduate Enrollment Growth, 1960-1989 Chart 2 Brazil - Total Higher Education Enrollment by Government Authority and Institutional Type, 1989 Chart 3 Education and Income in Brazil Chart 4 Projected Expenditures and Revenues for Proposed Brazilia!L Loan Program Table 1 Higher Education Enrollments in Private Sector in Select Countries This report was writen by a team led by Laurence Wolff and including Douglas Albrecht aud Alcyone Saliba. Tm addition, the following consltants wrote background papers to this report He Guaai Calban; Madu Go.nes; Pedro Lincoln Mates; Edson do Oliveimr Nunes; Lura Dantas; Violets Maia Monteio; Je4acques Pal; Jacques Schwaman; Simon Schwartzman; and Teodoro Rogerio Vabl. This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. Table 2 Enrollment and Estimated Cost by Higher Education Sector, 1989 Table 3 Enrollment by Type of Institution, 1989 Table 4 Background of Entering Students for Several Brazilian Institutions Table 5 Private Sector, Enrollment Breakdown by Type of Institution, 1989 Table 6 Higher Education: Student Teacher Ratios and Unit Costs in Selected Countries Table 7 Costs and Quality in Two Federal and Two Private Universities Table 8 Student Loans Per Year Under PCE Phase 1 (1987-1989) Table 9 Interest Subsidies on the PCE Iaa Programs by Phase Table 10 Cost Recovery in Public Higher Education as a P, -rcentage of Unit Operating Expenditure, Selected Countries Table 11 Scenarios for the Future Table 12 Single Year Financial Impact of Higher Education Reform Measures (esOt.- ed for 1995) BMLIOGRAPHY .................................. 58 ANNEX 1. CRErr A FOR ALLOCATING FUNDS TO PUBLIC INSTITU ONS.. 62 ANNEX2. LISTOFCONSULTANTREPORTS ........................ 71 ANNEX3. STATISTICALTABLES ................................ 2 1I-1 Higher Education Enrollment by Level, Institutional Type and Sector, 1960, 1975, 1985, 1989 11-2 Enrollment by Governing Authority and Institutional Type, 1960, 1975, 1985, 1989 II-3 Student Enrollment and Completers by Field of Study and Level, 1989 114 Undergraduate Enrollment by Governing Authority, Historical Series, 1960-1989 11-5 Undergraduate Enrollment by Governing Authority, Selected Years II-6 Higher Education Evolution Between 1983 and 1988 II-7 Undergraduate Enrollment Percentages by Region, Governing Authority, and Institutional Type, 1988 II-8 Public Education Expenditures by Government and School Levels, 1985 II-9 Ministry of Education Budget, 1985-89 in Constant 1989 Prices II-10 Parental Income of Federal University Students by Region, 1989 Sample Data 11-11 Summary of Case Studies of Two Confessional and Two Community- Based Private Universities 11-12 Private Non-University Tigher Education in the State of Rio de Janeiro, Case Studies Summary E-1 Unit Expenditures in Public Higher Education, Selected Countries m-2A Student-Teacher Ratios in Federal Universities, 1988 1-2B Student-Teacher Ratios in State and Municipal Universities, 1988 m-2C Student-Teacher Ratios in Private Universities, 1988 m-3 Federal Universities, Expenditures per Student in 1988 m14 Summary of Case Studies of Federal Universities m-s Comparison of Federal and Private Universities Quality, Efficiency Measures and Cost, 1989 III-4 Summary of Case Studies of Federal Universities 11-15 Comparison of Federal and Private Universities Quality, Efficiency Measures and Cost, 1989 V-1 Student Loan Schemes around the World V-2 Subsidies and Govemment Losses on 23 Student Loan Schemes V-3 Distribution of Enrollments and Expenditures for Higher Education With and Without Different Reform Scenario: SUMMARY AND RECOMMENDATIONS Bacnk=j 1. World-wide economic success is increasingly based on an ability to Incorporate new technologies into industrial processes rathxr than on the simple exploitation of natural resources. This requires, among other tings, a sophisticated labor force at technical and managerial levels as well as on the factory floor. Brazil is falling behind in providing its population the fundamental education basis upon which to build these skills. Brazil's current investment in primary and secondary education is inadequate and inefficient, as documented by pervious World Bank reports. In higher education, large public expenditures are inefficiendy spent and are not proviiding an adequate return to society; the result is that Brazil is not able to provide adequate higher level manpower to support economic growth. This report seeks to contribute to the current debate in Brazil on how higher education can better serve Brazil's economic and social development. It documents issues and problems in an international context and it proposes a variety of institutional and flnancing reforms. Description of Higher Education in Brazil 2. Higher education in Brazil is a large enterprise, with a enrollment in 1990 of over 1.5 million students, equivalent to 12% of the population aged 20-24, having grown from less than 500,000 twenty years ago. Currently, about 22% of enrollments are in federal institutions, 18% in state and municipal institutions, and 60% in private institutions. Brazil's current model of a small but higher quality public system is similar to that of Japan, Indonesia, Chile and Colombia. About 23% of all public education spending on education, equivalent to 1.3% of GDP, goes to higher education, although only about 2% of enrollments are at this level. The private sector and individuals directly spend an additional .4% of GDP on higher education. 3. Higher education in B&azil is diverse in terms of institutional mission, costs, and quality. Two state universities (in Sao Paulo) have high quality undergraduate programs and enroll one quarter of all Brazil's graduate students. Six federal institutions (92,000 students) and two private Catholic universities (25,000 students) offer undergraduate programs of relatively good quality and a signifcant amount of graduate education. An additional 24 federal institutions, enrolling 268,000, several of which also offer good quality undergraduate instruction, are scattered around the country. Private higher education can roughly be divided into three types: (i) the higher quality Catholic institutions, (ii) other private universities (310,000 students), mostly confessional and community based, and (iii) a large number of 'isolated facilities" enrolling 500,000 students, run essentially as businesses, and offering low cost education, usually through night classes. Graduate education enrolls about 50,000 students and is characterized by a competitive funding system and a number of high quality programs. Principles of Higher Education Reforn 4. According tJ economic theory, there are several broad justifications for a public role in higher education. Private credit markets are not structured to support investments by individuals in human capital; therefore public support is needed to ensure an adequate level of such investment. Since the generation of new scientific knowledge is highly risky and usually not appropriable by an individual or company, private individuals and entrepreneurs by themselves do not have adequate incentives to perform this research or to disseminate it; therefore the state needs to help to support a variety of non-appropriable types of research. The state also has an interest in helping to ensure that, by and large, access to higher education is determined by ability to achieve rather than solely on ability to pay. Higher education also has a number of other, albeit non-quantifiable social benefits, such as developing an educated citizenry, training leaders, and transmitting knowledge fimdamental for the functioning of a modern society. ii 5. There are also importat economic reasons why public intervention in higher education should oe limited. First, totally free higher education is inequitable. Those who can afford higher education isually come from the more effluent sectors of society, and they can forgo the income which could be gained by entering the labor market instead of attending hisher education more easily than the less affluent. Support by a general tax base of free higher education would mean that the poor would be subsidizing the rich who benefit most from higher education. Furthermore excessively high investment in higher education could draw funds away from primary and secondary education which in the developing world generate higher economic rates of reurn to society. Budget constraints and increased social demand for higher education mean that thn. government cannot afford to provide all the funds needed for higher education as well as for primary and secondary education. 6. It is unclear what the proper balance between public and private support of higher education should be. Nonetheless, it is clear that in most developing countries tCre is a need to increase individual and ptivate sources of funding for higher education. This can be achieved by charging tuition for teaching costs of undergraduate education in public institutions, by avoiding constraints on the growth of the private sector, and by channelling public funds to higher education mainly to correct for credit market failures, protect societyrs interests in equity, and assure adequate generation of basic knowledge. In principle many of these interests can be met dtrough subsidizing loan schemes for all students who need them as well as scholarships for the truly needy, while charging the full teaching costs. 7. No country in the world currently has an education system which charges full teaching cost in public institutions accompanied by massive loan schemes and scholarships. There are a number of reasons. Historically, Governments could afford the costs of an elite higher education system, and belief in this model persists even in a time of mass higher education. Cost recovery from student accompanied by loans has also not been fully implemented in many countries because there are a number of significant practical problems in the management of large loan schemes, as well as non- trivial transition costs in requiring institutions with long teaditions to accept radically new financing mechanisms. Nonetheless, many countries in the world are finding themselves forced to move in the direction of increased tuition charges in public institutions, expanded loan schemes, and the growth of private educ3tion. 8. In an age of increased specialization of knowledge, countries are also moving towards policies which encourage increased institutional differentiation and a variety of institutional missions (e.g., research, extension, technology developme, and professional, technical and teacher education). New policies emphasize decentralized decision making in public institutions, increased accountability of institutions receiving public fiuds, more open competition for resources, and increased access of the less privileged to higher education. The aim is to improve the quality and cost-effectiveness of higher education. Quality is here defined as "value added" -taking an entering student, whatever his native ability or prior knowledge, and increasing his abilities to reason, solve problems and communicate. Financing of Hi ihg 5 Edci 9. I. In the current system of financing federal higher education institutions, the Ministry of Education (MEC) has no criteria for allocating resources other than the prior year's budget plus an increment. This means th university rectors have no incentive to increase efficiency or improve quality. Moreover, approximately 94% of these funds go to personnel, who are paid directly by the federal Government, and a scant 6% is allocated to non-personnel items. Federal universities are almost completely dependent on the federal government for fiuds, since they are forbidden by the iii Constitution to charge tition and various regulations discourage them from charging for student services and from seeking funds for research and services from the private sector. 10. Are the unit cost of federal higher education excessive in terms of the level and quality of instruction and research in these institutions? This is currendy the source of impassioned debate in Brazil but little systematic analysis. One way of approaching the issue Is to examine student/teacher ratios. They average 8.8:1 in federal insttutions, a figure which is among the lowest in the world, and far lower than most private universities in Brazil. The ratio of students per non-teaching staff is also extremely low: 3.9 students per non-teaching staff person. Enrollments over the seven years through 1990 have been stagnant while the number of teaching and non-teaching staff have increased. Annuat federal higher education expenditures per students are estimated at US$7,930. Of this amount about $6,000 goes to teaching and research, with the remainder to university hospitals and for pensions. This cost is far higher than the unit cost of $4500 in the Catholic University of Rio (PUC- Rio), the best private university in Brazil, and four to eight times higher than in other private universities. The larger federal institutions have no economies of scale compared to smaller institutions. It is true that many staff in federal institutions engage in research, more students in these institutions are enrolled in high cost science and engineering courses, some of these institutions provide extramural services and training, and some of these institutions offer significant higher cost graduate programs. Nonetheless, even after taking these activities into account, the unit teaching costs of federal institutions could easily be reducea by at. least 33%. Furthermore, physical facilities of federal institutions appear to be grossly under-utilized and at least some of them cot!ld accommodate twice the number of students. 11. Free tuition for public higher education institutions in Brazil has had the perverse result of fully subsidizing children of the upper-middle income classes while lower-middle class students must pay to attend private institutions. In 1989, 44% of public higher education students had parental income of over ten minimum salaries, compared to 10% in the general population. Free tuition also encourages students to extend their period of study and reduces the pressure on public institutions to provide cost effective educational services in competition with private institutions. 12. Aside from fimds for research going mainly to the two best private universities, Government support to private education is provided solely through the student loan scheme which in 1989 amounted to about 10% of the income of private instittions. This program began in 1976 on a highly subsidized basis. In 1983 and again in 1989 high inflation and the inefficiency of the program brought it to the verge of bankruptcy. A new but inadequately funded program is now in place with a lower but nonetheless significant intest subsidy. 13. Ra . The Government has three financing tools with which to increase the equity and efficiency of the educational system-introducing cost recovery in public institutions, fiunding public institutions on the basis of formulas, and refocussing and expanding the student loan scheme. To increase the equity of the educational system and to provide resources for future expansion and quality improvements, the Government should recover costs from students ia public tertiary institutions, through tuition payments and other fees (for laboratories, examinations, etc.). The eventual objective of cost recovery in public higher education should be to recover a large portion of teaching costs from undergraduate students. However, cost recovery is only one of se, er4l means, albeit an important one, at the disposal of Government to increase efficiency and equity of thc system. Furthermore, recovery of costs in federal institutions will depend on a constitutional amendment, and even afterward will need to be gradually phased in. 14. Another very powerful tool to increase efficiency, which could be implemented without constitutional reform, would be for the Government to finance the public system on the basis of formulas derived from estimates of what it iIwu cost, on a unit basis, to undertake activities of teaching and research, rather than what institutions would like to spend; within these broad iv parameters, individual institutions could bave substantial flexibility to make adjustments within the budget. The federal government could differentie target unit costs by level (undergraduate, graduate) and field of study (science, engineering, medicine, social science, law, humanities, and education) and then assign funding to each institution based on its enrollments in each fie.. of study. Given that the hard sciences, medicine, and graduate education cost significantly more than other areas, institutions with more enrollment in these are.ts would receive proportionately more government funding. The overall objective wuld be to reduce average unit costs to about two thirds of present unit costs, by requiring institutions to increase the current student/teacher ratio of 8.8:1 by at least 50% and by similarly increasing the current student/non-teaching staff ratio. 15. The third financing tool #o increase efficiency is the student loan program, which should become one of the cent>' elements of the Government's higher educat'on policy. MEC should control this program. The average size of loans should depend on the level of tuition charges in public and private institutions. The Government should consider policy measures, such as income contingent loan repayment, that prevent repayment burdens from becoming excessive for low wage earning students-not only a matter of equity, but of efficiency, since excessive burdens lead to higher default. To target the program, the Government could provide scholarships to the truly needy and could set an explicit income ceiling on students' eligibility for subsidized loans. The Government could also use the loan scheme to encourage increased quality in private educeZion by setting minimum standards for participating institutions, such as a minimum share of faculty with advanced degrees and adequacy of library resources. The Government should also consider using loans to advance policy objectives in areas it considers of high social priority, such as forgiving portions of student loans to graduates who teach in public primary schools or provide basic health services in low income areas. Governance and Manageent of Federal Higher-Education 16. IsM. The current system of governance and management of federal higher education encourages inefficiency and results in inadequate public accountability. The federal government sets all personnel policy, including recruitment, promotion, and salary levels, much of which is encoded in the "Law of Isonomy' passed in 1987. Legally, once hired, a teacher can only be fired in the extreme case of "just cause," which is hard to document. Indeed, firing of teachers rarely if ever occurs. While the law requires peer evaluation for piomotion to full professorship, an advanced degree is not necessary. Lack of rewards and/or punishment for teachers encourages what is described as the "pacto corrupto* between teachers and students, where the teacher spends little time preparing lectures and correcting exams and the student need not spend much time attending classes and studying. Given this system, it is almost surprising that in many cases teachers are dedicated and quality is reasonable. The Federal Council of Education (CFE) establishes the minimum curriculum for all courses, which covers about 94% of the total of credits needed to obtain a diploma, thus inhibiting diversity and imovation. Legally, a diploma is required to practice a large number of professions in Brazil. Since the end of military rule, most universities have selected rectors on the basis of a popular election in which all students, teachers and administrative staff vote. This has resulted in a highly politicized process. Within universities, there is little institutional planning, almost no cost analysis or budgeting systems, and the rector has little power. 17. Recomm idaions. The federal government should abandon centralized control of federal higher education institutions. Subject to certain rules and guidelines, each institution should be free (i) to hire, pay and utilize staff accorAing to its own criteria, (ii) to seek funds from a wide range of sources, including industry, local govenment, and students, and (iii) to establish new programs of study. Federal institutions should be funded to a great extent through annual block grants and should be permitted to charge tuition. The "Law of Isonomy' should be repealed, since it is incompatible with autonomy, especially in terms of institutional responsibility for budgeting and for hiring and firing of staff. v 18. MEC should develop analytical capacity (a small but high level stat) to determine and revise as necessary the rules and formulas for allocating higher education financing, as well as to examine long term policy issues. Accountability should be increased by linking evaluation of the performance of higher education Institutions with the budgeting process. MEC should sponsor expanded programs of training and professionalization of university management. 19. To complement the changes in incentives brought about by a new system of governance, the current method of electing rectors should be replaced by one which gives appropriate weight to the most producdve members of the academic community, representatives of society as a whole, and government. The rector's office should have budgetary decision-making power; re-appointment of rectors should be permitted if their performance is satisfactory; and the development of a pennanent professional cadre responsible for budgetary and institutional planning should be encouraged. 20. To increase quality and diversity in both public and private institutions, the Government should extend its recently inaugurated program of dereguladon of professions. Ste-wide or national certification examinations (preferably managed by private non-profit agencies and institutions) should be developed for entrance into a restricted number of professions such as medicine and law. These examnations should be of sufficient rigor to encourage higher quality in undergraduate education. The CFE should not prescribe curriculum content, however, and at the most should ask professional societies to provide non-compulsory guidelines. MEC should also encourage improved quality through the establishment of an independent system of peer review and evaluation of higher education institution, both public and private, using a variety of quality and efficiency measures. It should also support research on the (public and private) economic rates of return to higher education by sector and profession, conduct tracer studies of graduates, and explore ways of Introducing pre- and post. achievement testing in higher education. The Government should continue to strengthen its thus far successful system of financing of graduate training and research while at the same time encouraging increased private sector investment in research and development. ncoraoement of Priat Educadon 21. Ise. Government today tre the private sector as marginal to the higher education system, while at the same time regulating it haphazardly. Ihe CFE is charged wih authorizing the existence of private higher education institutions, including specific courses and enrollment. However, from year to year the CFE makes authorizations on the basis of unclear and subjective criteria including 'labor market demand". Ihe Federal Government also continues to regulate tuition on the basis of frequently changing rules. Tuition control adversely affects the aspirations of the better institions to increase quality. 22. R tM . The Government should no longer use labor market or "social" needs as criteria for recognizing courses, since experience has shown that it is impossible to esfimate future manpower needs with any precision. Consistent with normal consumer safeguards, the Government should allow any institution to operate which has consistent interna rules of operation, staff with at least bachelors' degrees, and minimum physical facilities. Once accredited, private institutions should be permitted to increase enrollments aeely. Concurrently, a strong system of independent evaluation and assessment of all higher education institutions should be established, which would review accreditation stats periodically. 23. The Government should end all control of tuition. Concurrent with this policy, it should authorize new courses, permit enrollment to increase freely, and reduce certification requirements, so as to increase competition among schools. Government support of confessional and "community' universities, which are of particular importance to the public because of their relatively high quality and sense of social responsibiity, should be Indirect, through an expanded student loan scheme and vi funding of graduate education and research. Government should permit these institutions to increase tuition freely, and sbould consider revising the tax code so as to provide incentives for private and corporate donations to higher education. Financial and Other- Implications gfQrm 24. In the absence of reform, Brazilian higher education will constitute an increasing and ultimately unsustainable drain of the federal education budget, with litle or no results in terms of improved cost-effectiveness and increased quality of higher level human resources. The teaching and research cost of federal higher education of about US$6000 per student annually in 1989 could be reduced to US$4000 through improvements in internal efflciency which would not mean any loss of quality. A reasonable estimate of the share which undergraduate teaching costs represent would be about 60%, or US$2500. This one third reduction of unit costs, accompanied by a tuition charge equal to the fill costs of undergraduate teaching in public institutions, as well as a vastly expanded loan scheme, could reduce the costs to government for higher education in 1995 by more than 50%. A program of partial cost recovery, with tuition charges of about $700 per year (25% of teaching costs), accompanied by the above reduction in unit costs, and an increase of 50% in enrollment in federal institutions (to utilize excess physical and teaching capacity) would reduce total expected government expenditures in 1995 by 1 1 %. If enrollment were held steady rather than increased in federal institutions, the government's expenditures compared to no reform would be reduced by 35%. While these savings would be significant, a reform program would also bring important benefits to society in the form of increased diversity, cost-effectiveness, equity, and quality of the higher education system. Feasiblity of Reform 25. The time is ripe for action in higher education reform. The overall compression in government spending, declining real wages in universities as a result of fiscal constraints, continuous strikes by faculty and students, and the impossibility of improving quality under the current system have clearly put Brazilian higher education in crisis. The Government and the informed public have been discussing higher education reform. For example in January 1991, the Ministry of Economy announced Its desire to introduce tuition in public higher education. In 1991-92 MEC officials proposed a number of the reforms described above. In mid-1993 a parliamentary commission completed a study on higher education issues. Government reports have focussed on the need for providing more autonomy and accountability to higher education, and newspaper articles have reported a debate over the costs of public higher education. To succeed in reform, a program will need to be established which may, in the short rn, adversely affect the fortunes of pressure groups and lobbies, but will, in the long run, ensure that higher education serves national goals for renewed economic development and social progress. I. IN7RODUCIION T Me Ned for Hfigher Education Refor-m inBazi 1.1 Increasingly, world wide economic success is based on ability to incorporate new technologies into Industrial processes rather than on the simple exploitton of natr resources. The increasing knowledge content Af production means that human resource development at all levels will become the fundamental source of economic development. However, there is much evidence that Brazia is falling further behind the industrialized countries in human reoce development First and foremost, Brazil is far behind in its investment and output in primary and secondary educationL Innovation and discipline on the factory floor will come from workers, not just from managers. Not only is Brazil's current public investment in primary and secondary education inadequate and inefficient, but there Is a lack of public awareness of the importance of a cost-efficient primy and secondary sector. 1.2 Inequities of income between the rich and poor are higher in Brazil than in most other counies at its income level. Brazilian society wIll need to bridge this gap, not only for social and humanitarian reasons, but also because a modern economy requires a broad base of educated manpower. A social development policy should therefore be a fudamentd element of the Governments economic reform program, with its strongest focus on providing a quality primary and secondary education to the majority of Brazil's citizens as well as on expanding basic health services. A broad base of an educated citizenry will be essential for Brazil's internal tranquility as well as to esure intenational competitiveness. 1.3 Higher education also requires reform and improvement if Brazil Is to meet the demands of the twenty-first century. The large public expenditures at this level do not seem to be providing society with an adequate return to its investment in tem of the quantity and quality of human resources produced. The findamental need in higher education, as in all levels of education, Is to improve qualty cost effectively. Quality is here defined as *value added." An educationdl insDtitu tae the raw material it receives-the student, whatever his or her native ability or prior knowledge-and provides a product in the form of a graduate whose abilities to reason, solve problems, and communicate have been increased to the maximum extent possible. A high value added institi in higher educadon could well be one which accepts very poor students, or those with nadequate preparation, and educates and develops their capacity so that they become productive members of society. Many U.S. community colleges should therefore be considered of high quality, since they provide good remedial and/or technical education to young men and women with inadequate preparation. Instiutions such as the University of Ijui in southern Brazil appear to be providing this kind of education. But, in general, the current system in Brazil encourages excessive credentialism as well as what is called the "pact corrupto," where students and staff may have an unwritten agreement to demand little in the way of teatching and learning. 1.4 Brazil also needs to stive to establish higher education insttions which add directy to the stock of human knowledge of the world, similar to the great research universities of the United States and Europe. Developing countries must strive for similar excellece in those areas where their resource endowments are such that they could benefit from increased knowledge. For example, Brazil especially needs middle and high level expertise to enable it to exploit in a sustainable manner its rich endowment of physical resources and its diversified flora and fauna. Brazil also needs to keep pace with world wide developments in economics, management and other social sciences to strengthen management of its own society. Finally Brazil, as well as othet countries, needs to have an educadon and research system which will retain its best scientists and researchers within the country so that they can serve their country's needs. There is a world market in ideas-the fundamental building blocks of technological advances, 2 which developing countries need to be capable of tapping into. While it is difficult to measure direcy Brazil's position vis a vis the world, in l9R4 Brazilian scientists published out 1000 articles in scientific journals of international reputation, a number which, while higher than any other Latin American country, amounts to less than .04% of the 230,000 scientific articles published in the world in 1984. Using a larger data base-the number of books and papers produced by Brazilian authors in the social sciences, arts, humanities, science and technology, Brazil in 1986 produced about 7000 of such documents out of a world total of 1.2 million. Furticrmore, it is striking that Brazil's two world-class universities are the state-run universities of Sao Paulo and Campinas while the federal universities, many of which strive for this excellence, have relatively small numbers of graduate students and only islands of higher quality research in the federal universities of Rio, Sao Carlos, Minas Gerais, Pernambuco, Santa Catarina, Rio Grande do Sul, and a few others. 1.5 There is a growing awareness in Brazil of the need for reform of higher education, especially at the undergraduate level, as the burden on federal government budgets of federally funded higher education continues in a period of financial stringency. Since 1989 the Brazilian Government has focussed on macro-economic reform-restructuring the economy through trade liberalization and deregulation, reducing the fiscal deficit, and holding down inflation. For the Government to follow dtrougb on its objectives, it is essential to begin to treat higher education with the same insistence on efficiency and cost-effectiveness as other sectors of the economy-through reducing control and regulation. increasing the cost-effectiveness of public expenditures, and re-defining the role of Government so as to meet only those social needs which cannot be met by the private sector. Government will need to mobilize its most progressive and farsighted citizens to implement a program of reforms which will in the short run adversely affect the fortmes of pressure groups and lobbies but in the long run will ensure meeting the nation's goals of long term social and economic development. The Obiective of This Rprt 1.6 This report on higher education in Brazil can contribute to the growing debate on higher education reform by documenting the issues and problems of the system and placing these issues in an international context. In fact issues which are highly controversial in one country may be considered normal in a neighboring country, and much can be learned from a comparative perspective. For exaple in Brazil and other Latin American countries the possibility of cost recovery in public higher education is highly controversial; yet countries as Aiverse as the USA, Spain, Korea, Australia, Chile, and Indonesla recover significant percentages of costs from their public education students. Similarly, while Brazil. s weUl as Japan, Colombia, and the USA have strong private higher education sectors, the possible establishment of a private education system in countries such as Morocco, France and England is the subject of heated debate. 1.7 This report is the third in a series of education sector reports produced by the Bank since 1987 .on education in Brazil; the previous two reports covered primary and secondary education. In addition a report on Public Spending in the SocW Sectors completed in May 1988 provided a broad overview of expenditures in education as well as financing and management problems. Each report by level examnu the major issues and identifies policies that coula improve the allocation of public and private resourcc as that level. Two key themes of this work on the social sectors are (a) the importance to Brazil of gening a higher reurn from its already significant investment in human resource development through better targeting and management of fuds; and (b) the importance of improving the quality of prmary education, since this level is of fiudamental Importance for the evolution of a modern industrial and stable democratic society. 3 1.8 The main objective of this report is to review issues and make recommendations for more efficient and equitable use of public funds in higher education. The report focusses on internal efficiency, management and governance, financing, and public policies towards the private sector. It does not review in any analytical depth issues related to the filt between the labor market and higher education output. It also does not cover in detail graduate education which is reported to be functioning relatively well, nor does it cover state-run higher education. 1.9 This report was written by a team led by Laurence Wolff and including Douglas Albrecht and Alcyone Saliba. In addition, the following consultants wrote background papers to this report: Herbert Guarani Calhau; Malu Gomes; Pedro Lincoln Mattos; Edson de Oliieira Nunes, Laura Dantas, and Violeta Maria Monteiro; Jean Jacques Paul; Jacques Schwartzman; Simon Schwarzman; and Teodoro Rogerio Vahl. A team led by Laura Dantas gathered data on private institutions in Rio de Janeiro and a group led by Teodoro Rogerio Vahi developed paired statistical comparisons of five private and two public universities. The report also benefitted from comments by Adrian Verspoor, Barbara Bruns, Philip Musgrove, George Psacharopouios, Estelle James, Adrian Ziderman, and Michael Michaely. While this report incorporates many of the findings and conclusions of the consultants and readers, the final product is the responsibility of the authors and reflects the policies and opinions of the World Bank and not of the individual consultants. 1.10 This report also utilizes work already published on higher education in Brazil. In particular, the Nucleo de Pesquisas sobre Ensino Superior (NUPES) of the University of Sao Paulo has been documenting issues in Brazilian higher education. In addition, IPEA has produced several monographs on higher education issues. This report also utilizes standard stadstical data and reports prepared by MEC and other public and private non-profit agencies. 1.11 This report is directed at Government policy makers and higher education leaders. However, its analyses and recommendations are of interest to a wider audience, in particular public and private higher education administrators, the academic community more generally in Brazil, the student population, and the informed public, all of which have a stake in an improved higher education system. 1.12 The main body of this report consists of two sets of chapters. The first set provides (a) a summary description of the higher education system, and (b) a review of the main issues in the sector focussing on financing issues, including Brazil's student loan scheme, on costs, efficiency, and management of federal higher education, and on oversight of private education. The second set of chapters covers (a) principles for higher education reform, (b) pacific proposed financing and institutional reforms, and (c) alternative enrollment and financing scenarios. The annexes provide additional descriptive material, a review of formulas for financing public higher education in other countries, and additional statistical information. In addition consultant reports listed in an annex are available (in Portuguese) from the authors or the World Bank. 4 II. DESCRIPTION OF THE HIGHER EDUCATION SYSTEM OF BRAZIL Structure and Hstory 2.1 Higher education in Brazil is a very large enterprise, with enrollment in 1989 of about 1.6 million students, 97% of which attend undergraduate programs. As Chart I and table 11-1 show, enrollments have grown during the last thirty years from less than 100,000 in 1960 to 1.6 million by 1989, which is equivalent to 12% of the school age population and about 58% of secondary school enrollment. This includes about 50,000 graduate students. First-year undergraduate enrollment is equivalent to 74% of secondary school graduates in the previous year. About 60% of total enrollments are in private institutions, of which one third are in private universities. Federal institutions, most of which are universities, account for 22% of enrollments. State institutions enroll 13%, with the vast majority in Sao Paulo and Parana. Municipal institutions, also mainly in the Southeast and South, enroll 5% of students. BRAZIL: Undergraduate Enrollment 1960-89 2.2 By 1989 Brazil had 93 ! BRAZIL: Undergraduate Enrollment 1960-89 universities, of which 35 are federal, Thousand* enrolling over 300,000 students, 39 - eoc 1800 _are private, enrolling over 300,00 1400 .... ......... ... students, and 19 are state or Imunicipal, enrolling about 150,000. 1200 . , .... In addition to universities, the 1000 Brazilian higher education system includes multiple-faculty institutions 800 / .... that offer programs in more than 600.~fone knowledge area (such as social Goo ............. ... ................. ......... o e l o ld e a e s c ss ca sciences and technology) and 400 .... ....... .... .. ........... ...... single-faculty institutions that offer 200 - programs in only one knowledge area (such as social sciences). O0 er 0 ~~~~~~~~~~~~~80% of the higher education 1900 1ose sro 19rs 1808 19es 1sse institutions' are of the single-faculty 8ouice AnrinJ. Tabsesi type. Within these two categories. Chart 1 most of the single-faculty institutions and nearly all of the muitiple-tfault, institutions are private. Chart 2 and Table 11-2 summarize the current enrollment breakdown by governing authority and institutional type. 2.3 Programs of study vary from 4 semesters, such as in teaching programs that prepare elementary school teachers, to 12 semesters, such as in professional programs that prepare medical physicians. Degrees are offered in teaching, vocational areas, by discipline, by profession, for example), and for military or religious careers. I MEC statistics counts each *isolated" faculty as an institution, although, in the private sector, several 'isolated' faculties nay be under the control of a single governing board and, thus, actually constitute a single institution. It is estimated that the total number of private non-university institutions would decrease by as mu. h ^ 50% if the governmng board were the inclusion criterion. 2.4 Almost two thirds of students Total Higher Education Enrolilment Dy Government Authority are enrolled in the broad field of arts, and Institutional Type, 1989 social sciences and humanities. Exact Thousand _ sciences (111 %), engineering and technology (10%), and health sciences 800 ..... .. (12%) enroll about one third, with e00 | agriculture holding only 2-3% of the enrollment. In 1989 the system 400 produced over 220,000 graduates, about half from universities. Assuming 200 . a five-year average for completion of a Fe dl program by students, and given that Federal State Municipal Private the 1985 freshman enrollment was [NonUniversities 16.629 57.661 53.771 670.785 around 400,000, 50% is a rough FUnolvealts 9323.676 154.38 21.63 383.612 estimate of the graduation efficiency M Univerltites = Non-unlveretites rate in Brazilian higher education. This rate is relatively low considering outot Table 11-2 that Brazil requires competitive entrance examinations (vestiyA ihr) for Chart 2 all higher education institutions. Table 11-3 shows the enrollment breakdown by level and field of study for undergraduate and graduate students. 2.5 In the mid-1960s the Brazilian Government decided to meet the social demand for higher education by encouraging the establishment of more private institutions, mostly of the non-university type, and opening new federal universities, mostly in the capitals of the poorer states, where none existed before. The implicit policy goal was to consolidate a relatively elite public university system, offering undergraduate and graduate instruction free-of-charge to the country's best students and emphasizing research, while creating a private sector providing mass education. However, while federal institutions generally offer better instruction than private institutions, only a few of them undertake significant amounts of research. While Brazil was pursuing this model, other Latin American countries, such as Argentina and Mexico, were opening the doors of public higher education to nearly all secondary school graduates and creating mass public education systems paired with small private systems. 2.6 Between the mid-1960s and mid-1970s, Brazilian higher education exhibited continued fast growth, averaging around 20% per year, as seen in table 114. Although all sub-systems grew in student enrollment, private enrollment increased most rapidly, as table II-5 indicates. In 1965, 38% of undergraduate enrollment was in federal institutions, 17% in state, 1% in municipal, and 44% in private; by 1983, these proportions had changed to, respectively, 24%, 10%, 6%, and 60%. By 1989, federal and municipal enrollment shares had decreased to 21% and 5%, while the state and private shares had slightly increased to 13% and 62%. Since 1983 enrollment in federal institutions has declined slightly, while state and private institutions have grown steadily, as table 1-6 shows. 2.7 Brazil's contemporary higher education system of a relatively high quality public system is similar to the current higher education system in Japan, Indonesia, Philippines, Chile, and Colombia; it is unlike that of the United States, which is predominately public and where the highest quality institutions are private, and that of most European systems which have small or non-existing private sectors. Table I shows private sector enrollment ratios in these countries as well as some other countries with which Brazil is often compared. 6 2.8 Brazil's education system is Higher Education Enrolmments in Private Sector In highly differentiated by region, as table Selecd CQunte 1I-7 shows. In the poorest region, the Northeast, only 4% of the school age population attend higher education. Enroument in The public sector predominates in the _t _on/Count_ r Private Sator North and Northeast, while the private 29 sector holds the majority of the Africa - enrollment in the other regions, EMENA 3 especily in the Southeast, where it LatinI Arnenca 34 . has more than 2/3 of the students. France lThere is very little federal and very Aqmtina ~~14 strong private (90% of enrollment) USA presence in Sao Paulo, where nearly Mexico 23 all of public higher education Chile 52 enrollment is in state schools. India 58 Indonesia Ss Bazi S8 2.9 Although federal and state Colombia 60 institutions do charge small fees for Korea 66 student services, and also do contr.at Japan 78 specialized services (research, training. -- -ouroes Regiona and Aia counries: an atechn.cal assistance, and, in the casc m.t 19:9. Sr: gna sctri ingat, medical school, health care) to publi.. .......:.;. Fanand apan: OECD MonograplnHh on; . :.and private agencies, the constitution d : Educaon, 1989 and 1988, rcdicsctl specifically prohibits the charging of USA: US Department of Education Annual tuition. Therefore public universities Statistics, 198. are mainly financed directly with AChilen Baa, 1990. governmeDt funds. In contrast, privstc 'h4~. Bhwuier, 1990. institutions are financed mainly through tuition and fees, although Table I some private universities also reoet%e Table 1 government funds (mostly for graduJte programs and research) and also contract services to public and private agencies. Students attending private institudons can also benefit from public funds by applying for educational loans and/or scholarships. 2.10 Federal expenditure on federal universities2 in 1988 averaged US$7,930 per student, atbxut fifty times higher than expenditure at the primary and secondary levels (roughly US$150 per stuJcnt: This figure is estimated at US$6000 if the costs of university hospitals and pensions are excluded This level of higher education spending per student is higher than that of other Latin American cotunries and it is similar to that of many developed countries. Unit expenditures in private institutions range from about US$4,500 in the best Catholic universities to a low of US$500 in u'ce non-university private institutions. Table 2, which summarizes enrollments and estimates costs he sector, shows that total public spending in higher education is equivalent to 1.3% of GDP and it accounts for approximately 80% of total spending in higher education in Brazil. These figures do t include spending on research in universities by agencies outside the Ministry of Education, whi,h equivalent to another 10-15%. 2 Figues an based on official exchange mtes. .7 Enrollment and Estimated Cost by Hiher Education Sector. 1989 Estimated Estimated Unit Cost in Total Cost in Total Cost Sector Enrollment % USS USS billion as % of GDP Federal 340,203 (22) 7.930 2.7 0.9 State 211,946 (13) 4,000 0.8 0.3 Municipal 75,434 (5) 2,000 0.2 0.05 Private 943,276 (60) 1,300 1.2 0.4 Total 1,570,860 3.200 4.9 1.7 Source: Enrollment: Annex Table 11-2. Costs: Federal costs are based on Paul and Wolync. State costs sre pr-rated estmes based on Paul and Wolynec and on case studies undertake by Rogerio VahL Municipal costs are estimates based on Vahl. Private costs are based on Vahb and oDa tudic in Rio de Janeiro undertaken by Edson Nunes. Because of volati exohangerates, a.well as incomplete data, these estimates should be considered indicative only. Thy not include expenditures for scholarships, research and salary supplements for researchers, all of which are estimated at US$551 million, and which go to individuals and to publio and private universities and research institutes. Table 2 2.11 In Brazil a large share of total public spending on education goes to higher education (Table 11-8). Only 2% of total public education enrollment is in higher education, but an estimated 23% of total public (federal, state and municipal) education spending is directed to this level. The Ministry of Education alone spends about 60% of its budget in its federal institutions, which serve only 22% of higher education students (Table 11-9). 2.12 Assuming that enrollment in graduate education as well as percentage of faculty with advanced degrees are proxies for output quality, it is possible to differentiate several tiers of higher education institutions in Brazil. On this basis, as well as on the basis of informed judgments by relevant observers, the two state universities of Sao Paulo (USP) and Campinas (UNICAMP) are in a category by themselves. These are by far the highest quality institutions in Brazil and boast significant graduate programs (26% of the enrollment in graduate education in Brazil is in these two institutions). Much of the research done in these institutions is of international quality. Under- graduate education is also considered by far the best in Brazil. Science and engineering constitute major portions of enrollment. Total enrollment in these two institutions is around 50,000. 2.13 Another tier is formed by a relatively large number of public and private universities that provide undergraduate education of varying quality. Only a few of these institutions have significant 8 graduate programs: the federal universities of Enrollment by TYPe of Institution - 19891' Rio, Minas Gerais, Santa Catarina, Sao Carlos, Rio Grande do Sul, and Pernambuco, as well as the Catholic Universities of Rio and Two High Quality State Universities 50,000 Sao Paulo (PUC-Rio and PUC-SP) enroll 7% (USP and UNICAMP) or more of their students in graduate Six Federal Universities with Research programs. Total enrollment in these and Graduate Orientation 92,000 somewhat research oriented institutions is around 92,000 in the federal universities and Two High Quality Catholic Universities 45,000 around 45,000 in the private institutions. Enrollment in the remaining federal Other Federal Institutions 268,000 universities (that is, those without significant Other State and Municipal institutions 218,000 graduate education) is around 215,000. Some of the stronger institutions among this group Other Private Universities - are the Universities of Brasilia, Parana, Para, Confessional and Community 191,000 Juiz de Fora, and Mato Grosso dAo Sul; Other Entrepreneurial 188,000 state institutions of similar quality enroll Private Non-University Institutions 519.000 140,000. The remaining private institutions can be divided between universities Total 1,571,000 (enrollment 310,000), which generally offer undergraduate education of reasonable quality, and faculties, most of them of the /' In descending order of perceived quality. isolated type, which offer low quality, mostly Source: Authors' estimates, based on MEC statistics. evening, courses in the social sciences and law, mainly for a lower middle class clientele (enrollment 580,000). Table 3 is a breakdown Table 3 of enrollments by these tiers of higher education in Brazil. Equity and External Effncy 2.14 Eauiy. Several studies have documented that the vast majority of students in all higher education institutions, both public and private, are from the middle or upper classes. Within higher education, public institutions have students of the highest socioeconomic status. A recent survey of income of parents of federal university students (Table 11-10) shows the following breakdowns (1989) in terms of minimum salaries: 0-3 minimum salaries, 18% of students; 4-10, 37%; and 11 and above 44%. This compares with an estimated 10% of the general population earning more than ten minimum salaries.3 Socio-economic class within higher education varies by subject studied. A study (Paul, 1991) shows that over 30% of enrollees in day engineering, biomedical studies, and law in a private institution in Sao Paulo had fathers with higher education, compared to 13% in teacher training and 16% in night courses in law and social sciences. Fully 61% of students in the federal university in Ceara studying civil engineering had fathers with at least an undergraduate degree. This compares with 20% or less for areas such as literature, nursing, geography, and pedagogy. For Brazil as a whole about 8% of the adult population has attended higher education institutions. The student 3 Meyer (1982), Farret (1985), Saliba (1990), Gomes (1990), and Paul (1991) have do. jented these income inequalities. Estimates of income distribution in the general population are from Caniargo, et al. (1990). The top decile of the labor force earned an average of 17.1 minimum wages in 1989. The second decile earned an average Df 5.6 minimum wages. 9 profile of the moi . prestigious private institutions (some Catholic universities) is similar to that of the public institutions. Table 4 provides a summary of recent studies on the socio-eonomic status of higher education students. Background of Entering Students for Severs, Brazilian lnsktu;tLos AMended Father with PubUo Higher Secondmy University of Mogi das Cruzes 21% 58% (Private Institution in Sao Paulo) Day classes in Engineering, Biomedical Sciences and Law 34% 41% Day classes in Teacher Training 13% 42% All night classes 17% 62% Federal University of Ceara 33% 20%. Engineering, Medicine, and Data Processing 53% 12% Administation, Geography, Nursing, and Literature 1S% 31% State University of Ceara 20% 32% Catholic University of Ceara 31% 10% Source: Paul, 1991. Table 4 10 Educedon A d lucofer in Brazil F_ At0~~~~31 140 I~~~~~~~~~~S SA atO 32-s - . ,. a I X 61 6, nvuL*pdu f Mm yws Mm wubegs f_4 niburnaug qfaM& - (auwarnu (pamm) 5sl.w 1/ 211 63 0.8 'i4..II 3.1 ,40 7.4 IA_1 34 3.848 10.i4 1-2 4.1 4143 32.8 2s3 494 10,38s 1t7. 3-5 4.1 14.51 l S-I. 0. 2Mb 93 1015 10.9 51,34J 3.3 iS-IS_ 1.30 73SJU5 * NOs DM kw do sib a arm ab and. pb am u of _d 180 3W saw sllbll*|11 '& Cm.. S d. E, Imb.ma. sC33w I1 2.15 Earnings of Graduates. As can be expected the earmings of those with post secondary education of any sort are high. Data from the 1980 census shows that the income of those with 13 or more years of schools is more than fifteen times those with five years of schooling and more than 180 times the income of those with less than three years of schooling. Chart 3 below summarizes the situation. 2.16 Studies show differences in earnings between graduates from low and high prestige courses of study but little difference between public institutions and the higher prestige private institutions (Paul, 1991; Velloso e Bastos, 1984). A tracer study undertaken by MEC4 provides some additional understanding on the relationships between the labor market and higher education output. Overall there is insignificant unemployment among all higher education graduates. However, only one-third to one-fourth of higher education graduates are youths who have never worked. These generally attend the best institutions and then enter their chosen profession. The more typical graduate, especially in the lower quality private institutions, lives in the Southeast, is about 30 years old, worked for several years after graduating from secondary school, and then decided to go to school in the evenings. Most of these graduates continue in the same job they had while studying, possibly with some salary increment or some modest increased aspirations because of their degree. These people are paying for low-cost courses and getting modest perceived benefits from them. 2.17 There is little information on private and social rates of return to education in Brazil. Ona study (Psacharopoulos, 1981) reports private rates of return of 25% for secondary education and 14% for higher education and social rates of return of 24% and 13%. A more recent study (Leal and Werlang, 1989) reports private rates of return of 16% for both higher and secondary education, 12% for full primary education, and 16% for four years of primary education during the period of 1982-86. Graduate Education and Research 2.18 Graduate education is a small but important part of the higher education system. Current enrollment is estimated at about 50,000, the bulk of which is in master's programs. Roughly one third of the graduate students are pursuing a degree in the 'hard" sciences (including technology). About half of the graduate enrollment is in federal institutions and one third is in state institutions, mostly (97 %) in the state of Sao Paulo. There are more than 1,300 master's and doctoral courses, or an average of 40 students in each degree program. Graduate education has recently grown rapidly, and is of relatively high quality compared to undergraduate education, to a great extent because of a relatively effective competitive funding system operating outside MEC's regular funding of higher education institutions. 2.19 An entirely different funding mechanism exists for graduate education and research in both the public and private sectors, which is based to a great extent on open competition and peer review Public and private universities and non-profit research institutes get funds for graduate education jnJ research. These funds may not appear in their budget if they are received through "foundations'. The most important Government agency financing (and undertaking) science research is the Coun.ii for Scientific and Technological Development (CNPq). The CNPq provides funds for researchers I-n the form of salary increments and also provides support in the form of research grants, usually to individual researchers and for relatively small amounts. Along with its grant program, CNPq administers eleven separate research institutes and a large scholarship program (at present, there 4,re 30,000 scholarships inside Brazil and 3,500 overseas). CNPq has 1,000 employees, and has a mcik%gt 4 Reported in Claudio de Moura Castro (1990). 12 science planning and analytical capacity. Each year, 40,000-50,00O applications for grants come in to CNPq's offices (including travel grants), along with 100,000 applications for scholarships. CNPq provides for a third of the applicants in each category. The total budget in 1989 was US$291 million most of which went to top off researchers' salaries. An estimated US$50 million went for research grants. CNPq uses a peer review system to evaluate research grant requests, which are generally unsolicited, and approves such grants about four times a year. 2.20 The Agency for Financing Studies and Projects (FINEP) is a public corporation which acts both as a conventional research foundation and as a development bank lending funds for technological innovation in Brazilian industry. FINEP was set up in 1967 to develop Brazil's engineering industry. FINEP has 700 staff, more than 100 with doctoral or masters' degrees, and in 1988 supported 1,840 projects split almost equally between scientific research and technological development. FINEP also functions as the Executive Secretariat for the National Scientific and Technological Development Fund (FNDCT), a major government source of grant funding for infrastructure, new buildings and new equipment in university based science and technology research. In 1989 about US$60 million was expended under FNDCT. Traditionally FINEP makes awards under FNDCT on the basis of evaluation by its own staff, as well as some ad-hoc reviewers from the science community. In addition for many years, FINEP, a Rio based institution, has been providing large scale institutional support to an engineering center at the Federal University of Rio de Janeiro as well as support for science research at the Catholic University of Rio. Almost 90% of FINEP awards are for continuation of ongoing reseatch programs. In 1992, FNDCT funding was severely reduced. 2.21 The Agency for Training of High Level Personnel (CAPES) of the Ministry of Education and Culture (MEC) is the coordinating agency for graduate education in Brazil. CAPES's principal instruments of support have been grants and fellowships for study in Brazil and abroad for individuals preparing for careers in university teaching and research (in 1989, CAPES supported around 14,000 students at Brazilian universities and 2,000 students in North America and Europe). CAPES certifies and rates all graduate programs in terms of overall quality and provides support to 700 graduate courses (two-thirds of all the courses in the country). It also runs a program to help build up institutions so that they can offer new courses. In 1989 CAPES' total budget was equivalent to US$200 million. 2.22 The overall total budget of these three agencies in 1989 was US$551,000,000. While this amount is equivalent to 20% of MEC's annual expenditures, a large amount of these funds go to private universities, especially PUC of Rio, to non-profit research institutions outside the universities, and to individuals as salary supplements and for overseas scholarships. 2.23 The Program of Support for Science and Technology (PADCT), partially financed by the World Bank, emphasizes giving scientists a greater role in sectoral decision making, establishing a strong and open system of peer review for award of research contracts, encouraging greater interaction between basic and applied research, and supporting multi-year integrated research and training projects. The program is expected to fund a total of US$300 miltion in science projects over the next five years. FINEP, CNPq, and CAPES are responsible for contracting with individual researchers and for monitoring research programs. Private Higher Education 2.24 Private higher education has existed for over a hundred years in Brazil. Its participation in the overall system increased during the late 1960's and 1970's,when the military regime, confronted with strong pressures for increased enrollment in public institutions, as well as political opposition 13 centered in the public institutions, resolved to permit the opening of large numbers of new private institutions. During that time higher education, both public and pri /ate, increased rapidly, but the greatest increases were in private non-university institutions. 2.25 Since the 1970's,a majority of students have been enrolled in private institutions. The accommodation of social demand for higher education in Brazil now relies heavily on the private sector, which has expanded despite the fact that the federal government regulates the number of first- year students that a private institution can admit and the tuition value it can charge to students. 2.26 The private sector in 1987 comprised 72% of the institutions, offering 67% of the first year openings, enrolling 60% of the students, and supplying 67% of graduates. Compared to public institutions, the private sector emphasizes undergraduate education and humanities, social services, and teaching. It has 75% of the student enrollment in humanities, 66% of languages, 57% of exact sciences and technology, 14% of ag:icultural sciences, 43% of health sciences, and 51 % of arts. For teacher training the private sector has 67% of teaching programs, with 69% of students and 75% of graduates. 2.27 Although the private sector concentrates on undergraduate education, it is also an important partner in graduate education: 11% of programs and almost 20% of students. However 60% of private sector graduate enrollment is concentrated in two universities, the Catholic Universities of Rio and Sao Paulo (PUC-RJ and PUC-SP). In comparison, about 1/3 of the federal universities have no graduate education, and 50% of graduate enrollment in federal institutions is concentrated in three universities, in Rio, Minas Gerais, and Rio Grande do Sul. CAPES' program of peer evaluation and rating of graduate programs shows that graduate education in the private institutions is of quality comparable to that of many federal universities. Private graduate education receives 10+ % of the funds allocated by the government agencies CAPES, FINEP, and CNPq and supplies 15% of theses and dissertations. 2.28 In the 19z^'s and 1970's private higher education grew at 15% per year, which is much faster than the second&. system, with the result that a more heterogenous clientele, including students with deficient preparation, now attend private institutions. Growth over the past six years has only been 2% per annum as a result of the economic downturn as well as continuing restrictions by CFE on opening of new programs. 2.29 Private Universities.' PUC-Rio and PUC-Sao Paulo enroll 25,000 students, offer high quality undergraduate and graduate education, and produce research. PUC of Rio has a student teacher ratio of 10:1, which is similar to that of federal universities; 17% of its enrollment is in graduate education. PUC of Sao Paulo has 15% in graduate education and a 16:1 ratio. These two institutions were founded many decades ago, have large numbers of graduate students, student- teacher ratios lower than in many other private but higher than in public, some full-time faculty, and costs higher than many other private but lower than public ones. 2.30 In private universities other than PUC-Rio and PUC-SP, 80-90% of teachers do not have graduate degrees; only about 15-20% are full time; 63% are hired on hourly bases for teaching only (no research expectation) and receive salaries of about US$ 200 per month. In contrast, in public universities 55% of teachers did not have a graduate degree in 1988; and 77% were full time. With 5 This section draws (though not exclusively) on compatative case studies, undertaken by Vahl (1990), of four community universities, including 2 Catholic universities (PUCs). 14 regard to student flow, the ratio of 1985 first year openings over 1989 graduates was 78%, compared to a public university ratio of 63%. Private universities use facilities, libraries, and laboratories, more than twice as intensely as public institutions. FTE studentlteacher ratios are 24:1 and unit costs average about US$1,000 compared to 8.8:1 and around US$7930, respectively, in federal institutions. Table 11-Il provides detailed breakdowns of costs and efficiency in four private universities, as gathered by Vahl. 2.31 Primate Non-University Intitutions.' A large number of the non-university institutions offer all levels of schooling, from pre-schoo! to tertiary education, thus maximizing the utilization of their facilities and direct administration systems. They usually do not have campuses, simply occupy buildings, sometimes in more than one location, often offering night courses only in buildings operating as secondary schools during the day. Some have good facilities but the majority do not. Many single-faculty and multiple-faculty institutions aspire to university status, because of higher prestige and relatively trore autonomy. The vast majority of these institutions are located in Sao Paulo and Rio. 2.32 Table 11-12 provides detailed summary information on a survey of 13 private non-university institutions in the state of Rio. Like their public counterparts, private institutions do not maintain sophisticated record keeping practices and, hence, are ill prepared in the computation of unit costs and other efficiency indicators. Yet the richness of the first-hand data allows for the identification of some trends. Based on this data, 73% of staff in these institutions do not have graduate degrees. The graduating class is 18% of initial enrollment four years previous. Student teacher ratios average 26 students per FTE teacher. Only 8% of teachers are full time, with the vast majority (55-80%) paid on an hourly basis. Teacher salaries are directly correlated to the number of actual classroom instruction hours taught per week. Salaries are paid only during months of the academic term. Personnel salary costs consume about 60-70% of the budgets, of which 70% (50% of total) go to faculty salaries. Cost per student varies from US$350 to US$3,500 per year. 2.33 Income f.om tuition accounts for 50% to 96% of an institution's budget, with an average of approximately 80%. The federally-funded student loan-program brings in 5-30% of the budget, with an average of about 13%. Other sources of income are provision of services, such as health services in hospital-schools and at-large education programs contracted by local corporations for the training of their employees, as well as the investment of tuition money in the capital market. Since these institutions do not keep detailed accounts it is difficult to reconcile income and expenditure. Many institutions award partial or full tuition waivers to students in financial difficulty. Waiver levels vary from 5% to 45%, averaging 10%. About a third of the scholarships provide 100% discounts, the rest vary from 15% to 90%, with about a third falling at the 30% discount rate level. 2.34 CQmmunity Instittigns. One group of private institutions is that of "community" institutions, which see themselves as non-govermmental public entities providing social services. This group actually consists of two types of institutions, both sharing the same mission, but with distinct affiliations. The most traditional sub-group is that of the confessional institutions, founded and administered by consortia of religious orders, generally in existence many years prior to the 1968 reforms and consequent expansion; many of them are universities, but not all. A smaller and newer sub-group of community institutions is that of secular institutions, founded and administered by community-based groups with no religious affiliation. Total enrollment in confessional institutions 0 This section draws on case studies of 16 higher education institutions of the state of Rio de Janeiro undetA. n by a team of researchers led by Laura Dantas of SBI. The tables in Annex 3 show the detailed data collected. 15 was around 225,000 in 1989 (25% of total private enrollment). They enjoyed a great deal of public subsidy in their early years and have faced budgetary crises during the last decade or so, when that source of income dried up. They continue to seek a return to the times when they could benefit from direct government funding. They have been successful in ensuring that the 1988 Constitution permits public funding of certain private institutions.7 2.35 A second group, the majority of private institutions, can be defined as "entrepreneurial' institutions, which consider themselves as educational enterprises seeking a profit to all intents and purposes. M- iiy general directors of these schools are their own founders. Their background is diverse, from teachers to retired army generals to lawyers. Some were established by elite families (such as UNIFOR in Ceara, Mackenzie in Sao Paulo, and SBI in Rio de Janeiro), others are owned by business partnerships capable of cost-effectively delivering low-cost programs (such as UPns in Brasilia and Objetivo in Sao Paulo). Total enrollment in 1989 in entrepreneurial institutions was around 700,000 (about 75% of total private enrollment). Unlike in the primary and secondary education sectors, the law governing private higher education does not allow institutions to be profit- making. However, "owners" figure out creative ways to capture profit. One way is having many relatives employed by the institution, some hclding multiple full-time jobs. For instance, in the case of a Rio-based proprietary institution, a husband is president of governing board, the wife is vice- president of the same board and also general director of institution, and a son holds a directorship position in the board. Table 5 summarizes private sector enrollment by type of institution. Private Sector. Enrollment Breakdown by Tvye of Institution. 1989 (In thousands) Tvpe of Institution _Number Entrepreneurial Institutions University 188 20O Non-University 519 55% Community Universities Confessional 170 18% Secular 66 7% Total Private Student Enrollment 943 100% Source: Authors' estimates based on MEC statistics. Table 5 2.36 Demand for Private Higher Education. Considering that public .nstitutions have higher prestige and charge no tuition, they are almost invariably the first choice of students. Private higher education becomes the choice when: (i) the student is unable to gain admission to a public 7 Art 213 states that community-based, confessional and philandiropic schools may receive public funds. 16 institution; (ii) the student is a full participant in the labor force and, thus, cannot afford to attend daytime classes; (iii) the local public institution does not offer the course of study desired by the student; and, in some cases, (iv) the student sees the private institution as better quality in selected courses of study. Once the first-order decision is made to seek admission in the private sector, a second-order decision is often required, because most locations have more than one private institution. The profile of students in the best Catholic universities is similar to that of public institutions. However, most of the low quality, non-university institutions are attended by older, less prepared students, most of them working full time, and staffed by less prepared teachers, sharing low intellectual demands on each other the ("pacto currupto"). The widespread perception (largely unmeasured) is that the majority of these schools are "diploma mills" offering formal credentials but very little content. Hard data on this overall impression, through studies of learning or income after attending these institutions, is not available. 2.37 Several factors account for the phenomenon of people paying for low-quality private higher education. First, there is demand overflow; demand for higher education outweighs supply by at least 2:1, but the ratio of applicants by opening is as high as 60:1 in high prestige programs, such as engineering, dentistry, and medicine, at some institutions. Second, where higher education (and even education in general) has a high scarcity value, acquiring even low-quality higher education is better than acquiring no higher education, especially in a labor market where the majority have a less thin primary level educational attainment. Third, the public sector of the labor market is dominated by credentialism, in which career ladders for civil servants are defined by credentials and seniority (as well as patronage). Hence, -he Government actually "supports" low quality private higher education by assigning an intrinsic value to the diploma. Although the private sector of the labor market is generally more interested in skills than in credentials, it still has to comply with regulations whiih ire created by trade associations and mandated by the public sector. For example: the law regulating !e profession of journalism states that corporations producing any form of publication (such as client newsletters) must employ certified Brazilian journalists, whether they need them or not, an arttli;e t the joumalists union to guarantee jobs for its members; companies often work around the legislit; n by hiring a card-caring journalist who receives token compensation for agreeing to have his/her rxmt on the company's masthead, but who does not do the work. Similar legislation exists for many ott :t professions. Such labor market regulation guarantees that investment in any private higher educoti. n pays off, given that students attending private institutions receive diplomas with the same nominil certifying value as one obtained from a high quality (public or private) institution. 2.38 Internal Management of Private Institutions. The decision making flow in private institut : is simpler, the organizational chart is flatter, and resource allocation is leaner. A typical institutwi n governed by a general council, representing the "owners",whether a religious order, private cit zzv., or families, and it is managed by several area (financial, academic, administration) directors head .dJ a general director. The governing board, which commonly makes all the budgetary decisions. c!: s nominates the general director for renewable terms varying from 2 to 4 years, or even for indet,! - length terms. In several institutions surveyed, the directors have held their position for twenty ! or so. This process is quite different from the direct election for single non-renewable terms currently practiced in the federal and state universities. Generally board members are not remunerated for their service, except the administrative managers. Some boards include a stud.:rt representative. The general director nominates the other directors. Decision making in confe.. institutions tends to be more complex than in secular ontes, and the leadership turnover tends l.} higher than in secular institutions, apparently because the number of docision making parties t.*' be larger in the former. In the case of one Rio-based school, there have been 15 general dirtx,. the last 29 years. 17 2.39 Compared to the public system, private institutions have better internal efflciency and management, a leaner organizational chart, better decision making flow, less discontinuity of administration, agility in crisis solutions, higher student/teacher and student/staff ratios, flexibility to hire/fire teachers, determine their salaries according to market values and costs levels, and flexibility to adapt quicker to labor market needs and thus change curricula (in tLe absence of government regulations). Private institutions are often accused of paying inadequate attention to quality and of making excessive profits. Paradoxically, government restrictions on tuition increases as well as the pervasive credentialism of the system are probably the main causes of lack of attention to quality. There .ias never been a systematic demonstration of 'excess*profits, but Government restrictions on entry and control of enrollments most likely lead to "rent-seeking", as has been common in much of the Brazilian economy. 18 III. ISSUES IN HIGHER EDUCATION 3.1 This section seeks to document the main issues and problems of the current higher education system in Brazil. It covers the key areas of the costs, governance, and management of federal higher education, the relationships between the federal government and private higher education, and overall financing of the higher education system. Costs of Federally Funded Higiher Education 3.2 There is much controversy within Brazil on whether or not the costs of higher education are excessive. Based on comparisons with state and private institutions in Brazil as well as through some international comparisons, it can be concluded that, while federal institutions emphasize higher cost programs in science and technology, and some of these institutions provide significant programs of graduate education, overall the unit costs of these institutions are significantly higher than private institutions in Brazil providing similar programs. The nmain factors accounting for such high unit costs are the low student/teacher and student/non-teaching staff ratios. There are only minor differences in expenditures between research and graduate oriented federal universities and purely undergraduate institutions; nor are there any economies of scale among federal institutions. In addition physical facilities in the federal institutions surveyed are grossly under-utilized. 3.3 International Comarisons. Because of Brazil's rapid inflation and extremely variable exchange rates, it is difficult to arrive at reliable dollar estimates of unit costs. In the years 1988 and 1989, in particular, especially, the difference between the official exchange rate and the parallel exchange rate (more than 40% in 1988) makes reliable international comparisons very difficult. Estimated average unit costs in federal institutions in 1988 are US$7,930 per student, based on work undertaken by Jean Jacques Paul and Elisa Wolyneck (1990), which uses official exchange rates prorated on a monthly basis during 1988.1 This estimate is arrived at by dividing the MEC budget per institution by the number of students. Therefore, it may include a variety of non-educational and research activities, such as running hospitals, museums, etc. and paying pensions. Francisco Gaetani and Jacques Schwartzman (1991) argue that these unit costs wouid be about 40% lower if the parallel rather than the official exchange rate were used, which would put the federal institutions in a more favorable light in international comparisons. Gaetani and Schwartzman also argue that unit costs would be reduced by an additional 25% if the costs of university hospitals (about 10% in many institutions as well) as the costs of payments to retired teachers (estimated at 15% for some institutions) were discounted from the overall budget. However, retirement costs may be considered a part of wages. In addition, most developing countries have a fluctuating gap between the official and parallel exchange rates, and the cost of retirees as well as hospitals are often included in international estimates. Tables 111-1 and 6 provide unit cost estimates for selected developed and developing countries, and it compares them with estimates for Brazil of unit costs based on the official and the parallel exchange rates. Salaries in Brazil for professionals average one third of those in developed countries. However, even making the exchange rate adjustment suggested by Gaetani and Schwartzman, Brazil's uait costs are equal to about half of those in the USA, Japan, and Australia, and are significantly higher than those of its Latin American neighbors. While it would be inappropriate to compare Brazil with Mexico and Argentina, which have followed tht, path of opening the doors of public institutions to all secondary schools graduates, Chile and Costa Rica are two Latin s Unless otherwise noted, this report relies on the official exchange rate. 19 American countries which have retained relatively high quality public institutions at lower unit costs than Brazil. 3.4 Because the main determinant of costs in higher education world-wide is teaching, which normally accounts for 70% of higher education costs, the use of student-teacher ratios, which are not subject to problems of exchange rates corrections and are relatively stable over time, provide another way of doing cross-country comparisons. Full-time equivalent (FTE) student-teacher ratios in Brazil's federal institutions average 8.8:1. Table 6 compares this ratio with student teacher ratios in selected countries. With the exception of Colombia, Brazil's student teacher ratios are the lowest among the countries comparedg'. 3.5 Another element of importance is the ratio of students to non-teaching personnel. While no international comparisons are available, the current students to non-teaching staff ratio in Brazilian federal institutions is 3.9:1 (it was 5.1:1 in 1980). In short, despite the usual caution attached to international comparison, Brazil's federal system is undoubtedly one of the high cost systems in the world, once one takes into consideration Brazil's salary levels. The main determinant of these high costs is the large numbers of teaching and non-teaching staff compared to students. 3.6 Comgarisons within Brazil. Table 111-2 provides estimates of FTE student teacher ratios for all federal institutions and for state and private universities. The overall student teacher ratio in federal institutions is 8.8:1,compared to 10:1 in state universities and 29:1 in private universities. A review of the data shows the following: (a) Federal institutions vary greatly in student teacher ratios, from about 4:1 to about 12:1. A few small specialized institutions in agriculture and medicine have around 4:1 ratios. (b) In tenns of student-teacher ratio, there are no economies of scale among the 23 universities. In fact several of the largest institutions, such as the Federal Universities of Pernambuco and Paraiba, have student teacher ratios of less than 7:1, and the largest institution, the Federal University of Rio, has a student teacher ratio of 8:1. (c) The highest student teacher ratios in federal institutions appear among the youngest institutions. While this may in part be a result of a lack of graduate education, it may also be a result of an inadequate ability to lobby MEC to authorize new positions. (d) The state university of Sao Paulo (USP), the largest and best institution of higher education in Brazil, has a student teacher ratio of 10:1. UNICAMP, with the highest percentage of enrollment at graduate level (31 %) has a student teacher ratio of 7:1, and UNESP also 7:1. Most of the other state institutions have student teacher ratios of 10:1 or above. (e) Graduate education, which requires lower student teacher ratios than undergraduate education, affects the student teacher ratio in UNICAMP, with 31% of enrollment at the graduate level, and USP, with 25% of enrollment at the graduate level. Among federal institutions only the federal universities of Rio and Minas have more than 9% of their enrollment in graduate education 18% and 9%). 9 The argument made by Frncisco Gaetani and Jacques Schwartzman that comparisons with North American institutions are inappropriate, because much teaching in the US is done by teaching assistants, results in only marginal adjustments when student-teacher ratios are based on FT1E rather than headcount 20 Hicher ducaon: Student Tacher Batos and-Unit Costs in SglctLed Coutntris (a) Student-Te,ache7 Rato Total Student SIr Country Ewilti Eor°ilmuzt Bai year Venezuela (pub. univ.) 16,554 k 280,507 16.1 1989 Chile (pub. univ.) 10,388 W 112,193 10.8 1989 Argentina 41,804 707,016 16.9 1986 Chile 11,603 132,254 11.4 1984 Colombia 37,557 331,1S0 8.8 1987 Japan 195,276 1,965,023 10.1 1986 Korea 33,340 1,040,166 31.2 1990 Malaysia 4,717 47,946 10.2 1987 Spain 48,360 900,417 18.6 1986 UK 31,432 360,800 11.5 1986 Brai (b) Fderda 25,497 224,665 8.8 1988 Stae 14,658 130,481 8.9 1988 Private 10,989 323,766 29.3 1988 USA Publio 4-yr. InsL 319.000 5,544,000 17.4 1989 Private 4-yr. InsL 218,000 2.631,000 12.1 1989 Sounx#: Unemo SWadscal ewrbook (several yeu=) MEC Siose do Ensino Superior (1988) (a) Figures for some countries are agregate (private plus public). (b) Facuty figures sre in FME. --Uni Coss in Public Institutions. in USS 9gm ntCog CSye" Japan 5,968 1985 UK Unlwersiii 12,950 1986 Polytechnics 6,160 1986 USA 8,724 1984 PhFtinpas 3,492 1985 Spain 906 1985 Australia 6,126 1987 Che 1,700 1989 Venezuek 1,625 1989 Costa Rics 1,000 1985 Brazil - using officid exchange rae 7,930 1988 Braz - using paallel exchange rate 4,760 1988 Table 6 21 3.7 Table 111-3 summarizes unit costs in 1988 for 36 federal universities. By and large, it confirms the same variations noted in the student teacher ratios. The highest unit costs are in the Federal University of Vigosa, which is a specialized agricultural institution. The three universities in Rio also have among the highest unit costs. The lowest unit c' ^ s are in institutions in the North and Northeast with little graduate education. Paul and Wolyneck (1990) estimated that unit costs in federal institutions are 60% higher than the unit costs of the University of Sao Paulo (USP), in spite of the fact that 25% of USP's enrollment is in graduate education, and USP is considered the highest quality institution in Brazil. But Gaetani and Schwartzman have argued that more than 9,000 students in USP reported as full time graduate students are in fact not engaged in full time study or research. After eliminating these students and taking out funds for retirees, he argues that the difference is only about 13%. However, in a separate paper, Paul (1989) has noted that there are also significant numbers of "ghost" students in federal institutions. The conclusion, even after accepting Gaetani and Schwartzman's revisions, is that the highest quality public institution in Brazil. USP, does not appear to be particularly cost efficient but nevertheless has lower unit costs than the major federal institutions with which it is compared. 3.8 There is additional evidence of inefficiency in the federal institutions. During the period 1983-89, the number of teachers in federal universities increased from 43,000 to 44,500, while the number of students declined from 340,000 to 318,000. Thus the nominal student teacher ratio decreased from 7.9:1 to 7.1:1. During the period 1980-1989 the number of non-teaching staff increased strikingly from 65,000 to 92,000. 3.9 Institutional case studies undertaken by Rogerio Vahl break down expenditures in 1989 hk budgeting categories, by faculties, and by departments for two federal universities, Minas Gerais (UFMG) and Santa Catarina (UFMG). Table 111-4 summarizes the information on these institwnl r.i as well as four other private institutions. The average unit cost of the two federal universities %t*d J'J is US$10,000 per student (1989 data). The nominal student teacher ratio is 6:1. Full-time tea.:her% are required to teach 8-14 hours per week and are expected to prepare classes and undertake research during the rest of the time (compared to 24 hours in the private institutions). The top v s zi . for a full professor is around US$2,300 per month in the federal institutions (compared to US$1 t. ^') per month in the private universities surveyed). The number of non-teaching staff in the two td.- &i institutions is equivalent to about 25% of student enrollment (compared to 5% of enrollment in .-" private institutions). The average salary of the non-teaching staff is roughly US$1,000 per m.,nth l the two federal institutions (compared to $300 in the private institutions). 3.10 Vahl sought to break down costs by program of study. Since universities do not keep systematic records of this sort, only rough estimates, covering teaching costs only (but not maintenance and materials), are possible. Table fl-Il shows the following average cost breakdJ -" the two federal institutions: science and technology, $9,447; biomedical sciences, US$11,985 (e g medicine US$11,844, dentistry US$13,059, nursing US$11,755); and humanities and social scin:n_s US$6,854 (e.g., law US$6,681, mathematics US$7,290, pedagogy US$9,863). It is striking that :c differentiai between science and say, law, is about 50%, compared with a USA differential rang !4 from 2 to as much as 5:1. It suggests that the decision making system is such that student tej. h.- ratios tend to even out among all disciplines in spite of the possibility of utilizing savings in lh'A . t programs to strengthen high cost programs. In fact federal universities rarely use the combinti,. very large lectures accompanied by smaller classes common to the USA. 22 3.11 Another element of interest, which has never been studied in Brazil, is that of space utilization. The two federal universities surveyed provide 28 square meters per student, compared to 6 square meters in the four private universities surveyed (rables 1-11 and 111-5). More importantly, the former hias an average of 1.6 seats for every student enrolled in the system compared to 0.8 in the private institutions. Assuming that classes can take place eight hours in a day (a reasonable expectation, with classes running 8-12 a.m. and 2-6 p.m.), and students attend about four classes of one hour per day, on average seats in the two federal institutions are utilized about 30% of the time, compared to 85% utilization in the comparable private institutions. The US average in both public and private institutions is 60%. Based on this analysis, the physical facilities in the two federal universities surveyed are adequate to serve about twice the number of students currently being served without initiating any night classes. Put differently, if the cost per square meter of construction and furnishing is about US$400 then the Government has invested about US$190,000,000 in excess physical facilities in the two universities studied. While it is difficult to generalize to all federal universities, if all federal Institutions were in fact similarly over-built, then the total excess capital investment would be on the order of more than US$1.6 billion. Since private institutions normally put aside capital funds or pay mortgages for construction and refurbishing, the costs of the physical facilities in the public sector should be annualized and added to the recurrent cost to get an economic cost per student which would be comparable to the private institutions. Assuming a thirty year life of physical facilities, a 10% discount rate, and a cost of physical facilities per student of US$11,400, then the unit costs per student in the two federal institutions surveyed should be increased by US$1,200 or an additional 15%. 3.12 The argument has been made that the higher unit costs of the federJ institutions are a result of higher quality staff, many of whom undertake research, of programs in science and engineering, and of graduate programs. It is certainly true that the two federal institutions surveyed do undertake research and extension and do have many programs in higher cost areas such as medicine and engineering. In particular teachers in the two federal institutions surveyed reportedly produced 1,034 books, chapters of books, and articles in 1989, which is an average of .24 per faculty member. Almost 10% of the students at the two federal universities surveyed are the graduate level. 55% of the enrollment in the federal universities is in high cost courses, defined as science, engineering, agriculture, and medicine, compared to 36% in the four private universities surveyed. Furthermore in 1989, the feder: universities provided for 421 specialized training and extension courses compared to about 176 in the private universities. Finally around half of the federal university faculty in the federal institutions have advanced degrees (doctorate or masters). These figures all represent higher quality input and more research output than the vast majority of private universities in Brazil. 3.13 PUC-Rio and PUC-SP are the two private ins.itutions acknowledged to be among the highest quality institutions in Brazil, with high percentages of graduate students, and with input and output quality measures similar to the two federal institutions surveyed. Available data (Table 7, from Tramontin and Braga, Vahl and Paul) show that unit costs in the federal universities are twice as high as PUC-Rio and about four times as high as PUC-SP, which have equally high quality staff and major graduate programs. It should be noted that PUC-SP focusses on the social sciences and has very little enrollment in medical or hard sciences. Therefore its unit costs should be compared with estimated unit costs of over US$6000 for the two federal universities in humanities and the social sciences (Vahl). PUC-Rio's enrollment profile is similar to the two federal institutions surveyed, since it has a very strong science and engineering program (but no medical school), much of it supported by grants from FINEP. In short the two best private institutions in Brazil offer programs similar to those of the two federal institutions at less than half their costs. 23 Costs -and Qualtv in Two Federal-and Two Private Universities- UFMO UFSC PUC-Rio PUC-SP Percent Oraduate Students 11 10 17 is Percent Staff with Doctorate 20 18 30 15 FlE Student Teacher Ratios 8.1 8.1 10.1 16.1 Student/non-teachingStaff Ratios 3.1 4.1 11.1 14.1 Unit Costs in USS 9,179 7,852 4,535 2,089 Source: Data from MEC for 1988 on student teacher ratios and unit costs are used for the public institutions (Table 111-2 and 1113). Data on the other items are, for the federal institutions, from Vabl, and for the private institutions, from Tramontin except for student teacher ratios, which are from MEC. Table 7 3.14 It should be noted that the two federal institutions surveyed are among the oldest and best federal institutions. In contrast, the Federal Universities of Ouro Preto, Mato Grosso, Juiz de Fora and Espirito Santo have negligible graduate enrollment and are reported to undertake little or no research but have FTE student teacher ratios of 9: 1 or less and unit costs ranging from US$6500 to US$9000. Governance and Management of Federal Institutions 3.15 The system of governance and management of federal higher education encourages high costs and discourages eMdency, and diversity. Under the current system federal institutions have little autonomy er decision making and the federal government has inadequate knowledge of the performance of these institutions. Vahl and J. Schwartzman have documented these issues and problems. This section summarizes their conclusions. 3.16 Personnel Po!j W. The federal universities have no policy with regard to personnel. All permanent staff a&e contracted by the universities but paid by the federal government through the Civil Service Syste,m according to civil service rules. However universities are permitted to contract personnel on a temporary non-civil servant basis. No university is permitted to establish its own plan of positions and salaries, nor to establish its own criteria for promotion, nor to establish salaries for administrative/technical personnel on the basis of market factors or productivity. Without a doubt this lack of flexibility leads to low motivation on the part of faculty. Universities are not permitted to hire staff, since the federal government sets overall hiring policy, and recently has enforced freezes on hiring. Furthermore the universities cannot fire staff except under the strict limitations of federal law for civil servants. 3.17 The "Law of Isonomy" of 1987 regularizes employment of all teaching and non-teaching staff in federal universities on a national basis. The law establishes 354 non-teaching (technical/administrative) categories as well as the regulations for teaching staff advancement in the career ladder (that is, being promoted from auxiliary, assistant, and associate to full professor). The law includes salary relationships for all positions. It requires payment of the same salary for each category throughout Brazil. Under the law promotion may be gained through getting higher degrees (masters or doctorates) or through time of service. A teacher can reach the associate rank solely on the basis of time of service. The law requires evaluation of professional output only for entering the 24 profession and for becoming a full professor. In no case are advanced degrees required (advanced degrees were required before 1987). Among those currently at the associate level, 23% have doctoral degrees and 53% have master degrees. Without the requirement for advanced degrees, these percentages will fall and many full professors will not have advanced degrees. 3.18 Based on the law, faculty are hired on three bases-20 hours per week, 40 hours per week, and 40 hours per week and exclusive dedication. All three of these types teach between 8 and 14 hours per week. It is expected that the full-time teachers, especially those with 'exclusive dedication," will undertake research. After a teacher gets one of the "exclusive dedication' posts, no formal check is made on his research output or on other employment. Furthermore it is generally not possible to take a teacher out of the "exclusive dedication" category if he does not do research, since the present legislation permits reduction in hours or firing only for 'just cause," which in Brazilian labor law would mean faults such as total inco-npetence or corruption. Legally a teacher cannot be fired because his academic work is deemed mediocre or weak. Furthermore, if a teacher were fired, there would be no assurance that a replacement could be hired, since such decisions are subject to the ongoing federal hiring regulations. 3.19 Academic Decision Makiqg. While the law states that federal universities have 'didactic, administrative, financial and disciplinary autonomy," in practice there are many restrictions. The internal statutes of each university must be approved by the Federal Council of Education (CFE). The area of greatest autonomy is in the area of teaching and research, where, since the end of the military dictatorship, the universities have been quite free. The CFE, besides accrediding all universities and faculties, establishes the minimum curriculum for each course in all public and private institutions in accordance with national standards. On average the minimum curriculum covers 94% of the total of the credits needed to obtain a diploma. Within Brazil scores of professions require a valid higher education diploma to operate either publicly or privately. With the exception of law, there are no objective national examinations for entrance to a profession, which is solely determined by the holding of a valid diploma. Sao Paulo is currently experimenting with a professional examination for medicine. UTniversities are highly constrained when making innovations in courses. since they must obey the federal guidelines on minimum curriculum. Furthertnore, while the universities are permitted to increase or decrease the number of students per course, they are presently prohibited from contracting new professors and generally they are prohibited from firing any teachers, nearly all of whom are civil servants with tenure. Finally the University level Council of Teaching, Research and Extension, composed of representatives of all teaching levels, must approve the closing down or extinction of any course. This practice helps explain the existence of faculties and/or departments with a large number of teachers and a very small number of students in areas where demand is down. 3.20 Internal Management. The federal universities are formally managed by a university council. consisting of the rector and his staff, representatives of the main academic bodies, and representati' es of students and society as a whole. A council of overseers includes representatives of academia, the overall community (including the private productive sector), and MEC. Other councils are related t,l teaching and research. There are also councils on teaching and for each department and faculty 3.21 The rector of the university is selected by the President of the Republic from a list of throd presented by the university and can be rector for only one term. Since the end of military rule, nw%t universities have a popular election in which all students, teachers, and administrative personnel ' te The candidate with the highest vote total is sent up to the President as the first on the list. With tI exceptions, the President selects the first on the list. Students and faculty have gone on strike %kh.n the second or third on the list has been selected. The proposed new Basic Education Law 'AulJ formalize this requirement. The democratic election of rectors has in some campuses resulted in h.e election of highly qualified and dynamic rectors. However, there is a strong tendency for the prr.$.% 25 to become highly politicized, with resulting political battles between various parties. In addition many candidates find it necessary to make commitments to their constituencies which at a later time make it very difficult to introduce tough requirements or measures to increase cost-effectiveness, such as requiring students to attend classes or eliminating excessive numbers of non-teaching staff-L'. 3.22 The rector of the university selects his technical administrative staff from among existing faculty and staff. These staff do not act as managers but rather as interlocutors between faculty, st-udents, technical personnel, the rector, and the government. There is no permanent planning staff in the university. The staff who work with the rector generally handle only short term budgeting problems. There is no institutional planning mechanism. Therefore institutional planning occurs by accretion based on the interests and strengths of the various departments and faculties. 'Tlhe office of the rector has no authority or capacity to make difficult decisions in resource allocations between programs. To a great extent this lack of central planning is a result of history. Universities have never had a strong centralist tradition and they have grown as accretions of independent faculties"". In addition the proliferation of large numbers of committees with overlapping jurisdictions has made it difficult for university leadership to take any controversial decision. 3.23 Effects on Oualitv. The lack of evaluation of teachers and subsequent awards or punishment based on such evaluation, as well as the "credentialism" in the system brought about by the requirement for diplomas, leads in both public and private institutions to what has been described as the "pacto corrupto" between students and teachers. For many students, obtaining a degree is more important than learning. For the teacher, the "pacto" means he does not have to prepare challenging lectures, carefully correct tests and exams and/or spend his time working with and advising students. The "pacto" is most common in programs of lower social prestige as well as in night classes. In fact, given the nature of the system, it is surprising that many teachers (to some extent in all areas but especially in the sciences, engineering, and graduate education) are dedicated and that quality is adequate. Filancing of Higher Education 3.24 The current system of financing higher education produces inefficiency, and inequity and also discourages efforts at improving quality. Specifically (a) the legal prohibition against charging tuition restricts institutional flexibility and is both inefricient and inequitable; and (b) the student loan scheme, the main means the Government has for assisting private education, has only partially achieved its goals of helping poorer students to attend higher education and has been operated without regard to efficient use of public funds, and (c) the highly centralized and inadequately planned mechanism of the federal government for financing federal institutions strongly discourages individual institutions from cost-effective use of funds. 3.25 Public Financing of Federal Institutions. MEC is the main funding source for federal universities, providing approximately 85% of the income of the federal universities. MEC's total '° The most flagrant example of politicized election of a rector was reported by various Brazilian newspapers. In this case the rector of the Federal University of Rio de Janeiro (UFRJ) hired, through temporary contracts, large numbers of administrative personnel who supported him and helped to ensure his re-election-in spite of the fact that university by-laws did not permit a second term. After a year of turmoil the rector resigned and the situation was normalized. " The case of the rector of the University of luiz de Fora, also reported in the newspapers, illustrates the lack of power of the rector. This rector resigned after the. University Council passed a resolution to increase the restaurant subsidy from 30% to 70%. With the passing of this resolution, the subsidy for the restaurant was equivalent to B2 % of the university's discretionary funds for non-salary operating costs. 26 budget for higher education in 1989 was around US$3 billion, of which US$2.5 million went to federal universities. The way MEC determines each institution's budget has an important impact on the behavior in each institution. Additional funding, especially for graduate education and research, is provided by other government agencies, as well as by private industry and by local government. A few federal universities (e.g. Brasilia, Minas Gerais) have their own endowment (patrimonia), usually in the form of real estate which can be sold or developed. 3.26 The budget of federal institutions can be broken up into two basic compone.ts, salaries for staff, who are civil servants, and much smaller discretionary budgets which institutions use to fund everything else. Personnel payments are made directly through the National Treasury and do not enter the budget of the institution, since MEC has complete control over them. Overall 94% of payments from MEC go to personnel and 6% go to other costs and to capital. The university has no control over the 94% since they are simply based on payment to staff. The 6% of budgets which institutions can control covers utilities, travel, materials and equipment, maintenance, restaurant, and library. Ordinary maintenance generally accounts for 70% of these funds. Major capital funds are also provided by MEC, often partly financed by IDB. Current capital investment is minimal. 3.27 At the beginning of each fiscal year, each institution is budgeted only a portion of its expet^ed requirements. Each institution receives these funds on the basis of 'historical criteria', e.g., the percentage of the total budget it received the previous year. The first payment is made on MarJn l. with inflation corrections. After six months, MEC provides a supplemental budget. Universities at not permitted to spend funds which have not been budgeted, so it is very difficult to plan the purchase of materials on a yearly basis. Furthermore, the universities are not allowed to keep furidi in interest bearing accounts, so in a high inflation period the real value of their funds is rapidly eroded. Purchases must therefore be made on an emergency basis. While universities have a f3ir amount of flexibility in determining how to apply non-salary funds, they appear to exercise little vr - controls on use of these funds by staff. For example universities have no internal control on use l energy, telephones, copy machines, etc. 3.28 The budget negotiations between the federal universities and MEC generally result in bud, .1 increases. But the number of staff and faculty members, the physical capacity of institutions and t.c allocations for non-salary expenditures are made without any reference to what is expected of the institutions. The five year period between 1983 and 1988 exemplifies the problem. During this pri budgets rose while enrollments fell an average of 7%. Institutions received extra funds but de%adiJ maintain their elite status. The extra per student funding could have been used to improve te..h ! or research quality, but there is no indication that this has been accomplished. Furthermore, a significant part of the non-salary budget is used to provide extra student subsidies, such as dis u .-A meals, health services and other welfare services. 3.29 The determination of budgets without any consideration of cost or of output results in A Public money has been used to purchase teaching and physical capacity that is not being used . the budgeting is done without any attention to outputs or costs, the funding increases have nit' resulted in perceived quality improvements. 3.30 Sources of Funds for Federal Institutions. Including Cost Recovery. Public higher edu.. ' i institutions are legally prohibited from charging tuition. They do, however, charge a variety *t fees for services to students as well as to the community and industry. Overall the two federad institutions surveyed get 80-85% of their funds from MEC. The remaining 15% comes from . fees charged to students (graduation fees, restaurant charges, etc.), contracting agencies (e.g . service tra-ning programs, technical assistance and studies provided to government and industr% health authorities (reimbursements for health services provided by university hospitals), and r. donors (CNPq, FINEP). Because they are subject to complex civil service regulations regardi r 27 of funds, many universities have established independent "foundations" to more efficiently manage non-MEC funds, especially those coming from FINEP, CNPq, CAPES, and PADCT for research and graduate education. The budgets of these foundations have been growing and they are providing an important element of flexibility. 3.31 Traditionally public institutions subsidize medical and restaurant services provided to students. The average restaurant subsidy in UFMG was 48% and in UFSC was 70%. The average yearly subsidy in the two institutions per student was US$31. It has been reported that the subsidy at the University of Juiz de Fora was 70% in 1989. Assuming a $31 annual subsidy for all federal students, then the total annual cost to the Government for the restaurant subsidy amounts to about US$10,000,000. 3.32 The legal prohibition against charging tuition reduces both the efficiency and equity of the higher education system. Since students do not pay for their education, they have little incentive to complete their course-work in a timely manner. Because students in Brazil's public institutions are disproportionately from the upper economic classes, the government spends far more of its money for the benefit of the rich rather than the poor. Since the provision of basic social services such as primary education and health care are not effectively reaching the poor, funding higher education only from the broad tax base results in a regressive transfer of funds from poor to the rich. 3.33 Public Financial SupDort of the Private Sector. Until the end of the 1960s, the federal government provided a significant amount of direct public support to private, especially Catholic institutions, covering as much as 50% of the costs of these institutions in some years. In the period 1970-1980 this amount changed to around 10% of expenditures. In the early 1980s, the Government effectively ended all direct support of private institutions. However, in the period 1986-1989, the Government again indicated its willingness to support "community" institutions, increasing to 7% of the total expenditures of about 20 community institutions in 1987. In 1989-90 such direct support has again fallen to almost nothing. The Government continues to support graduate education in a few private institutions through FINEP, CAPES, and CNPQ. Of particular note is FINEP's long term support of science and engineering in the Catholic University of Rio, which has helped make it the best private institution in Brazil. 3.34 The Student Loan Program. Recently the main source of public financial support to private institutions has been the provision of subsidized student loans. In 1976, the Brazilian government implemented the first student loan program to provide financial support for students to pay fees and maintenance costs in private and public institutions. Established by the Ministry of Education, executed by the Caixa Economica Federal (CFE, or Federal Savings Bank), the Pro,eama de Crddito Educativo (PCE, or Education Credit Program) has since provided loans to 712,688 students. Although presently PCE funds are available for the payment of tuition by (undergraduate only) students attending private institutions, during the first phase (1976-83) of PCE funds were also directed to stipends, so students attending private as well as public institutions benefitted from the credit line. The program has experienced severe financial problems since early 1980s and it has been reformulated twice. 3.35 Eligibility was prioritized according to a household income formula adjusted for the number of dependents. Loan money was intended to meet tuition costs at private institutions, and the accommodation, food and book expenses for the most needy students in both public and private institutions. Because in 1976 resources were plentiful, loans were granted to almost everyone who applied. Once students were awarded a loan in a given year, they were automatically eligible for loan renewal for a period equivalent to the average length of course-work, with a one year automatic extension. if necesarvy The rensavmpnt t0rme gnItia - fv ninm1nud0l fffoA,rg A%-- the annual inflation rate averaged almost 100% during the period 1976-83). Students were granted a 28 1 year grace period and repayment was to be no longer than the period of the loan itself. Although students signed a loan contract, they did not have to provide a guarantor. 3.36 The student loan program was administered by the Caixa Economica Federal, and losses were to be paid to the Caixa by the Ministry of Education. The Caixa. a banking structure wvith branches throughout the country, was regarded as the most appropriate body to administer the program since it had extensive experience with private lending. Furthermore, it was believed that the Caixa's control over credit ratings would be a significant deterrent against default. 3.37 By 1983, the program was essentially bankrupt. As inflation exceeded 100%, the nominal interest rate of 15% on the loans implied that the loans had a subsidy of around 90%. This was so high that students found themselves obliged to pay only nominal amounts per month. The amounts were so low that many did not bother paying back their loans, nor did the Caixa pursue them since the costs of recovery outweighed the value of outstanding debt. When administrative costs and default losses are included in the financial assessment of the program, it is clear that the loans were more expensive than outright grants. 3.38 Contributing to the rapid depletion of resources was a poor means for targeting student credit. Despite intentions to limit availability to the most needy students, the means test did not impose a firm limit on eligibility, and the program issued loans to most students who asked for them. The means test that was used stipulated three degrees of priority for lending, but did not put any limit or ceiling on who could get money. The PCE simply allocated all funds availabhn m the first few years to students, regardless of need. The recipients included students in the least needy group. 3.39 Despite the collapse in 1983, a strong demand remained for educational credit. The PCE's effectiveness as a means to bring relatively poorer students into the education system, particularly the private sector, led the government to reformulate it. The first task was to secure a financing source for the program. The government dedicated a fixed percentage of annual income from the federal lottery to the new loan program. In an attempt to make the program self financing, both the eligibility criteria and the terms of repayment were altered. Support was restricted to needy students in the private sector, for the purpose of paying fees only, and credit was limited to the maximum official length of course-work. Between 1987 and 1989, PCE granted approximately 40,000 loans to new students each semester. Table 8 summarizes the experience of those three years. 3.40 During the second phase of PCE, the terms of repayment were tightened somewhat, but a large subsidy remained. The interest rate assessed was to vary each month at 80% of the ORTN (which shadowed, but did not exactly match, inflation, which averaged well over 100% per year, reaching over 300% in 1988), plus an additional 6% of nominal fixed charge. Considering the high inflation rate of the period and the (modified) rep nt terms, the interest subsidy on the program amounted to over 55%. This subsidy level was a sut,, antial reduction over the previous one, but not enough to make the program financially viable. There were two principal factors accounting for this subsidy level: (i) interest rates remained well below the inflation rate, and even further below the cost of capital; and (ii) the period o. repayment was extended to twice the lending period, and thus the subsidy was extended over a longer period of time. 3.41 A few other reforms were introduced mn the second phase of PCE, including the requirement for a loan guarantor and the paynivnt of an up-front insurance fee. Both these changes were believed to be important steps to minimize losses through default and evasion. Nevertheless, the program continued being financially unsustainable because of the high interest subsidy. Hyperinflation exacerbated program losses and the insufficiency of dedicated lottery income again brought the program to the veree of bankruntev in 1QRQ when the npreram, %*' 2g0 - rcf3r - 29 - 3.42 Two important changes Student Loans Per Year Under PCE Phase I have been made to reduce the (1987-1989)* interest subsidy. First, the interest charge will be 6% above the BTN (a new New Total No, Total Loan AverAge Loan indicator of inflation). Second, Loans Loans_ AmcS Amourg the period of repayment has 1987 41,679 139,456 $48,845,042 S350 been cut in half, to equal the 1988 40,974 188,592 $111,125,472 S590 period of borrowing. With a 1989 39,345 224,722 $175,540,228 $782 time lag between price inflation and BTN adjustments, Total 121,998 552,770 $33S,S10,742 S600 the real interest rate is probably between A% and 2%. * Dollar figures are calculated using the average exchange rate in the One can compare this interest given years. rate charge with that on the US guaranteed loan program which is approximately 3 %.'Y Table 9 summarizes the experience of the three Table 8 programs. 3.43 The current PCE structure has an estimated interest subsidy 8% and therefore will require continued infusion of public funds. In addition, it. remains to be seen how effective the Caixa Economica will be at collecting the loans from graduates. The requirement of a loan guarantor, the control that CFE has over other access to credit, and CFE's experience in education lending suggests that default problems may be containable. The essential problem now confronting the PCE is that it has no money to make initial loans. The current allocations from lottery income are insufficient, either for start up loans, or to repay the losses that will result from the remaining subsidy, default and administrative costs. 3.44 The central problem of the first two stages of PCE were huge subsidies because the interest charges on debt was far below the rate of inflation. The interest subsidies on the loans have been 89.7%, 55.4% and 8.0% of the three loan programs respectively. Thus, the most serious problem seems to have been corrected. In order for the program to continue as p lanned, however, it will he necessary for the government to provide access to initial capital funds and a commitment to prw ide funds for about 25% of the annual lending. 3.45 The PCE has facilitated access to higher education for many students from lower income backgrounds. However, two important factors have reduced access to loans among the neediest students: (i) the lack of funds available to sustain the losses of the program has meant that there has been less overall credit available, and (ii) poorly conceived target restrictions, particularly the ahsen,;e of a firm limit on who is eligible for a loan has meant that the program has not been as effective In reaching the neediest studt,nts as it could have been. During the first few years of PCE, many loians were made to wealthier students in the public sector to support their living expenses. The eligihlht%t restriction on loans to students in private institutions was certainly a targeting improvement, but at is not yet clear that the targeting is as effective as it could be. 3.46 A third factor that will become increasingly important is existence of adequate mechani-sms tu minimize losses due to default. The Caixa's requirement of a guarantor and its ability to bar a..e%% n. 12 The US Guaranteed Student Loans Program has a larger overail subsidy of approximately 18%, hower t. because it does not assess interest during the period of study or the grace period and repayment is over 10 % eArt 30 other credit are powerful tools that will certainly minimize the problem. There are still several important details, however, that need to be addressed. Requiring a guarantor may imply thai those who need loans the most will be unable to receive them. The repayment policy may force too many graduates into default unnecessarily, as a result of the rigid policy ct categorizing anyone 60 days late in payment as in default. Studies have consistently shown that the inability to pay due to temporary shortfalls in income has been one of the major causes of high default rates in the US loan programs. It has also been shown in other countries, particularly in Sweden, that allowing any graduate whose income falls below a threshold to defer payments (while still accruing interest charges) is an effective way to minimize default. 3.47 A fourth issue is the quality of education that public credit is supporting. While some private institutions are of high standards, a wide range of institutions are being subsidized through the loan scheme. The PCE has enabled some of Brazil's worst institutions to have access to public subsidies that were necessary to their survival. Other institutions have used the loan funds to increase their h-1den profits rather than improving the quality of education. Ensuring that public subsidies are used by institutions to promote quality improvements should be a major task of reforms, as discussed later. Another issue would be the extent to which the public is adequately informed of PCE. Rglationships between the Federal Government and Private Higher EducatonU' 3.48 The various arms of the Government regulate private Institutions through a multipUcity of often conflicting rules, regulations, and instruments, especially those related to estabUishing new programs and to regulating tuition, which are, in theory, supposed to ensure minimum levels of quality and public accountability but which generally have had the effect of discouraging public accountability and encouraging low quality. Nunes and others have documented these problems. This section summarizes the legal background as well as actual practice. 3.49 Legal Backgroun. By law the CFE establishes the conditions and limits for the existence of Brazilian institutions of higher education, both public and private. The CFE authorizes programs of study and sets first year enrollments in all institutions of higher education. Two years after such authorization, the CFE "accredits" them, that is, gives them a more or less permanent right to exist. Legally the CFE provides greater autonomy to universities than for isolated faculties, although in practice the difference is minimal in authorization of new programs and increasing enrollments. The CFE authorizes the establishment of universities which must fulfill more rigorous criteria than isolated faculties. The CFE authorizes the ir.nimum curriculum necessary for all 'recognized" courses, which normally accounts for 95% of all course work. The CFE also has the obligation to review the quality of higher education institutions and to require their disbanding if necessary, and is supposed to review all institutions after ten years. The Government has the right to regulate tuition at all private institutions and has exercised this right through the CFE, MEC, the Ministry of Finance and other Government agencies. By law all private higher education institutions must be non-profit, with ultimate authority in each institutions held by a "Maintaining Authority" or "Council". In fact, many higher education institutions are profit making in all but name and many are run by individual "entrepreneurs" with a council under the direct control of the entrepreneur. The constitution of 1988 permits direct public support of "community, philanthropic, and confessional" institutions. This possibility is provided for in all the constitutions since 1946. Tax laws exempt private institutions from paying certain social costs of employment and also permit tax deductions for tuition payments. 3.50 Actual Practice. A review by Nunes et al. (1990) of legislation issued since 1960 shows that the various laws and decrees regularly require that new programs of study should be created only in 13 The following discussion is based on a consultant report prepared by Edson de Oliveira Nunes, Laura Dantas. and Violeta Maria Monteiro. 31 Interest Subsidies on the PCE Loan Promrams by Base No of Years Average Real Interest PCE of Loans Nominal Annual Interest Repayment Subsidy Phase Disbursed Interest Innlation Rate Period on Loan Phase I 5 15% 77% 35.0% 5 90% Phas3 1 4 306% 391% -17.3% 8 55% Phase IJ 4 1.0fe 4 8% All loans are assumed to be $1,000 per year, adjusted for inflation. Subsidies are calculated by using a real discount rate of 3% and include a grace period of 1 year. For Phase 11 and III loans, the ORTN and BTN adjusments used by the Government to calculate interest rates are assumed to be 96% of the inflation rate. Inflation mates used are avseges over the period of the loan programs. Since Phase 111 is being initiated, no inflation rate Is assumed. These figures include NO consideration of losses due to default, or of the administrative costs involved with the program. Source, Authiors'calculations based on Calhau (1990). Table 9 areas where there are identified "shortages" in the labor market in relationship to national or regional development needs, although these are never systematically defined. Other laws and decrees since the 1960s have at times emphasized "social needs", "non-proliferation' of professions, emphasis on quality, and emphasis on research and extension. In reality, formal and systematic criteria for establishment of new institutions have never been established and the CFE's authorization of private institutions has been based on ad-hoc criteria. As one member of the CFE has stated, the CFE basically legitimizes existing courses. However the delays and uncertainties are such that private institutions hire "despachantes" whose job is to ensure that their requests for approval are reviewed favorably and in a timely manner. The CFE has also never terminated any course for cause and ho never reviewed any institutions after ten years as mandated by law. A review of the data on the percentage of requests approved by the CFE shows a wide variation in the percentage of approvals In the period 1971-1974, for example the CFE approved on average 85% of more than 1000 requets for new .ourses. During the period 1978-1981, the CFE approved only 20% of more than 1000 requests. After 1982 the CFE again returned to a policy of approving over 80% of such requests. In January 1991, the Minister of Education announced he would not approve requests for 3000 new programs of study because the Government needed to put more "rigor"into its criteria for accreditation. The CFE's own data confuses the number of "authorized" and 'recognized" courses aniJ also contradicts the data on courses prepared by MEC's statistical office (according to the CFE th.re are 6,435 programs of study, while MEC statistics report only 4,288 such programs). 3.51 The history of federal regulation of tuition in private institution is also one of lack of transparency and changing regulations which have made it difficult for these institutions to plan. Nonetheless over the long run Government control of tuition has not discouraged entrepreneurs 2: continuing to expand their offerings and from all evidence higher education in the lower quality institutions is a lucrative operation. In 1989 a report by IPEA stated that the use by the Governm:X.t of homogenous criteria for tuition increases had particularly put the confessional institutions in .A dilemma since they were unable to keep quality up. The system therefore expressly encourages 32 expansion of lower quality institutions. Historically the CFE has been responsible for establishing the "formulas" for tuition increases. An expected tuition level is specifled on the request that is submitted to CFE, based on number of openings, number of teachers, teaching load, etc. Once CFE approves this price level, changes cannot be made without CFE's approval (inflation adjustments are also subject to Government approval). 3.52 In the period 1987-89 a variety of contradictory decrees were issued under the Cruzado and Bresser plans. In the last two years responsibilities have become blurred, with the CFE, MEC, and the Ministry of Economy playing major but unclear roles. In 1990 there were twelve Government decrees on tuition. Most recently the Government has required private institutions to negotiate and reach agreement with students and staff on the amount of tuition increases. If agreement is not reached then the federal government decides on suck increases. Private institutions have had to hire lawyers solely to deal with the federal government on getting permission for such increases. 3.53 New Proposals. Recent proposals for a new "Basic Education Law" may worsen a difficult situation with regard to federal regulation of private education. The draft law proposes to establish a new National Education Council (CNE) with 28 members each of them specifically representing a segment of society, as well as a National Forum of 60 members which would provide advice on national policy. The CNE's structure would make it very difficult to act decisively, since it would need to balance politically the needs of each pressure group represented on it. The new draft law proposes that the CNE would determine "criteria for allocation of federal funds among higher education institutions"-but it does not provide a mechanism for the CNE to implement such criteria and budgetary control would still remain with MEC. 3.54 Private Sector Reactions. A lobbying organization for the private sector of higher education was formed in 1982. The Brazilian Association of Maintaining Bodies (ABM) has over 140 mantenedoras representing over 200 institutions (or about 1/3 of the private sector). It defends the principle that education is neither a pure enterprise activity nor a purely social service activity. It lobbies government for reform in public policy towards the private sector, with the following agenda: (a) less regulation: let (public and private) prices be defined by market; give institutions freedom to create/close programs, increase/decrease first year openings (in one case of a Rio-based medical school, MEC wants to decrease number of openings from 150 to 50, due to pressure from the Medical Association), to change curriculum timely; (b) public funds for research and community services; (c) profit provision. The present situation is hypocritical: the school is treated as a regular enterprise, just like a cement plant, in terms of duties; for example, it gets no special financeal credit terms fromn banks; but the picture changes when the school is to set prices, then it is a special enterprise; (d) change the public role from that of "concessionaire" of privileges to that to "gauger" of quality and "guardian" of public interest (make sure that there is "truth in advertisement"); (e) explicit, consistent and transparent public policies; and (f) expediency in processing requests: 3 years turn around time from CFE is commonplace! 33 IV. PRINCIPLES FOR HIGHER EDUCATION REFORM The Economics of Higher Education and Implications for Reform 4.1 Under the tenets of welfare economics, private competitive markets are called for unless specific types of market or other failures occur that require government intervention. In the case of higher education, this would mean that there would be no justification for public support or provision unless there were important externalities, failures in credit markets to individuals and institutions, and/or problems of inequity. Since these do exist there is at least a limited justification for public intervention in the provision of higher education. 4.2 Justification for State Intervention. 'Public goods" are commodities, such as clean air, or services, such as national defense, where the benefits to society cannot be easily divided among individuals, and in which those individuals are therefore likely to under-invest. Education is a "quasi- publice good, yielding both private benefits (in terms of increased income-earning opportunities) and social benefits. However, the mix of public and private benefits varies across educational levels and products. The social benefits or "externalities' of higher education include (a) generating innovation to improve all peoples' lives through scientific research leading to technological applications; (b) preserving, transmitting and improving upon bodies of knowledge in various disciplines needed for social and economic development; (c) training the necessary leaders for a society who can manage the economy and the society, as well as being a source of innovation for economic development and resolution of social problems; (d) nation-building, and (e) promoting a free society through criticism as well as well as thoughtful and participating citizens. Many, but not all, of these benefits are those coming at the level of graduate education and research, especially in science, which has been shown to have major long term benefits in technology development. Public funding at this level is needed to support the brightest minds in seeking new knowledge which often is not appropriable by individuals or individual enterprises. 4.3 At the same time, as a quasi-public good, individuals directly benefit to a larpa extent from higher education through increased income, so many of the returns to the investment in higher education are captured by individuats. For example, in Brazil individuals with 13 or more years of schooling earn twenty or more times the minimum wage, compared to individuals with 0-4 years of schooling, who earn the minimum wage or less. While the private returns to higher education in terms of increased income to individuals can be measured, it is very difficult to quantify the social benefits and much depends on the value society places on these them. Nonetheless, it is generally acknowledged that these exist and, therefore, some degree of government support for higher education is justified in order to ensure adequate generation of new knowledge. 4.4 Public intervention in higher education is also needed to overcome failures in human capital markets, which prevent private individual and institutional investment in higher education. Higher education is epive, and institutions often lack access to start-up funds necessary for them to open or expand. Furthermore, while individuals would normally be a source of funds for institutions through tuition payments, most individuals lack access to credit because human capital investments are regarded as inherently risky, and banks tend not to accept future earnings as collateral for a loan. Without enabling individuals to get access to credit, society will under-invest in higher education. Educational institutions could not afford expensive instructional equipment and materials, given students' limited access to the necessary capital to cover tuition costs of their education. The under- investment would be greatest in fields which are both capital and labor intensive, such as science and engineering, and in fields where the social rate of return is greater than the private rate of return, 34 such as basic research in the social and physical sciences and areas such as public health. Privately financed institutions will be biased toward less expensive fields that generate high private returns, such as law and business administration.15 4.5 A final reason why the state should intervene is to pursue goals of equity. Societies have an interest in equity not only as an ideological principle, but also from an efficiency standpoint. Education systems are most efficient when access is determined by ability to achieve, not ability to pay. Thus, public intervention serves to ensure that qualified poor individuals, who in most cases have no access to credit, and who cannot afford to forego income while attending school, have access to education. It also can ensure that poorer regions of a country, which are most likely to lack the capital resources to begin an institution, can benefit from the services provided by a local higher education institution (through attracting commercial development, improving local industry or agriculture, etc.) Adequate resources are often concentrated in a few central cities. This situation further worsens regional inequalities, as there is little attention and research addressing the needs and concerns of remote areas. Transporting individuals to central institutions also results in a "brain drain' of these regions. 4.6 Limits to state intervention. Although there is a case for public intervention in the financing of higher education, there is also a strong rationale for limiting that intervention, for requiring direct beneficiaries to pay part of the costs, and for encouraging the private provision of higher education. This can be argued in terms of increasing both the efficiency and equity of the educational system. 4.7 Important extemal efficiency issues are the government's tradeoff between investing in higher education and investing in other potential areas of public expenditure, and how efficient its investment in higher education is compared with the private sector's. As noted above there are externalities related to higher education. However, there is also a strong argument for directing Brazil's public investment on the margin into primary and secondary education rather than into higher education. In the first place the lower levels offer the greatest externalities by providing basic litcracy and numeracy to the population as a whole, a goal of all societies. In addition the lower levels generally offer higher returns in terms of increased productivity as measured by income as comparcd with the costs to society. A useful, although imperfect, means to assess the impacts of education is a rate of return analysis. Studies have consistently shown that primary education yields the highcst social rate of return, followed by secondary and then higher education. Data from 45 developing countries yields rates of 24%, 15% and 13% respectively (Psacharopoulos 1985). Finally, private sector financing of primary education would be highly inequitable because of family income constraints and the long lead time before children become economic assets. Therefore, while all education warrants public investment, at the margin, primary education is the most efficient public investment. It also appears that graduate education and research have greater public good propcrtics tban undergraduate education. 4.8 Government also has an interest in encouraging eouity in society. Complete subsidization of higher education, at the general taxpayer's expense, is inequitable t -cause those who attend higher education usually come from wealthier families (on average 44% of students in federal univcrsifics Is A common mnisconception is that quality teaching and research in scientific fields occurs in US pnvatc institutions without statoe support. These institutions are an exception to what are commonly conceived of as pri r iZne institutions. Por example student tuition makes up only about 25% of Hrvard University's operating income. Nv of the remaining funds come through government support for researcb, and through an endowment fund, hs.hih accounts for over 25% of annual income. 35 have family income equivalent to 11 minimum salaries or more, as compared to 10% in the general population) and they usually earn the highest salaries after graduation. As per capita expenditure on higher education is so much greater than at lower levels of the education system, public subsidies represent a net transfer of income from the poor to the rich.' In addition, in Brazil, free tuition in public institutions is being provided to childrfn of the middle and upper income classes, while those worldng full time, and with lower incomes, attend evening courses in fee-paying institutions of lower quality. Therefore, user charges in public institutions, when combined with mechanisms to assist those who cannot pay (loans and scholarships), will result in a more equitable distnbution of public funds. User charges shift costs onto those who receive benefits rather than spreading them over a broad tax base of people who receive fewer and more indirect benefits. 4.9 Another limitation to state intervention concerns growing budget constraints in the wake of expansion. The social demand for higher education has exploded over the last few decades because of rising numbers of secondary school graduates and the perceived private benefits to university degrees. Many governments have responded to this pressure by rapidly expanding public higher education institutions and/or by increasing funds for higher education at the expense of more pressing priorities. In Latin America, for instance, the percentage of youths aged 20 to 24 entering higher education increased from 3.0% in 1960 to 16.9% by 1987. The impact on government investment in education has been alarming. In 1960 Latin American governments spent 14% of their education budgets on higher education. The share of resources climbed to nearly 24% in 1980.'7 Meeting the continued demand for enrollment in higher education would mean that primary and secondary education as well as other pressing social needs would have fewer public resources. 4.10 An argument against state provision of higher education is that it encourages inefficiency. Throughout the world costs of public institutions are found to be higher than those of private institutions which provide similar services. Because of bureaucratic rules and regulations public institutions are less agile than private ones in responding to changing demands. Furthermore low or non-existent fees lead to longer periods of stay by students in public institutions and also to a lack of pressure for increased internal efficiency. 411 In summary, based on economic theory, Governments should seek to maximze the mobilltion of Individual and private sources of funding for higher education. This would include charging tuition to cover the teaching costs of undergraduate education In public Institutions, supporting the growth of the private sector, and trying to get the public sector to focus most on the areas within higher education which have the greatest externalities, Government support should especially go to research, with externalities In terms of encouraging economic development through technology development, and to institutions In backward or underdeveloped regions. To deal with equity Issues and capital market failures, governments should support large-scale loan schemes as wel as scholarships to the needy. Subsidizing students through scholarships and loans rather than directly subsidizing institutions would mean the Government would place choice of institutions In the hands of students, with resulting encouragement of diversity and cost eMciency of higher education systems. ' Birdsall and James 1990, Psacharopoulos 1986. ' Psacharopoulos, George. 'Higher Education in Developing Countries: Tle Scenario of the Futue', in Highlt Education, Vol 21. 1991. 36 4.12 While these arguments are persuasive, no country in the world has fully established a higher education system of this sort. This is partly a result of history, when Governments could easily afford the full costs of a small public higher education system training the nation's elite. Another reason that this model has not been fully implemented is that there are a number of complex practical problems which have to be considered, including, for example, how to effectively manage large loan schemes, as well as important and non-trivial transition costs in introducing a possibly excessive uncertainty in funding leading to changes in the missions and clientele of institutions. A final reason lies in the power of associations and pressure groups to resist such changes, which may involve a loss of privileged positions. This often leads governments to finance private rather than public goods, and to undertake inefficient programs that yield large benefits to small groups of politically influential actors, rather than small bencfits to widely diffused groups, even though the social return to the latter may be greater. Since producer groups are likely to be better organized than consumer groups, and middle and upper income groups are probably more articulate than poorer groups, it is likely that producer and middle class groups will benefit dis-proportionately from government policies. Nonetheless, most countries in the world are finding themselves forced, mainly for financial reasons, to move in the direction of increased tuition charges in public institutions, expanded loan schemes, and the growth of private education. Other Elements in the Reform of Higher Education 4.13 Any reform program will also need to take into account how higher education institutions ought to operate in this present age of rapid technological change. These include, especially, needs for diversity and autonomy, accountability, and quality. 4.14 Diversity. Prof. Burton Clark18 has observed that there are now more than 8000 specialties in the sciences alone. The explosion of knowledge at the top end is accompanied by a wide variety of new jobs and tasks at middle and lower levels. With each passing decade there are more and more varied students, more and more varied labor force connections, and more and more varied bodies of knowledge that flow from research. This trend will increase in the future. In response, throughout the world there has been a multiplication of types of institutional instruments: polytechnics, regional colleges, short cycle units, specialized research institutes, open universities, and several types of universities, both research and non-research based. The unitary model of higher education--with one budget system, one personnel system, and one curricular system--is already severely dysfunctional in the few countries, such as Italy, where it continues, or in modified versions, such as French higher education, which consists of only three major systems. Currently the two most diverse highcr education systems are those of Japan, with a huge and varied private sector and types of public institutions, and the US system, with 1800 private institutions and 150 public universities and a widc variety of clientele and institutional missions. Most large developed and developing countries will need to incorporate some variation of these open, diversified systems as the model of their future higher education systems. 4.15 Autonomy. To ensure diversity, it is fundamental that the higher education enterprise be, to the greatest extent possible, decentralized and autonomous. Some areas of government require natural unitary control, or even top-down bureaucracy, such as the collection of taxes or the control of a national military establishment. But other tasks do not tolerate unified command. Higher education cannot operate well under central control because of a need for a diversity of teaching, disciplinary, and analytical approaches. This is unlike individual primary and secondary schools u hi h 18 The following discussion is based in part on an unpublished paper by Prof. Clark. 37 fundamentally provide basic common skills, a common culture, and common socialization needed for national building and national identity. The cries for autonomy therefore do not simply reflect yet another interest group seeking to be given .esources and left alone. Rather there is a basic need for localization of decision-making which is intrinsic in higher education. The most dependable way to guarantee this localization is to have a thorough decentralization of control, to multiple public authorities at regional as well as national levels, to multiple private authorities, and to the greatest extend possible to individual higher education institutions which should be free to allocate funds as they see fit.' They should be given the maximum freedom to open and close courses and to experiment with curriculum and programs. 4.16 Accountability. The idea that higher education need not justify its activities to government or to society at large came at a time when the research institutions of Europe and America took a very limited portion of the public budget and therefore society could afford to pursue their results only from afar. With the increasing social demand for higher education, such a utopian position is no longer possible. Therefore, any institution directly receiving public funds has an obligation to show how such funds are used. The issue here is the clear enunciation of academic, research and other goals and an accoumting of how and under what conditions and costs such goals have been met. The need for a strong dose of accountability is particularly important because of the highly vocal nature of the academic community, many of whose students are the children of leading figures in political and economic life, who may wish to retain their privileged position and may resist being subject to outside scrutiny. Accountability requires a strong effort by Governr ent, wv ich must strengthen its analytical and decision-making skills to review and evaluate the perfe nance of the institutions it is funding, and to establish public policy priorities. 4.17 Ouality mprovement. As noted in the introduction, the definition of quality in higher education should be ttat of "value added": the extent to which, within a specific time frame, an educational institution builds on the raw material it receives-the student-and increase his or her knowledge and skills. This includes ensuring that modem rather than outdated theories of science, economics, business administration, etc. are taught. As more and more economic development becomes knowledge-based, it becomes fundamental to ensure that the population increases its stock of such knowledge. A high-quality institution therefore could be one which accepts very poor students, or those with inadequate preparation, and educates and develops their capacities. Many US community colleges as well as the UK Polytechnics should be considered of high quality, since they provide good remedial and/or technical education to young men and women with often inadequate preparation. In Brazil, an institution like the University of Ijui could be considered among the highest-quality institutions in the country, since its clientele is poor rural students from the interior of Rio Grande do Sul, in whom it reportedly instills intellectual discipline as well as a strong dose of social consciousness. The lowest-quality institutions would be (a) the private "diploma mills" attended by students for the sole purpose of achieving a 'credential" which would entitle them to a salary increment in the public service but which provide little or no increment to their stock of knowledge, and (b) those ,ublic institutions, which, while they may have better students, have non- performing faculty and reulre litle from the students. The "pacto corrupto" in both public and private institutions is the fundamental enemy of higher quality. Credentialism at best has no effect on economic development, probably slows it down because of wasted energy and can increase inequity by rewarding those with the most financial resources and "connections." Quality as defined as "value added" can be measured by increased learning and achievement or increased income. From this point of view encouraging quality would be important to help ensure that students earn enough in the future to pay back loans. 19/ With flea ntinn.n r 1nteA tn arnuntnlnitv d;stus below. 38 V. PROPOSED REFORMS IN HIGHER EDUCATION 5.1 Brazil should set out as its long term goal the establishment of an equitable and efficient higher education system in accordance with the principles listed in Chapter IV. Specifically, the goals should be to: e charge students for a large portion of teaching costs in public institutions; o expand loan and scholarship schemes to assist students in all higher education institutions; o ensure cost-effectiveness, autonomy and accountability in all institutions receiving public funding; and 0 provide a large portion of financial support to higher education through subsidized loans to students, and through support for high public good areas such as basic research and related graduate education, with both public and private institutions competing on an equal basis for research funds. Brazil is very far from this model. No tuition is charged in public institutions, the loan scheme does not achieve the goals of increased equity and efficiency, and the current system of governance and management of public institutions and of oversight of private institutions actively discourages cost effectiveness and accountability. The only area where, until recently, Brazil was achieving relatively good results was in its support of science and engineering at the graduate level, where the current competitive system with multiple sources of funds has worked relatively well. 5.2 Because of constitutional and political impediments, as well as the high costs of traumatic changes in the nature and mission of institutions, it is expected that under any circumstances Brazil will move gradually on the reform program described above. The proposals in this section are, therefore, mainly designed to identify medium-term programs which would be politically and technically feasible and which would move Brazil significantly in the right direction. Ihese proposals cover the two major areas of financing of the higher education system and governance, management and oversight of the system. 5.3 The Government and the informed p'iblic are already discussing major reforms in higher education. Recent government reports have focussed on the need for providing more autonomy and accountability to higher education, and newspaper articles have reported a debate over the costs of public higher education. Senators and other public figures are seriously debating the possibility of charging tuition in public institutions. Furthermore, in 1991-92 MEC officials proposed many of the reforms discussed in tis paper. In mid-1993 a parliamentary commission completed a study of higher education. The time is therefore ripe for action. However, to succeed, Govermnent will need to mobilize its most progressive and farsighted citizens to get behind a program which will in the short run adversely affect the fortunes of pressure groups and lobbies. Government should keep in mind that its efforts will be repaid to society through helping to ensure that higher education meets national goals of renewed economic development and social progress. Higher Education Financing 5.4 Reforms in the finance of both public and private institutions can help ensure that Brazilian higher education serves the principles of diversity, autonomy, accountability, cost-efficiency, quality 39 improvement, and equity. Reforms in the financing of higher education are of the utmost importance because they are most likely to change the behavior of decision-makers at the institutional level. Experience in other countries, especially in Europe and America, is of particular value. Financing reforms fall under three main categories-establishing cost recovery in public institutions, strengthening the student loan scheme, and budgeting on the basis of formulas to improve efficiency. 5.5 The c.urrent practice of providing incremental budgets to in-titutions based on last year's funding, while not paying attention to costs or outputs, should be ended. This method of financing of federal higher education has resulted in among the lowest student teacher ratios in the world and in costs which are too high by a factor of one third to one half. The government has and continues to give institutions funds to buy more capacity-in terms of teaching, building, equipment, staff-but it does not tell the institutions that this capacity has to be used for any specific purpose-such as teaching more students. However, any new approach to financing of federal higher education will require the provision of real autonomy to universities, the dismantling of the 'Law of Isonomy', and the provision of CFE or a new agency with budgetary power and analytical capacity. 5.6 Cost Reove in Public Higher Education. To Increase the equity of the education system and to provide resources for future expansion and quality improvements, cost recovery should be Implemented at Brazilan public tertiary Institutions, through tuition payments and other fees (for laboratories, examinations, etc.). The long term objective of cost recovery in public higher education Is to recover a large portion of the teaching costs from students. Cost recovery is only one of several means, albeit an important one, at the disposal of government to increase efficiency and equity of the system. However, recovery of costs in federal institutions will depend on a constitutional amendment, and even afterward will need to be phased in. Therefore, in the short term the Government's main financing tools for reform will more than likely revolve about changing its budgeting systems and improving the student loan program. 5.7 The analysis in Chapter m showed that the current teaching and research costs of federal higher education of about US$6000 could be reduced to US$4000 without any significant lost of quality. Of this figure US$4000, a reasonable estimate of teaching costs would be US$2500, on the assumption that the teaching load of 8-12 periods per week permits staff to spend about 40% of their timne doing research. Iherefore the goal of cost recovery in public higher education should be to charge this amount to students, along with a loan and scholarship program (see below). 5.8 In practice this goal can be achieved only gradually, especially given the practical problems of vastly expanding the loan scheme. A modest and more politically feasible start might be to set initial tuition at a level similar to the current charges in private secondary schools. Since an estimated 60% of matriculating students come from private secondary schools which charge fees ranging between $400 and $1200 per year, there would be no reason that students should not be expected to maintain a contribution of US$700 at the University level. It would be difficult conceptually for students to argue against and would require only a relatively modest change in the student loan scheme. Tuition of US$700 would be equivalent to 28% of estimated teaching costs and 18% of total costs (after total costs are reduced by a third). The table below summarizes the extent of cost recovery of total costs in public institutions in a number of countries with which Brazil is often compared. In Spain, for example, institutions charge students 20% of the total costs of their education. The figure is currently 26% in Chile, an average of 15% in the US and 26% in Indonesia. Data on the percentage of costs solely related to teaching are not available. However, one of the few studies on this subject argues that public universities in the U. S. recover a large proportion of teaching costs from the first two years of undergraduate education (James, 1979). 40 5.9 Eventually, government funding to each public institution could be reduced to a level which would assume cost recovery of full teaching costs, and public institutions could be permitted to decide how to finance these teaching costs. Each institution could then set its own tuition levels and could diversify its funding base through seeking additional funds from industry, local government, state government, and from private citizens through donations. 5.10 The extensive subsidies provided for student services use up a considerable portion of discretionary budgets. For example, the subsidy for student meals at the public universities averages $31 per student per year (60% of the cost of meals), or an estimated US$10 million for all of Brazil. This subsidy should be eliminated by charging full cost for meals or by charging an additional $31 per year to all students. Federal institutions should be encouraged to charge full cost for all such services to be provided to students. Reforming the Student Loan Program Cost Recovery in Public Higher Education as a Percentage of Unit Operating Expenditure, 5.11 To the extent that the Selected Countries Govermnent increases cost recovery in public higher education, the student loan 9C Cost scheme (ECP) should become oP. of the S211ntIL. Recovery Year central institutional elements ef higher B3zil 0 1989 education policy reform. Over the long Argentina 0 1987 term, an efficiently run student loan Nigeria 0 1989 scheme will significantly lower government Colombia 5 1988 expenditures while creating incentives for India 5 mid 1980, institutions to respond more rapidly and Thailand S maid 19809 Japan 9 1987 cost-effectively to both student and labor Philippines IS mid 1980s market demands. Overall loans should be USA 15 1984 used to promote access of students- to Israel 17 1987 higher education and choice between a Spaia 20 1988 variety of public and private institutions. chile 26 1987 Because of its importance, the budget of Ch__________2____1987___ the ECP should be largely expanded and Source: Mingat and Tan, 1989; OECD 1989; Brunner 1990; under the direct control of MEC. The World Bank Sector Study on Nigeria, 1989; World Government should ensure that the credit Bank Staff Appraiusal Report on Indonesia 1988; World program is financially viable by Bank Staff Appraisal Report on Colombia 1988. guaranteeing access to start up capital, recurrent funding to offset losses from the remaining subsidy and default, and collection procedures that are strict enough to minimize default losses. The Table 10 average size of student loans would depend on the level of tuition charged in both public and private institutions. As federal institutions move towards full cost recovery, the loan amount available would increase and would eventually operate as a kind of voucher which the student could use to study at the institution of his/her choice. 5.12 Selection of students. Only needy and deserving students should have access to scholarships and subsidized loans. Need should be judged both in terms of family resources and academic ability. The neediest students with high potential should be given grants, as opportunity costs will otherwise 41 prevent them from attaining access. The definition of need, in terms of the number of recipients and in the amount of credit available, will depend on the level of public and private tuition. If full cost recovery for instruction were implemented in the public universities, the loan amounts would probably average US$ 2000 per year. The initial loan program provided subsidies for the living expenses of wealthier students in the public sector, with the result that funds were quickly depleted. The Government should explicitly limit subsidized loans to needy students as defined by clear and objective criteria. Fully self-financing loans (with no explicit subsidy) could also be made available to those with no demonstrated financial need. 5.13 Since students can sometimes hide their true ability to pay, the criteria for determining financial need should not be income alone. To avoid problems of erroneous income reporting, family assets (houses, cars) should be reported, as well as parental occupations and education levels. In addition, attendance in a private secondary schools (without scholarship) can be considered as strong evidence that a student has the ability to pay a significant portion of higher education tuition. To ensure accurate reporting, the ECP could spot check students, and deny access to any student found to be misrepresenting his/her need. 5.14 The loan amounts should encourage choice for students. As an example students could d;cCJe to apply it either to tuition costs at more expensive institutions or use it for living support at leis expensive institutions. This will encourage institutional differentiation. 5.15 Repayment terms. A recent World Bank review of loan schemes world wide has shown thAt to date only a very few schemes recover more than 60% of their real value, after taking into accuunh interest subsidies, administration costs, and defaults (Albrecht, 1991). Tables V-1 and V-2 summu ze the experience of the loan schemes reviewed, including experience with loan subsidies, default rst. and administration costs. However, a wide variety of innovative means of implementing student I ans schemes are under discussion, and future loan schemes are expected to be more efficient. There .^ therefore no reason why a student loan program in Brazil could not recover 60 to 75% of the lo-n value. This can be achieved through reducing or eliminating the interest subsidy while at the ssme time introducing a variety of innovative repayment schemes. 5.16 Specifically, interest rates should be positive in real terms. Repayments should be on a ; real scale, to account for varying patterns of graduate wages. Graduate earnings typically start l. 1 and rise quickly within a few years. The real value payments should reflect this pattern so th3t . payments do not represent an excessive portion of earnings. The first payments should therefore relatively smaller than later payments. One way of achieving this would be to move towards a l. Jn program in which graduates repay their loans as a fixed percentage of annual income (an incOme- contingent loan), rather than having a prearranged payment schedule.' Income contingent loins .&: e preferable because they minimize risk to both the borrower (because no payment is an excessl%c portion of income) and the creditor since default will be reduced. As the income tax system strengthens, collection would be more efficient, lower cost and lower evasion, by having student% submit their payments on top of their annual tax payments. Such income contingent loan progrx- * have been successfully implemented in Sweden, Australia and Ghana. 20 An altemnative would be to establish a graduate income tax, in which all graduates of public institution . pay a smaUl percentage of their income as a tax to the government. This would be similar to repaying an in, r contingent loan. This approach has been discussed in several countries but has not yet been implemented. 42 5.17 As an alternative, income deferral clauses, which stipulate that a borrower can defer payment if his/her income falls below a certain threshold (such as average earnings) are a useful means to minimize default among graduates than cannot pay. Similarly, the program could stipulate that no graduate would have to make a payment in excess of 15% of his monthly income. It should be noted that the US program, which has no deferment or income contingent clause, has forced many students into default due to temporary income falls. Default in the US is approximately 14% compared to 1% in Sweden. 5.18 Minimizing default from evaders who can afford to pay means stepping up procedures to bar access to further credit, taking legal action against defaulters, and requiring payroll deductions by the employers of students who are late in th&tr payments. Tle Caixa has the means to enforce repayment, in part because it provides largo amounts of housing and other lines of credit, as well as because of the practice of requiring a guarantor for the loans. A further step which can prove useful is the use of private agencies to locate and collect payment in the event of default. Such measures have proven effective in the loan programs in both Honduras and Colombia, where default rates have dropped dramatically in recent years. 5.19 Meeting Needs in Priority Areas. The Government should also seriously consider using the loan program to meet policy objectives in priority areas. The most fundamental and urgent area is that of primary and secondary school teaching, especially in the rural Northeast, where low pay and inadequate working conditions have lowered the quality of those entering the teaching profession. Pending such improvements, portions of the loan could be forgiven for all graduates who teach a minimum number of years in the public primary or secondary school system in key regions of the country. This approach would serve to attract motivated students with inadequate financial means into the profession and would illustrate to the general public the importance the federal government places on improved basic education. 5.20 Funding. There are several alternatives that can be used to fund the loan program. In the first years of reform, the money saved from reduced unit costs in federal universities, and the tuition revenues transferred to MEC could be used to finance the increased size of the loan fund. If these policy changes are implemented, the loan program could fully function without any injection of funds from outside MEC's current budget. As the government will now be receiving a significant amount of this money back according to the repayment terms, there would be an overall net savings. As an alternative, the program could borrow money to fund itself in the first fevw years, while the government provides the recurrent funding to offset losses. 5.21 The graph below shows what it would cost each year to run a loan program that would provide an average loan amount of $2000 to 25% of the entire higher education system. It assumes a repayment scheme with a 1% real interest rate, a real cost of funds of 4%, a default rate of 10% and administrative costs of 3% of outstanding debt per year. During the first few years, effective outlays will be high (although the government could borrow against these outlays). After 10 years, about 75% of outlays would be met through repayments to the program. A variety of other alternative projections could be prepared based on differing assumptions of interest rates, defaults, and administration costs. The World Bank has available a simple computer program to test sensitivity to changes in the key parameters in a proposed future loan scheme. 5.22 The Government could also use loan funds in a way that recognizes their potential to create incentives for quality education. Funds should no longer be used to subsidize poor practices, or to encourage the expansion of the worst institutions. Rather, funds could be channeled in a way that effectively promotes quality in private institutions throu, h linking eligibility for loan money to quality 43 indicators. One simple indicator of quality would be the percentage of staff wit Projected Expenditures and Revenues for master's or Ph.D's. Proposed Brazilian Loan Program manter's sp quality Another simple quaity US S thousand indicator which could 1600 be used is the number 1400 .................................. ...........................I ...... of library books, which 1200 .......................would be a proxy for 1200.1 ...............I.............. the availability of non- 1000 .: ...... >_f .salary inputs at the 800institution. Institutions 80 .................... ....................................I...I........ inttto.Isiuin / ~~~~~~~~~~~would be required to 60c0;. certify,under penalty of 400 . ........... law, that staff with 200 ...............................................advanced degrees were 200 /.e .. - .... teaching or that library o books were available 0 s l0 lS 20 25 and circulating to students. Subsequently Expenditure -t-I Revenues more subtle measures of quality could be Assumes siana toenn for 252 of utudonle dfault 102, Is annual costs and 35 utilized, witfh a variety afoual @xpanstofl. of financial inducements for further increased quality. Public universities or Chart 4 other independent professional bodies could carry out general assessments of the staff and curricula. Private institutions eventually could be asked to pay for this assessment. Initially the Government should set the rules for eligibility in such a way that a significant portion (say, 25%) of private institutions would not have high enough standards of quality to be eligible for the program. This would be a powerful incentive for quality and would especially benefit the higher quality religiously affiliated and community institutions. Improving quality would also result over the long run in higher income of grartuates who would be more likely to repay loans. Qgovernance. Mang ent and Oversight of Higher Education 5.23 Budgeting for Public Higher Education. As noted in Chapter IV, in principle Govenunent funding should mainly be directed to student support and competitive funding for research. Eventually this could result in breaking down the differences between public and private institutions. However, public institutions are expected to continue to exist in more or less their current configuration for some time. Therefore it is fundamental for Government to use its budgeting power to ensure that federal institutions increasingly openly compete for funds on the basis of cost-effectiveness and public utility. 5.24 A funding formula should be devised to encourage such cost effective behavior on the part of public institutions. The primary signal to send is that the government will pay institutions on the basis 44 of what it considers it should cost to perform its functions after subtracting the amounts to be paid by students through tuition and fees, rather than what the institution would like it to cost. Specifically the government would calculate the expected costs for teaching (including overhead such as administration, libraries, maintenance) on the basis of recommended student teacher ratios (differentiated by field of study), student staff ratios, necessary classroom space, etc. On the basis of these data, the government would calculate a funding formula, in which the expected costs per student (or graduate) are multiplied by the number of students an institution has. And thus the Government would determine a budget on the basis of cost rather than on the basis of last year's budget. This funding formula should be based on a per student unit costs in public institutions equivalent to two thirds of the present cost, since as noted in Chapter III, the unit costs in public institutions are excessive by at least this amount. In addition, the Government would continue to support specific research programs on the basis of open competition and peer review of the quality of proposals. 5.25 Annex 1 provides a summary of issues and options in funding formulas, with summaries of the procedures in eight countries. Briefly, there are three broad categories of funding criteria: (i) incremental/ political budgeting-e.g. the current situation in Brazil and most of the developing world; (ii) formula based input funding-e.g. enrollment formulas with costing norms, weighted by course, type of institution, institu.lonal profiles, etc.; and (iii) performance/output based funding which addresses issues of the efficiency and quality of education. No formula is fully satisfactory. Simple formula based input funding--payment on the basis of unit cost per student in a particular discipline- does not promote further improvements in efficiency or quality, nor does it account for the possibility of economies of scale. Unless there are other restrictions, institutions may be tempted to keep taking on students and receive more money than it costs to have them, especially since the marginal costs for Institutions to take on more students is usually less than the average costs of students already there. Therefore a formula should pay a lower amount for additional students. Recent experiments have sought to incorporate efficiency and performance incentives into formulas. In Holland and Finland, for examplf .qunding is based on number of graduates rather than enrollments. The aim of this approach is to improve student flow rates. However, unless there are other constraints, this could well result in reducing standards. 5.26 In view of Brazil's lack of experience with funding formulas, as well as a lack of institutional data and minimum analytical expertise, funding formulas in Brazil at the start could be simple and easy to measure and enforce and based on desired per student costs, with some weight given to number of graduates. Even the simplest formulas will require determining current and target unit costs by major field of study-science, agriculture, engineering, medicine, social science, law, humanities, and education-and then assigning funding to each institution based on weighted enrollments in each fields of study. The hard sciences and medicine will cost significantly more than law aad humanities, and therefore institutions enrolling more engineers will on average receive more funds than those enrolling larger number of law students. In the absence of data on unit costs by course in Brazil, it might be possible to rely on international norms. Faced with funding by formulas. institutions may enroll "ghost" students who do not attend classes and never graduate. In fact, it has been reported in Ceara that many students are double enrolled in two or more institutions but rarely appear for classes. Possible ways of dealing with this problem include: asking institutions to provide evidence that non-attending students are taken off the enrollment books and not permitting students to enroll in more than one institution at a time. 5.27 Governance of Federal Higher Education. The main objective of proposed changes in governance of federal institutions would be to enable them to act with autonomy and accountability 45 within the context of a system in which Government continues to provide a large portion of their operating costs.2" Each institution would be encouraged to set its own rules and regulations for its internal operation. Subject to certain rules and guidelines (discussed below), each institution would hire, pay and utilize staff according its own criteria, seek funds from a wide range of sources, including industry, local government, and its own students and alumni, and establish new programs of study. Federal higher education institutions would be funded on the basis of formulas as described above. The Government would guarantee, at the beginning of the year, their full funding for the rest of the year and in real terms. 5.28 To implement a decentralized public system, the 'Law of Isonomy" should be repealed, since it is incompatible with institutional autonomy, especially in terms of institutional responsibility for budgeting and for hiring and firing of staff. After such repeal, higher education staff would no longer be part of the civil service. Rather, they would be employees of the specific institutions they are working for and would have no guarantees beyond those negotiated directly with the institutions and which would be available for all private sector employees. Each institution would be free to set its own hiring, pay and promotion criteria, in accordance with certain general rules to be set by Government. 5.29 To increase accountability and cost-effectiveness, MEC should be strengthened and professionalized. The highly politk'al orientation of MEC would need to be reduced and there would need to be greater continuity of leadership (in recent years there have been new ministers of education about every twelve months). An alternative would be to replace the current Federal Council of Education (CFE) or to restructure the CFE to be the main agency managing the higher education system and acting as an intermediate agency or buffer between Government and higher education. By standing between Government and the institutions, the new agency would be able to define the balance between autonomy and accountability and counterbalance the demands of higher education institutions for more funding with fewer strings attached and the desire on the part of Government for cost effectiveness and adherence to national goals. In principle, higher education could be protected from political pressures and funding would be put on a sounder and more technical basis. However Brazil's experience with autonomous agencies has often been that the representatives become spokesmen for various pressure groups, lobbies, and political groups. For the new agency to work, its members would have to be appointed on the basis of nominations by professional and other groups, both public and private. They would need to be distinguished citizens drawn from a broad spectrum of society representing themselves and the public good rather than specific interest groups. This new agency would need to have an analytical capacity (a small but high level staff) to determine and revise as necessary the rules and formulas for annually financin,g higher education, as well as to examine long term policy issues. 5.30 In addition to using funding formulas as a means of encouraging cost-effectiveness, MEC would set additional guidelines to encourage quality and efficiency which would be reviewed during the annual budget exercise. In principle only a few crucial policies should be made mandatory=. 21 Governance would take on a different dimension if and when Covernment would recover full teaching costs from undergraduate students in public institutions. 22 States in the USA vary greatly in the nature of their oversight of public higher education, with some providing institutions with great discretion in the use of their funds, and others, notably New York State, keeping a tight control over line items as well as many institutional policies and practices. 46 MEC would need to monitor the actions of federal institutions and revise the guidelines, especially those made mandatory, to help insure best practice. Possible guidelines might include the following: (a) Guidelines for hiring and promotion of faculty. The major change from current practice might be to encourage all associate and full professors to have masters or PhD's. Another change might be to revise the requirements for "exclusive dedication" to encourage part-time teaching in areas such as law and medicine, to identify staff who are not doing research and to require them to increase their teaching hours significantly or to take equivalent pay cuts, and to establish simple rules and regulations for outside consulting. (b) Efficiency guidelines. The basis for the Government's funding formula would be student teacher ratios for courses of study. This information would be made available to all institutions. Individual institutions would not be obligated to follow these ratios if they could make other savings or find other sources of income. (c) New program guidelines Institutions would have the maximum flexibility to redirect the focus of existing programs of study and to increase or reduce program enrollments. However, the Government would need to approve new initiatives with major financial implications, such as establishment of a new faculty of engineering or medicine. Given the present situation of financial stringency coupled with under-utilization of facilities and staff, it would be prudent to defer for some time (at least three years) authorizing such major new programs. (d) Guidelines for student performance, Institutions could be encouraged to develop mechanisms to ensure that no student enrolls in two institutions at the same time and that all students attend classes and take final examinations. Students not attending classes or not completing their course-work within a specified period of time would not be counted as students. A portion of the funding formula could depend on timely graduation of students. (e) Rporting. Each institution receiving public funds would prepare an annual report, in accordance with a specific format, on its sources and uses of such funds. 5.31 To enhance cost-effectiveness the Government would declare a moratorium on establishment of any new publicly funded universities, at the very least until the existing facilities are adequately utilized (e.g. by a doubling of enrollment). During this period any new capital funding for federal Institutions would go solely for equipment and library books and periodicals. MEC would coniniss%i n studies and support training on technical issues in physical facilities utilization, including how to utilize computerized scheduling programs as well as varying the size of classes (eg., combining large lectures with small seminar groups or classes) as a means of getting maximum facilities utilization. 5.32 Management of Federal Institutions. Students have too short a time frame and administrat i e and technical personnel are fundamentally not involved in the business of the university, which is tie creation and transmission of knowledge, to have decisive roles in the selection of rectors. Therefore. to complement the changes in incentives for increased cost-effectiveness to be brought about by reforms in governance and financing, the current method of electing rectors of federal institutions (after which a short list is presented to the Minister of Education) should be replaced by one whi.h gives adequate weight to the most productive members of the academic community, representatives' 47 society as a whole, and government. A viable alternative practiced in numerous countries is for the University Board, consisting of public citizens of renown, to conduct a search and recommend to the President or Minister of Education the most highly qualified candidate. The current model in Sao Paulo would also be an improvement, albeit partial. This model provides for a two step election procedure, the first, open balloting by all members of the university community, the second, balloting by full and associate professors, after which a list of three is presented to the Governor. 5.33 To further increase quality and cost-effectiveness in federal institutions, the rector's office should be strengthened by giving it budgetary decision making power, subject to review by a university board, by permitting re-appointment of the rector if his performance is satisfactory, and by encouraging the development of permanent professional staff responsible for budgetary and institutional planning. The rector would be expected to take the lead in determining the mission of his institution and in proposing a strategy to achieve the agreed upon goals. The excessive number of committees and boards taking decisions should be eliminated and one or two boards should be established. One should be the board of overseers which currently exists, and a second should. be the academic senate. Students and non-teaching staff should have representatives on both of these. 5.34 Encouragement of Private Higher Education. The Government should play a major role in encouraging a wide range of private institutions with differing clienteles and missions. Public policy should make clear that the distinction between public and private institutions is artificial, since both meet a variety of social and private goals and are supported by a variety of public and private funds. 5.35 Specifically the Government should end the backlog of 3000 applications for new private higher education institutions. It should establish and publicize simple, coherent, easily verifiable, and minimum criteria for opening new private courses, essential to protect consumers, such as the existence of formal rules of organization and a formal curriculum, staff with at least a bachelor's degree, and minimum facilities, such as classrooms and an operating library. For technical and science subjects, there would need to be minimum laboratories and operative agreements with hospitals for practical training of health professionals. The Government should no longer use labor market or "social' needs as criteria for recognizing courses, since experience has shown that it is impossible to estimate future manpower needs with any precision, and should resist the lobbying efforts of professional societies to restrict entry to their professions. Consistent with the above minimum standards, the Government should encourage the opening of new private institutions as a means of putting more competition and choice into the system. It should permit private institutions to freely increase enrollments, provided they show evidence of adhering to the basic minimum requirements. 5.36 Finally, the Government should end all control of tuition in private institutions. At the sane time the Government should move to authorize new courses, permit enrollment increases, and reduce certification requirements for many professions, so as to increase competition and encourage diversity. 5.37 MEC should control the budget of the student loan scheme, as described later in this chapter, and should use student loans as its main instrument to encourage quality improvement in the private sector. Under a revised loan scheme MEC would require all private institutions to meet certain minimum requirements of quality. The requirements would be significantly higher than those for the opening of new courses, since the Government has an overriding interest in ensuring that its own funds are spent effectively. 5.38 The confessional and community universities are of particular importance to the public because of their relatively high quality and social responsibility. However, Government support should be indirect, through the expanded student loan scheme, through funding of graduate education 48 and research institutions, through permitting these institutions freely to increase tuition, and through reviewing and possibly revising the tax code to encourage private and corporate donations. In their turn these institutions should focus on seeking independent and permanent sources of income from philanthropy, private giving, endowed chairs, and industry support, setting as their goals the permanent funding base and the high quality of the better private universities in the USA. Other Reform Measures 5.39 Dereglation. To increase quality, the Government should extend its recently proposed program of deregulation of professions, begun in 1990, when the Government proposed ending the requirement for a university degree to practice in thirteen professions, including journalism, economics, and architecture. Eventually Government should regulate only the medical professions, law, accountancy, dentistry, engineering and education. As has been experimented with in Sao Paulo in the area of medicine, and as exists in law, state-wide certification examinations (preferably managed by private non-profit agencies and institutions) should be developed for entrance into the above professions. These examinations should be of sufficient rigor to encourage higher quality in undergraduate education. The Government should publish information on the percentage of graduates from different institutions passing these examinations. The CFE should end its prescription of curriculum content and at the most should ask professional societies to provide non-compulsory guidelines. The existence of certification examinations would ensure that institutions adhere to minimum levels of quality in the remaining regulated professions and that diplomas alone will not suffice. 5.40 Evaluation. MEC should encourage improved quality through supporting a program of evaluations of higher education programs, both public and private, including support of independent peer review and evolution agencies. Since there is no single accepted measure of quality or cost- effectiveness, support should go to the annual publication (by a non-profit research group) of information on measures such as the following: (a) articles and books written by faculty and by departments and research institutions; (b) library books and periodicals per student; (c) percentage of students accepted by faculty and percentage of those accepted who enter; (d) percentage of graduate students by department; (e) number of professors by department with masters and doctorates; (f) ratings by peers of quality of undergraduate instruction, in much the same manner as the CAPES ratings of graduate education; (g) percentage of undergraduates going on to graduate education; and (h) percentage of graduates passing state-wide certification examinations. In addition MEC should support a program of research to measure quality in higher education, focussing on increases in knowledge, skills and income brought about by different higher education institutions. This would help to confirm or deny the judgments of informed observers about the quality of Brazilian higher education institutions and could eventually influence policies on financing, including using the student loan scheme to improve quality. 5.41 Training. To improve institutional cost-effectiveness, MEC would sponsor a program of professionalization of university management, to include seminars and short courses, utilizing the master's degree program at the University of Santa Catarina as well as the NUPES study group at the University of Sao Paulo. A fund could be provided, to be managed by CAPES, for graduate training overseas in university management. 49 5.42 Strengtening Graduate Education and Research. The current system of graduate education and research works well. Until recendy funding was adequate but should keep pace with inflation." Increased interaction with private industry should be encouraged. Furthermore, while overall the method of supporting graduate education is working relatively well, the first priority should be to Increase the numbers of students per course rather than to establish new courses. 5.43 A number of science and technology issues remain for Brazil. One issue is related to the adequacy and appropriateness of existing programs and funding agencies. In particular CNPq has become an excessively large bureaucracy which the Government will need to down-size. The long- term role of PADCT in the science funding system, after Bank financing is ended, is uncertain. In principle scientists should have a variety of sources for funds to ensure that new and innovative ideas receive a hearing. However an excessive number of agencies and programs may result in a dispersion of effort. 5.44 Another question of importance is that of the r active long-term roles of the states and the federal government in supporting S&T. The potential rapid increase in state fund!ng of S&T raises critical questions related to the role of the states in S&T and their institutional capacity for decision- making in the sector. It may be appropriate for the federal government to encourage states to reduce their unrealistically high funding objectives and to focus on primary and secondary education, with the federal government retaining its preeminent role in providing the bulk of the public funding effort in science. 5.45 The Government has created a number of highly directed research institutes in areas it considered important, such as space, aeronautics, agriculture, and the Amazon. Some of these institutes may have outlived their usefulness. The Government therefore needs to examine the balance between directed and non-directed programs. Similarly, there is a need to examine the extent to which the emphasis should be on ensuring that existing centers of excellence are of world class quality, on supporting emerging centers, on supporting regional development efforts, and on encouraging institutions to cooperate, so as to help ensure a critical mass of researchers and students and to maximize the use of expensive equipment. It may also be uecessary to encourage the weaker research groups to disband. 5.46 The most fundamental issue in science and technology is that of uncertainty with regard to whether and how private sector investment in R&D will increase. If this does not increase over the middle term, then Brazil's public sector investment will have an inadequate payoff. With regard to private sector investment in R&D, the Government which took power in March 1990 set as its goal the opening of the Brazilian economy to international competition. It has ended most quantitative restrictions on imports, and is setting up a system of tariffs which, while very high at present, is to be progressively reduced. The Government also intends to privatize the major state enterprises. The Government has recently become aware of issues related to intellectual property and technology transfer and recendy established a commission to examine overall issues related to strengthening its intellectual property and trade secrets legislation and enforcement as a means of encouraging industrial innovation. With regard to informatics, key changes are underway to end protection. 5.47 In addition to changes in govemment policy itself, a successful transition to a more competitive industrW sector will require changes in behavioral patterns of firms, relations between 3 Funding in 1991 and in 1992 of FNDCr (the National Fund for the Develoment of Science and Techno l.gy I has been severely cut. 50 management and employees, and relations between the productive sector, government, education, and the science and technology community. The Government is currently preparing a report, to be completed in two years, by an independent commission, which will review these types of issues. Some issues and possibilities include: linking the level of Government support to the level of private funding of individual R&D centers; determining the degree of public cost-sharing of technology development which has potential commercial value; and developing appropriate models for government/industry/university collaboration in establishing science parks and other mechanisrms such as "incubators" to support development of new technologies. Key issues related to the role of Government in encouraging private sector financing of R&D also include: the extent to which Government should financially subsidize private sector R&D; costs and benefits of using fiscal/tax incentives, directed credit lines, or public equity participation in R&D joint ventures; and needs to restructure capital market regulations to promote more appropriate financial instruments supporting innovation. The results of the report are expected to have a substantial impact on corporate strategic planniuig especially in the areas of research and development and technology adaptation, government policy in science and technology, program direction and linkages with the productive sectors in Brazilian research centers, and on the science and engineering curriculum in the education sector. 5.48 Ka Initiativ. One future possibility would be to encourage post-secondary two and three- year programs as a means of increasing manpower in areas such as nursing, primary and secondary education, and technology. Compared to neighboring countries, Brazil has very low enrollment in two- and three-year technology institutes. The area of primary and secondary education teacher training requires particular review, since the quality of entrants into the profession is reported to be declining because of low salaries and poor working conditions. To encourage students of low socio- economic status to get a higher education, special remedial programs could be set up. In addition an open university, using mass media and correspondence, would be, in principle, an inexpensive way of increasing access to higher education, provided a private non-profit institution was set up or contracted and fees were charged. 51 VI. FINANCIAL AND OTHER IMPLICATIONS OF REFORM 6.1 The government has an important choice to make with regard to higher education. It can do nothing and continue to waste large anounts of resources, assign an increasing proportion of higher education students to inadequate quality institutions, and fail to meet the economic and social demands for highly trained manpower of the 21st century. Or the Government can take the necessary steps for reform. 6.2 This chapter discusses the financial, enrollment, and institutional implications of higher education policy reform. It should be emphasized that the quantitative estimates of the impact of policy changes on higher education developed for this chapter are indicative only-designed to provide policy makers with rough orders of magnitude. The chapter analyzes two basic scenarios - a "with reform' scenario and a "no reform" scenario. 6.3 Underlying the quantitative estimates of the impact of the "reform" scenario are the following assumptions: * Unit costs of federal institutions will be reduced, on average, by one third (33 %) through efficiency gains. It is not possible to project variations among institutions or programs, but considerable differentiation can be expected depending on the types of programs of study offered by each institution. * Cost recovery will be introduced in all public institutions, with the eventual objective of charging for full instructional costs. * Government programs of support for research and post-graduate training will remain intact. But direct institutional subsidies for undergraduate education will be reduced pari-passu as institutions recover a higher share of their costs directly from students and take steps to improve efficiency (i.e., reduce unit costs). * As institutions increase cost recovery, the Government will expand the coverage and size of the student loan program, subject to reforms to improve its financial efficiens. In addition, a limited number of scholarships to assist the poorest students will be made available. Loan support will be determined by tuition fee levels in both public and private institutions. * Private sector institutions will be free to set tuition fees and enrollment levels as thIw choose. * Finally, over the next five years, the higher education system as a whole will grow h! about 3% per year. This rate would keep pace with increases in secondary school output, as forecast in a recent World Bank survey of secondary school education.' 3 World Bank, Issues in Secondary Education. The current trend in enrollment growth in secondary schood to 3.4% per year, while reforms leading to a moderate increase would lead to increases of 6.6% per year. But estimates of output growth are 3 % per year. 52 But growth in demand could well be less than 3% per year, especially if the current situation of under-investment in the lower education levels continues, and if tuition is introduced in public institutions. 6.4 Table I I summarizes key aspects of each of the scenarios discussed below, as well as some of their likely advantages and disadvantages. 6.5 Pailure to Reform. If the government continues on its current path of financing an elite and high cost system of federal higher education, federal expenditures would continue to be around $7,930 per student per year, largely spent for the salaries of teaching and non-teaching staff. While students in public institutions would receive a moderately good education, large amounts of financial resources would continue to be wasted, since student teacher and student staff ratios and physical capacity are substantially in excess of what is needed to achieve quality education. 6.6 Public enrollments would most likely remain close to what they are today, and expansion would take place in the private sector, most probably in the lowest quality institutions. Because of government controls on private higher education tuition, the higher quality institutions would not be able to accept more students without jeopardizing quality. The most likely result of umnet demand for public higher education would be continued expansion of low cost, low quality private institutions, leaving Brazil without adequately trained manpower to meet the economic and social demands of the twenty-first century. Furthermore, there would be virtually no assistance mechanisms for needy students. Finally, there would be a growing shortage of resources for public primary and secondary education. 6.7 As can be seen in Table 12, if no reform steps are taken and public expenditures per student continue to grow at the current pace, public funding is likely to reach $3.6 billion per year in 1995 in constant 1989 prices, compared with $1.4 billion of private expenditures. Despite the high percentage of students in the private sector, the federal government would spend large amounts of funds on inefficient public institutions serving a minority of students. 6.8 E.IIRfdnrm. Full higher education reform would include recovering from students the full costs students of undergraduate teaching costs. To maintain equitable access to higher education, the government would support a large-scale student loan program, combined with a limited number of scholarships for the poorest students. Thus, government expenditures on higher education would mainly be directed through the loan and scholarship scheme and through support, on a competitive basis, for research and graduate programs. The distinction between public and private higher education would be blurred. Reduced government spending on higher education would make funds available for investment in the lower levels of education, to cover any shortfall in financing research, or to other sectors. 6.9 The results of such a system could be large public savings allowing the government to invest its social budget in more pressing areas that represent a more equitable and socially efficient use of public resources. The higher education system itself would be run more efficiently, since competition promoted by ful cost recovery would encourage institutional discipline. A demand- driven system would be more likely to ensure that the types of training and education offered re- sponded to the needs of the labor market (as expressed through students' choices). 5casr for the Fture Seeao D Action Federal Private Federal Lan Pblic Enrou Enroiiass Unit Costs Program Ewpaditure Budget 1. No Reform Increasingly poor quality None Relatively Con- Growth at 2-3 % Remain Constant Remain Constant Remain Constant of instnton at very stant per year or Inctease or Increase high cost to government. 2. Full Reform Dynamic higher educa. Full cost recovery of Uncertain over Rapid Growth Fall by at Major increase Large Swings tion system that would undergraduate teaching long term. and differentia- least 113. be efficient, equitable costs. All institutions tion according to and responsie to stu- free to set tuition. Gov- market demands. dent demand. emntent support to sttu- dents via loans and 3. Refonr with Partial Cost Recovery 3a. Expand Public This option wotld ex- Require fedal isti- 50% Gtwt Slow Grwth Fall by 11 Increascs Modemte sav- Enollments while pand the system and tutions to increse en- ings Holding Spending stimlate efficitecy, rolmens by 50% witb Consmnt while achieving soam no increase in fending. budgetary savings. 3b. Reduce Spending This option would elirni- Reue fea insti- Remain Cout a Rapid Growth Fall by 113 Increaes Larg Savins while Holding atm waste in public tufions to maintain Enrollments Con- nstUon existing enrollmen s1ant and generate lae bud- levels and to swl extra getaly svings facilities and elimnat exss sff faculty. Table 11 54 6.10 Full reform would require major and difficult institutional changes and requirements. An efficient and income-targeted student loan and scholarship program would have to be established. Demand for some courses of study (such as basic sciences) may be suboptimal, because students' subsequent earnings will not fully reflect the "externalities" or spillover benefits which these individu- als generate for society, but which cannot be directly captured by either these individuals or their employers. Exacerbating this problem is the very high. cost of some types of education, particularly in the sciences. In some instances, students would not be able to afford the full cost of their education, even with a loan. Institutions would therefore have to cross-subsidize these programs by charging higher tuition for some of the lower cost programs. But competition from institutions which do not offer high cost programs may limit the capacity of the other institutions to cross-subsidize in this manner. Therefore, there may be a need for Government to maintain special programs of support for certain types of undergraduate education, in order to assure an adequate supply to graduates. As with support for graduate education and basic research, this government assistance could be channeled either to individual students or to institutions on a competitive basis. Another problem is that of the pervasive problem of "credentialism". The government may have to deregulate most professions and introduce objective entrance examinations in areas such as medicine to strengthen the awareness among students of the need for quality. Finally, the government would need to expand the amount of consumer information available to students through supporting a variety of peer and other evaluations of higher education institutions. 6.11 As can be seen in Table 1224 and Table V-3, if the government were to implement a full scale reform program, eliminating direct institutional subsidies for teaching and focussing its efforts on student loan and scholarship programs, annual government spending on higher education would fall by over 50% to $1.9 billion. Over half of this would be in the form of institutional subsidies for research and infrastructure and the rest would be the costs of student scholarships and of subsidized loans. The estimate of only 27% in annual real losses in the loan program from subsidization, default, and administration (lower than in all current programs) is achievable provik ed administration is efficient and income contingency schemes are used. Making up some of the expenditure difference would be a rise in private expenditures of US$1.0 billion to US$2.5 billion. Overall enrollments would grow approximately 20%. 6.12 Reform with Partial Cost Recovery. Recognizing that full cost recovery cannot be achieved overnight, there are important steps that should be taken to move in the direction of more efficient and equitable use of public funds in higher educaticn. A first step would be to reduce unit expenditures in federal institutions by one third through one of the following two strategies: (a) increasing enrollment in federal institutions by 50% while keeping public expenditures constant; or (3) keeping enrollment in federal institutions constant, and reducing teaching and non-teaching staff by one-third. Concurrently, moderate cost recovery would be introduced in public institutions - perhaps US$ 700 per year initially. This figure represents the current average tuition fee at private secondary schools in Brazil, and about 60% of the federal university students come directly from such 24 This table estimates the financial implications of the reform possibilities To illustrate the potential impact of reform, public and private spending are estimated for a snapshot one year period, in 1995. Again, it must be emphasized that these projections are indicative only and do not account for many of the complexities involved. The table projects public and private spending for the 'no reform" and afull reform" scenarios, as well as for the two proposed partial "reforms". 55 schools. US$700 would be equivalent to approximately 30% of estimated unit instructional costs after the above-mentioned cost reductions are made. In addition, tuition and enrollment controls in private universities would be eliminated. To offset the impact of higher tuition fees and also to promote increased choice amnong educational institutions, the existing student loan program would be expanded both in size and coverage, providing about 25% of students with loans equal to tuition fees in public institutions. 6.13 The strategy of increasing enrollment in federal institutions by 50% ((a) above) would bring more people into high quality institutions, create incentives for the elimination of some of the lower quality institutions, increase the level of financial support to the neediest students and generate a moderate financial savings for the government. Overall, federal institutions would be required to admit 50% more students while receiving no additional fiscal support. Incentives in public funding allocations could be established so that institutions would have to respond to student demand in order to receive funding. However, it would be crucial to ensure that public sector expansion occurs only where cost-efficient, with expansion differing by institution and by field of study. The 50% target would be an average for public higher education, not a target for every institution. The government would need to develop a funding mechanism to ensure that expenditures remained in line with true costs and to provide incentives for efficiency and quality improvements. 6.14 On this basis, overall (state and federal) enrollments in the public sector would- grow by about 35%. Despite these enrollment increases, overall public expenditure on higher education would decline by 11 % to about $3.2 billion. An annual savings of $400 million would be generated over the expected 1995 costs. The overwhelming portion of public spending, however, would still be in the form of direct institutional subsidies. Resource mobilization from private sources would increase to $1.9 billion per year. Significant enrollment expansion in higher education would be achieved with less public funding and without a loss in quality. Although expansion would be primarily in public institutions, it would be accomplished through an increase in private resource mobilization and a redirecting of public funds to the student loan program. The reductions of unit expenditures in public institutions would not affect the quality of teaching, since institutions would utilize spare capacity in terms of teachers and physical facilities. 6.15 An important advantage of 50% enrollment growth in the federal institutions is its feasibility. Supporting the claim that expansion is feasible, in 1990 federal institutions promised 30% growth in their enrollments over three years". Excess teaching and physical capacity could be used without resorting to the complex and politically difficult task of firing staff and selling off facilities. 6.16 A criticism that is likely to be leveled against rapid growth is that standards of admissions of admission would have to be lowered. This argument is somewhat overstated since federal institutions today turn away 86% of applicants. Assuming the number of applicants to these institutions remained constant, the institutions would still be turning away 79% of all applicants, t'r accepting I in 5 applicants - still a high degree of selectivity. Nonetheless, there will most likely be some fall in the standard of students entering public institutions. I As of this writing, it is not clear how and under what conditions this expansion will take place. Single Yeaw Fiuacial bapact of higher Education Refocn Measr (esat f 1995) Cot to Aal Public Goverment Costs of TOW Governue base cx- associated scholar- public ex- Private cx- Toa ex- savings du peaditur with loas ships penditure peaditurs peaditue to rdorni 1. No reform 3,530 86 0 3,616 1,524 S,140 - 2. Full Refom 1,018 484 251 1,7S3 2,517 4,270 1,S63 3. Reform with Partial Cost Recovey a. 50% erollmet growth in federal iastutions, contant federalbudgd 3,030 84 88 3,202 1,906 5,110 414 b. no enollment growth in federal instutions, federal budget decreased by one 2,250 84 83 2.422 2,198 4,620 1,194 third NOTES: Reforms refer to seenaos prsted in Tale I1. SYstewide enrollment in 1995 is proecWted as 1,795,000, baseod on 3%S Ima growth fiom 195. See daso Tabe V-3 for detailed projections of enrllmt ad expeitumrs by type of goventing authority. I. Public base expenditures for 'no remtf assmu that all expansion occuzs in the private sctorad ano change in unit cos at public insuions. F rfom asnUmes that tho government's ony direct expendiure to higher educato wil be bt current lvel of fixed coas for hospital and besea, which ae esimated at USS 50 per sudent for hoptls anW $1500 for reseach. Psial refontprgam(a) assumespublic ee to all insitutionsdoesnot icra,the goverm cnts US70pes, and ero t incasesby 50% in public instions. Pasia itfom progam b) amsumes that public expenditures for innwtion (i.e. net of esearch and hospt esiats) are educed by one-dhird, akog with cost covey of $700. and enutats ae fm ia publi institutions. 2. Loans exedriture calcuaed as follow: for no rsfotm, it is the curnt estima annual Wpending for the loga pram. For AuU reform, the lon progrm charge 1% rea interat rate, loans cover 50% of studets, with aveage loan sze of $2000, and sudent have a one yeargawc period and scit year to rpy. WhitW the oan program would not be receiving full revenues In relation to its anual expendiur with years, tb cakatoc ere [am dos so t it is assumed th th4 progra is yidin revenues. le estimate is that the progsmwout4 equi27% of uan capia dseach year. Th meld tae for paal fotni except thah4cedk> an average of $700and covers only 25% of thde sdents. 3. Scholarship spending is assmd to cover 7% of suet with scholarsips equal in size to theloo mut. a woluld be directed atsudent coinin from famailis with earigs between 0 and 2 minimum salanries. 4. Private expenditures are the smn of iWons paid in both public and pivate sectors. No efor assumes no changc in tn poicy inher scto.r. Fu reform aues fees of S250 in public sector and 20% avecage fee rie in privae sector, !t inceasd spending on loans an wChOlahnips.til vfom swmpblic sWCtsor fee go to $700 wbhil pnvat sector feea rise by 20%. S. Annual public savings is the difference between public expeadiure for fth with and without meforn rcenaios. Talble 12 57 6.17 The strategy of holding enrollments constant while reducing staffing by one third ((b) above) would require public institutions to fire excess staff and faculty members as well as possibly sell off under-utilized facilities. Some mechanism would have to be set up to ensure that cuts are not across the board but rather in the areas of lowest efficiency. Some faculties might well be closed down. It would be necessary to fire university staff and assist them to find employment in the - expanding private institutions. Innovative and entrepreneurial institutional managers would need to ensure retention of the best teachers. 6.18 This strategy would allow the public sector institutions to remain elite, but require them to be more efficient by eliminating redundant staff and facilities. There might be significant enrollment increases in the best private institutions, as well as a rise in private resource mobilization to US$2.2 billion. As noted in Table 12, public spending would fall to $2.4 billion--a savings of $1.2 billion per year, or 35% off the no-reform scenario. 58 BIBLIOGRAPHY Albrecht, D., and Ziderman, A. "Deferred Cost Recovery for Higher Education,' World Bank, 1991. BraziL (1988). ConstituicLo da Repnblica Federativa do Brasil. Brazil, Ministdrio da Educagio. (1989). A Isonomia no Contexto da Polftica de Recursos Humanos das Instituicoes Federais de Ensino Superior. Brasilia: SESUiMEC. Brazil, Ministdrio da Educa;ao. (1990). "A Educaqao no Brasil na DWcada de 80," Brasflia: MEC Brazil, Ministdrio da Educaglo. (1990). "Organizaggo das P,incipais Iddias Contidas nos Documentos Apresentados pelos Grupos de Trabalho Institufdos pelo Ministdrio da Educaqgo." Brasflia: SENESUIMEC. Mimeo. Brazil, Ministdrio da Educaglo. (1990). Sistema de Acompanhamento e Avaliacfo: Resultados da AvaliacLo vor Area do Conhecimento - 1978-8. Brasiia: CAPES/MEC. Brazil, Ministdrio da Educagao (1985). "Uma Nova Poiftica para a Educagao Superior. Relatdrio Fmal da Comissio Nacional para a Reformulagao da EducacAo Superior. Brasiia, mimeo. Brazil, Ministdrio da Educacio. (Several Years). Sinopse do Ensino Superior. Brasilia: SEEC/SG/MEC. Brunner, J (1990). 'La Educacidn Superior en Chile: 1960-1990: Evoluci6n y Polfticas." I Pocumcnt presented to the meeting of the regional project for comparison of higher education policics. Buenos Aires: FLACSO. Camargo, J.M., et al. Distribuicao de Renda no Brasil. Paz e Terra. Rio de Janeiro, 1991. Clark, B. R. (1990). "Basic Categories in Cross-National Understanding of Higher Education.' Paper prepared for the Ford Foundation meeting on higher education in Latin America, Novcmhcr 15-16,1990, New York, NY. Castro, C. de M. (1990). "Specialized Education for General Jobs: The Market for Higher Education in Brazil." In International Journal for Educational Development. 10:4, pp. 245. 252. Cunha, L A (1990). "Public Policies for Higher Education in Brazil." In Higher Education Policy. 3:2, pp. 21-25. Divonzir, A. G. et al. (1990). EducacAo e Cultura - 1987: Situagco e Poifticas Govemamentais. S4rie IPEA No. 128. BrasSlia: IPEA/IPLAN. Durham, E. R. (1989). A Autonomia Universitaria. 0 Princfgio Constitucional e suas ImplicaicSs. Document 9/89. Sao Paulo: NUPESIUSP. Durham, E. R. (1988). "O Ensino Superior: Contribui;o ao Debate sobre os Fundamentos de urns Nova Lei de Diretrizes e Bases." Paper presented at the 461 meeting of the Conselho de Reitores das Universidades Brasileiras, March 1988, Florian6polis, SC 59 Durham, E. R., & Schwartzman, S. (1989). Situacflo e Perspectivas do Ensino Sunerior no Brasil: Os Resultados de um Semin$rio. Document 3/89. Sao Paulo: NUPESIJSP. Efectos Sociales y Econdmicos del Credito Educativo. (1989). Bogota: APICE. Farret, M. M. L (1985). Exoanslo e Elitizacgo do Ensino Superior no Distrito Federal. Unpublished master thesis. GaCetani, F., & Schwartznan, J. (1991). Indicadores de Produtividade nas UJniversidades Federais. Document 1191. Sao Paulo: NUPES/USP. Gomes, M. (1990). "Personal Background of the Brazilian University Student." Mimeo (Preliminary results of doctoral dissertation). Wuja do Estudante 90. Sio Paulo: Editora Abril. Guimaries, E. A. (1989). 'A Evolugxo dos Custos e das Mensalidades das Instituio8es Privadas de Ensino Superior." Mimeo. Heyneman, S. P., & Etienne, B. (1988). Higher Education in Developing Countries: What. Ho_w and When?. Washington: The World Bank, EDI Working Papers IBGE. (Several Years). Anuario Estatfstico do Brasil. Rio de Janeiro: IBGE. 1DB. (1988). Economic and Social Progress in Latin America. Special Section Science and Technology. Washington, DC: IDB. IPEA. (1990). Para a D&ada de 90: Prioridades e Perspectivas de Polfticas Pdblicas. Volume IV: Polfticas Sociais e Organizagio doTrabalho. Brasflia: IPEA/IPLAN. James, E. (1979) "Product Mix and Cost DLaggregation: A Reinterpretation of the Economics of Higher Education." Journal of Human Resources 19,79, XII, 2. James, E. (1989). Differences Between Public and Private Higher Education: An International Perspecv. Mimeo. James, E. (1988). Philippines Education Sector Review. Washington, DC: The World Bank. IAal, C. L S., & Werlang, S. R. da C. (circa 1989). "Retomos em Educagio no Brasil: 1976-86." Mimeo. Leal, C. L S., & Werlang, S. R. da C. (circa 1988). "Polftica Educacional no Brasil." Mimeo. Ivy, D. C (1986). Higher Education and the State in Latin America. Chicago: The University of Chicago Press. Mendes, C., & Castro, C. de M. (Eds.) (1984). Qualidade. Expansio e Financiamento do Ensino Superior Privado. Rio de Janeiro: EDUCAM/ABM. Meyer, V. (1982). An Analysis of Alternative Tuition Policies for Brazilian Public Hig-her Education. Unpublished doctoral dissertation. 60 '0 Ensino Privado e o Poder Pilblico.' In Estudos. No. 2. Brasilia: ABM. "0 Ensino Superior Particular." In Estudos, No. 1, Braslia: ABM. OECD. (1989). Evolution des Modes de Financemente de l'Enseignment Superieur. OECD Monographs on Higher Education. Paris: OECD. OECD. (1988). Chatg Patterns of Finance in Higher Education: Japan. OECD Monographs on Higher Education. Paris: OECD. Plank, D. N., Sobrinho, 3. A., & Xavier, A. C da R. (1991). Born Old: Why Brazil Lags Behind in Educational Deelo2ment. Unpublished draft manuscript Paul, I-. (1991). "Brazilian Higher Education: An Economic Approach to its Rationality". Draft mimeo. Paul, I-J., & Wolynec, E (1990). 0 Custo do Ensino Supenor nas Instituicoes Federais. Documento 11/90. SAo Paulo: NUPESIUSP. Paul J-J., & Ribeiro, Z. D. (1990). n As Condi6es de Vida e de Trabalho dos Alunos do Ensino Superior Brasileiro." Mimeo. Paul, J-J., & Ribio, Z. D. (1989). "Diversidade da Oferta e Estratdegias dos Vestibulandos: 0 CAU) de Fortaleza! In Cadernos de Pesguisa. Rio de Janeiro: Funda;io Carlos Chagas. Psachampoulos, G. (1987b). "Education and Development: A Review", Research Observer Vol A. No. 1. Washington D.C.: The World Bank. Psacharopoulos, G. (1981). "Returns to Education: An updated International Comparison", Worhd Bank Reprint Series: Number 210. Washington, D.C.: World Bank. Psacharopoulos, G., Tan, J-P, & Jimenez, E. (1990). 'El Financiamento de la Educacidn en LAX PaLses en Desarrollo.' In Docencia Post-Secundana. 18:2. Psacharopoulos, G., Tan, J-P, & Jimenez, E 1986. Financint Education in Develoging Countnrw Yi loration of Polic Options. Washington, D.C: World Bank. Roth, 0. (1987). The Private Provision of Public Services in Developing Countries. Washingtn. DC: The World Bank, EDI Series in Economic Development. Saiba, A. (1990). ITe Vestibular Examination and Access to Higher Education in Brazil: Ev'Jicnt £rom Braslia Unpublished doctoral dissertation. Schieelbein, E 1990. "Chile: Economic Incentives in Higher Education", in Higher Education Policy. Vol. 3, No. 3. Schwartzman, 3. (1989) "A Selectividade S6cio-Economica do Vestibular e suas Implica;oes p:.t a Polftica Universitria PgblicaW In Educacao e SeleIo No. 19. Rio de Janeiro: Fund3a4U Carlos Chagas. Schwartzman, S. (1990). "A DiferenciaCio do Ensino Superior no Brasil" Mimeo. 61 Schwartzman, S. (1988). "Brazil: Opportunity and Crisis in Higher Education." In Highe Education, 17:99-119. Schwartzman, S. (1989). "Ciencia, Profissoes e a Questao da Autonomia. Documento 8/89. Sao Paulo: NUPES/USP. Souza, E. M. de (1980). Crises e Desafios no Ensino SuRerior do Brasil. Fortaleza: UFC. Tan, J-P & Mingat, A. (1989). Educational Development in Asia: A ComDarative Study FocusingEon Cost and Financing Issues. Internal Discussion Paper #51. Washington, D.C.The World Bank. Tramontim, R., & Braga, R. (1988). As Universidades Comunitgrias: Um Modelo Alternativo. Sao Paulo: IPEA/IPLAN/CEC. Tramontim, R., & Braga, R. (1990). "Ensino Superior: Perspectivas para a Ddcada de 90." Brasflia: IPEA. Mimeo. UNESCO. (Several Years). Statistical Yearbook. Geneva: UNESCO. US Dt)epartment of Education: Annual Statistics (1989). Vahl, T. R. (1980). A Privatizaclo do Ensino Superior no Brasil: Causas e ConseqUencias. Floriandpolis: UFSC/Editora Lunardelli. Veloso, ..R. & Bastos, V.L. (1984). "Ensino Superior e Subemprego: Um Estudo de Caso em Brasilia'. In C. Mendes & C. de M. Castro (Eds), Oualidade Expansao e Financiamento do Ensino Superior (pp. 129-155) Rio de Janeiro: Educam/ABM. Winkler, D. (1990). Higher Education in Latin America. World Bank Discussion Paper #77. Washington, DC: The World Bank. Woodhall, M. (Ed.). (198?). Financial SuR,ort for Students: Grants,. Loans or Graduate Tax? London: Kogan Page. Woodhall, M. (1983). Student Loans as a Means of Financing Higheflr Education: Lessons from International Experienc. Staff Working Paper No.599. Washington, DC: World Bank. Wolff, L. and Albrecht, D., ed., Higher Education Reform in Chile. Brazil. and Venezuela, "A View from LATHR' No. 34, December 1992. Washington, D.C. The World Bank. The World Bank. (1986). Brazil: Finance of Primary Education. Washingtoin, DC: The World Bank. The World Bank. (1989). Brazil: Issues in Secondary Education. Report No. 7723-BR (green cover). Washington, DC: The World Bank. The World Bank. (1988). Brazil: Public Spending on Social Programs: Issues and Options. Report No. 7086-BR (gray cover). Washington, DC: The World Bank. The World Bank. (Several Years). World Development eport. Washington, DC: The World Bank. 62 ANNEX 1. CRfTERIA FOR ALLOCATING FUNDS TO PUBLIC INSTITUTIONS 1. Recent discussions of public finance for higher education have begun to examine the criteria and mechanisms through which govemments channel money to institutions. To further that discussion for Brazil, this annex attempts to provide a comparative interaional framework with regard to what other countries have been doing in this area. The annex examines some of the key issues facing countries, and the different strategies used by some more innovative governments. 2. In general, there have three methods through which governments have allocaed resources: (i) incremental or political budgeting, in which budgets are decided on the basis of the previous years budget; (Hi) formula or input budgedng, in which the govenment uses costs norms and coeffients to establish a cost per student. This cost is multiplied by the mber of students at an institution; (iii) performance or output funding in which the government tries to promote efficiency gains or quality improvements through its allocative criteria. In this last category, gov ens detmine expected outputs from an institution and allocate fuding as instimions satisfy those output expectations. 3. It was argued in the body of the report ta incremental budgeting is the poorest method for allocating resources since government expenditures do not bear any relationship to the true costs of education. Incremental budgeting also fails to provide institutions with incentives to improve the way they conduct their teaching and research. It was recommended that the government move towards a normative type of funding formula, which may include output (performance) incentives. Tle analysis below tries to explain some necessary considerations in both formula (input) allocations and performancetoutput budgeting. A final section summarizes the procedures and criterias that some innovative countries have utilized. 4. hlafndin seeks to determine what it costs to teach one student per year and with that figure, determine a budget based on institutional enrollment. Formulas can be extremely complicated, depending on the way the unit cost coefficients are detemined and adjusted. These calculations can send powerful signals to insttutions about the types of activies they should engage in. Weightings to the unit costs have been developed because costs vary by field, degree, and types of activeies. The most common weightings are listed below: (a) By field of study: Engineers are more expensive to teach than lawyers, and most formWas are weighted by field of study. (b) By level of education: Variation by degree or program type. (c) By type of institutions (Institutional profile): institutions may incur different costs due to their location, their overall size, their mission within the overall education system. Each institution may receive a category of weighting to reflect these differences. (d) By type of students: it is important to distinguish between part and full time students since part time students cost less. 5. Straightforward formulas give institutions tremendous incentives to expand, since higher enrollment implies a larger budget. While formula funding worked well during expansion phases of university systems, some imrtant problems began to appear. (i) Expansion, as well as government commitments, became open ended at times when resources were becoming scarce. (ii) Formulas were sometimes too confusing or conveyed unintended signals. (iii) A third problem was that institutions 63 were becoming inefficent. Formulas were designed without and expectations of efficiency improvements or economies of scale. (iv) A final issue arose with regard to research funding. Most countries include some or all of their research finding in the formulas. Therefore, research funds are dependent on the number of students, and not the research activities themselves. While developing a research base may be a function of the number of students, it is not clear that all research money should be determined on the basis of enrollments. For this reason, most developed countries have begun to channel research funds on a selective basis, and through different institutions and criteria. 6. The problems listed above have led some governments to allocate money to institutions on the basis of performance or output. Two major categories of output bave been considered. One has been the need for efficiency. Such a funding mechanism ties allocations to the ability of an institution to reduce its unit costs or move students through their institution more rapidly. The second output funding approach attempts to link fimding to institutional quality. 7. With regard to reducing unit costs, the following can be done: (a) A marginal cost enrollment formula recognizes that it is possible to attain of economies of scale at institutions since expansion at the margin need not be funded at the same cost as the average. That is, a funding formula can be modified so that institutions institutions are paid less to teach an extra student. The goal of this kind of formula is to promote expansion only when it is cost efficient. This strategy has been utilized in Ontario, Canada, where expansion of institutions is now at a discounted unit cost, rather than the whole unit cost. (b) Contract/price bidding as in the UK, forces institutions to compete for fimds. In a bidding scenario, institutions submit prices at which they are willing to teach students. The funding body then purchases the bids from instiutions on the basis of their price and their quality. For the past two years in the UK, higher education institutions have submited prices to their fiuding bodies at which they are willing to teach students. The funding bodies assess these bids which are adjusted for the quality of an instiution. The bodies then purchase the lowest bids. The result of this approach is that institutions have been required to lower their unit costs in order to obtain more fimding. 8. The second area of efficiency that some countries have sought to improve through their fuding mechanisms, regards student flows - the rate at which studens compelete their studies. That is, in many institutions it is necessary to curb inefficiencies that result from students who drop out, or who take longer than the official length of their course to complete their studies. The following strategies have be used to redress these problems. (a) Funding per graduate rather than student: An enrollment formula can be adjusted so that e calculation does not pay on the basis of what it costs to teach a given student per year, but rather, payment can be made on the basis of what it should cost to graduate a student from a course. Funding per graduate implies that for each student matriculating in a four year course, the government will pay only four times the annal unit cost. Such calculadons of what it costs per graduate rather than per year of teaching have been implemented in Finland and Holland. Denmark has taken steps to reward institutions that have successfully promoted efficient student flows. 64 (b) Differential funding for graduate and dropouts: A further reform in this direction is to distinguish between graduates and drop outs. In Holland, such a distinction is built into the fimding formula - the government pays one rate for a graduate and another (discounted rate) for a dropout). The differences are severe, and institutions are under pressure to determine early on which students are most likely to drop out. 9. Quality funding: Besides rewarding efficiency, allocations systems can include incentives for quality. But quality is perhaps the most difficult characteristic to control for. Nevertheless, the Chileans, Japanese and Israelis have implemented three different methods to promote the quality of their undergraduate educadon through their allocative decisions. Ihese are discussed below. (a) Student choice as an IndIcator of quailty: Using student choice to promote education quality defines and rewards quality on the basis of student perception. This strategy sm from the ideas of 'voucher funding' whereby students receive money (or a voucher) from the government to attend any instituton they choose. In theory, a funding formula acts as a voucher, since institutions receive money on the basis of the mnmber of students they are able to attract. However, many insttutions have ceilings on their enrollments, as expansion may not be a desired policy goal. In this case, enrollment formulas cannot guarantee quality. The alternative is is to differendate the amount of money given to each student on the basis of his or her quality. That is, the individual students in the fimding formula can themselves have weights, according to their quality (acadenic success). The Chieans system awards funds to insitions on the basis of its ability to attract the best studens. Competition among institudons is expected to raise the overall quality of the system. (b) Peer review: A second approach defines quality in terms of the subjective opinions of experts. Allocation decisions are then adjusted on the basis of peer review assessments. ndependent audit bodies cary can submit repous to the body in charge of fiuding decisions. This approach has been used in Israd, where institutions are evaluated both for their quality and efficiency by independent audk groups, and in the United Kingdom, where the bid prices submited by polytehnics and colleges are adjusted for the evaluation they receive by Her Majesty's Ispeorate. In Brgazil, sh a system is in place for graduate education, which is funded on a peer review basis adminstred by CAPES. (c) Empirical performance Indicators: A third approach is to use empirical statistical indicats to evaluat quality. This has been attempted in Japan, but only for public support of private instittions. In this case, a base level of fuding is calculated using an enrollment formula, and this result is adjusted between -30% and +50% on the basis of performance indicators. ITe four performance indicators that are used for the adjustment are: (a) total enrollment in retion to the goverment's taget enrollment; (b) student-teacher ratios; (c) the ratio of non-salay expenditues to total expenditures; and (d) whether budgets are balanced or not. Ihe results of this approach have been somewhat diappointinS. While in' tutions were saved from bankruptcy, most of the money was used for increasing teachers salaries rather than quality improvements. 65 Cae Vexamples: 10. More detailed summaries of the mechanics of resource allocation are descnrbed below for seven countries: these include Chile, Denmark, Holland, Israel, Japan, Ontario Canada, and the United Kingdom. The summaries try to give some background information that is necessary to understand the context of the allocations procedures. CHILE 11. The Chilean higher education system underwent major fnancial reform in 1981. These reforms changed the allocations criteria, increased student contributions, and imposed broader structural reform of the systems by promoting truly prvate vocational institions. Before the refors, there had been parallel private and public sectrs which w treated identically in tem of public fiance. 12. Publicly fimanced institions receive funds through several sources which include: direct and indirect fiscal support from the Ministry of Education, fees, research grants from the national commission for scientific and technological research (CONICYT) and from semvce contra. The objectives of the fincial reforms were to introduce competitiveness into univesity fiuding (to promote quality), and to increase the contribution from students. 13. The reforms established three channels of public 'core fimding'. Under the proposed reforms, the old system of giving direct grants to universities was to be reduced to 50% of is 1980 level. These funds would still form a base for teaching and research. Tley had been determined both on enrollmen and historical contingencies. The second channel of public funds would be "indirect", on the basis of an institutions, ability to atact the 'top students" as determined by their performance on the national erance exam. The thi channel of funds would be to provide for institutionally managed student loan programs. Each institution would receive funding to create student support mechanisms which they would administer. 14. Since 1981, several modifications have been made to the original plan, mostly in the way the indirect funds wer determined. At first the indrect finding was based on an institution's abiliy to enrol students who scored among the top 20,000 grade (out of approximay 30,000 students who matriculate each year) on the secondary school exam. This sytem, because it reinforced eitin quality differentials, was replaced by a more flexible system, whee all new students carried funding, but were weighted according to their performance in secondary school. Another change was the elimination of the weighting of indirect fimding by field of study, It was felt tha the weightings promoted too many people in fields with best ii onal rewards. DENMARK 15. Denmark uses a norm based enrollmen formua so calculate a mndmim allocation for all its institutons. Supplemenl fuds are avaflable on the basis of performance in tms of student flows. The enrollment formula allocation is derived from student quotas for each field of study, which are set by the Ministry of Education. The MOE calculates an insitioD's stag and research needs based on these student quotas. Budgets are then divided by main fields of activity - research 66 and teaching and development plans. These calculations form a minimum allocation for each instddon. 16. Since 1987, the MOE has adjusted the enrollment formula allocations on the basis of student graduadon performance. Overall funding remains within a constant budgetary ceiling. The adjustment funds come from a pool of fimds that was established by withholding 5% of the fiuds normally allocated by the enrollment formula. The performance fnds are awarded on the basis of the number of courses completed. HOLLAND 17. Holland's higher educadon system consists of two categories of instidutions, all of which are public. There are Universities and the more vocationally oriented Institions of Professional Study (HBO's). 18. The Dutch fimding formula is extremely complex, and has therefore been problemaic both for institons and the govenmment. It includes a base amount to be allocated to Insdtutions for teaching and research that is enrollment independent, an enrollment sensitive component which is reassessed using a performance formula which distinguishes graduates from dropouts. This calculation has created very strong incentives for the rapid selection of potential drop-outs and a sdmulation of rapid graduation. 19. The university allocations are composed as follows: (a) 23% is fixed year to year, and is based on institutional profiles - size, number of faculties, etc. *b) % is vaiable, and is based on student numbers, using nor such as student teaher, and administrive staff to academic staff ratios. The cufrent trend is to separate funds for research, and make them independent of student mbers. (c) The student based component (b) is adjusted by allocating 4.5 student years worth of funding per graduate that completes his/her course of study and 1.5 years for dropouts, with the intnon of speeding up any "weeding out" of poorly performing students. 20. Thre main problems have arisen with the Dutch allocation system, despie the atempt to establish a rational system to promote the external and interna efficiency of institutions. (a) The complexity of the system makes institutions uncertain about the impact of their behavior. The consequences of changes in the parameters used in the many formulas is difficult to assess. (b) The system seems to reinforce (mirror) the current state of afairs, rather than allowing flexibility in innovation. (c) Instiudons are unable to identify activities which should be part of their specific mission. TBe system emphasizes conformity of institutions rather than diversity. 67 (d) The performance system considers time, but not process, how learning occurs. It inhibits "later in life" learning, and does not correspond to the reality of study. LSRAEL 21. The Israeli higher education system consists of 7 Universities, 1 Open University, and 7 non- Univerity institutions. Funding is through a buffered organization, the Planning and Grants Committee (POC). In 1980, PGC funds accounted for 75% of HE fiuds, but this has fell to 59% in 1986. 22. PGC finds are broken up into the following areas: Direct allocations for operating expenses, earmarked funds for research and equipment, matching funds, and development funds. Direct funds constitute 75% of the PGC total grant. 23. Allocations are awarded on the basis of both input and output criteria. The input criteria are essentially enrollment based, the PGC compares university budget proposals with a set of indices (parameters) for s/t ratios, and overhead costs. An independent examination of each institution is conducted to assess its productivity - both in terms of graduate and research output. If the results of the output assessment contradict the calculations by the enrollment formula, then funding can be reduced. 24. Research funds within the PGC are competitively awarded, and are based on the reports by external review committees. JAPAN 25. Japanese higher education consists of a reladvely small and elite public sector and a large private sector containing high quality and lower quality institutions. About 80% of enrollments are in the private sector. During the 1960's, the government encouraged rapid expansion of private insttions. Many, however, confrontel severe financial problems during the 1970's, and the government embarked on a policy of fiscal support for the private sector. 26. Public (National) Universities are funded from a 'quasi" buffer body called the National Schools Special Account (NSSA). The Ministries of Finance and Education approve an overall budget to be put into a National Schools Special Account (NSSA) each year. All fee and service incomes at universides also go to this fund, i.e. fees do not go to individual instiutions, but to the national account. Each year, the total income to the fund is divided among the national universities on an enrollment formula basis. The NSSA, founded in 1964, funds universities, attached hospitals and research institutions. Fees constiute approximately 15% of all public revenues. 27. The enrollment formula used by the NSSA includes weights by faculty and institutional needs. The current public sector funding mechanism is not thought to be very effective in encouraging institutional efficiency or quality. Also, the policy of transferring fee money and extrn funding to the national aount, has discouraged institutions from raising their own income. 28. Since 1972, the government has provided subsidies to the private sector through the Japan Private School Promotion Founion. The goal of this fimding was to raise quality and limit further expansion of institutions, while also restoring financial solvency. The mechanism for assessing the 68 allocation to a particular institution relies both on input and performance criteria. The base subvention figure is calculated on the basis of institutional enrollment and assessment of its cost. It is then adjusted (between -50% to +30%) based on the institutions conformity to four output based norms. 29. Performance is judged in four areas. (i) The ratio of actual students enrolled to number suggested by muinstry. (ii) Student teacher ratios for each faculty (iii) the ratio of non-alary educational expenditures to student derived revenue (higher the expenditre and lower the charges, the better - quiality without fees). (iv) budgetary balance. Instiutions are penalized or rewarded according to their performance in these 4 areas. 30. The maXimu allowable subsidy is 40% of Operational expediture. Although the MOE ineded to penalize insutions for overshooting its enrolment targets, it is interesting to note that the private sector admits about 50% more students than authorized. 31. The results of the intervention have been: (i) private institutions are now financially stable; (ii) enrollments have levelled off; (iii) quality improvements have been uncetain - most of extra money Went to big increases in staff salaries; (iv) fees have sharply risen (despite measures taken to limit fthis) ONTARIO, CANADA 32. Ontario's university system consists of 15 Universities - ranging in size from 4,000 to 60,000 students - funded through the Ministry of Colleges and Universities (MCU). Since 1967, the MCU has allocated funds on an enrollment formula basis. During the 1970's, the formula was intended to provide incentives for cost-effeetive expansion of the university system. Enrollmens and goverment expenditures in the sector gew rapidly. In the 1980's, however, there has been a strong need to curb expansion and contain total cost. Nearly 90% of all university support comes from the government, and 92% of operational expenditures has been disWibuted on the formula basis. 33. Allocation decisions were initially based on a fnding formula that was weighted by field of study %ud degree level. However, as the issues confronting the system changed (i.e. as resources became more scarce), the funding formula had to be modified. The most significant change was a discoundng of enrollments toward. marginal rather than average cost. Ihis was intended to eliminaae any icentive for expansion. A first experiment reduced funds for new enrollments by 50%. But expansion sil continued. The latest formula applies a more flexible approach to marginal cost fimding. The MUC witholds funds to reward specific policy agendas such as maintning quality of research and post-graduate education. There is a maximum level of cost-recovery allowable (13% of cost) before which any additional fee revenue will be deducted from the grant. The MUC expecs efficiency improvements from year to year ( and examines the previous year's grant before determining the new one). To maintain stability in fimding - and prevent fluctuadons in enrollment from upsetting planning - the MUC now uses enrollment bands of 3% in which an institution's funding will not change. 69 UNTrED KINGDOM 34. Over the last 25 years, the funding system in the UK has undergone several reforms, and it is best to understand the funding system chronologically. The British HE system is binary, with funding and management distinction between Universities, and Polytechnics and teacher training colleges. There is one minor private institution that receives no public support. There are theoretical fees, but these are paid indirecdy by the government for all British students. 35. The Universities have been funded, until this past year, by a buffered funding body, the University Grants Commission (UGC), which has been replaced by a less autonomous state body, the University Funding Council (UFC). The Polytechnics and Colleges were funded separately through a National Advisory Board which has also been replaced by the Polytechnic and Colleges Funding Council. Universities 3 6. Before 1980, the UGC served as a buffer between the government's broad policy agendas and the goals of autonomous Universities. The funding body developed five year that g. 'ranteed funding over that period. Until the late 1970's, the main emphasis was on expansion of the HE system. Grants were almost entirely block grants, to allow institutions full freedom in internal allocation. Specific grants for capital expenditures were more restricted. The block grant contained an unspecified portion for research activities. Research funding was therefore dependent on enrollment. The guaranteed five year plan of fimds collapsed in the wake of economic recession and falling birth rates in the late 1970's. 37. Several changes were initated in 1980 and 1981. First, overseas students were charged full fees. At first, this caused a severe cutback in overseas enrollment, but as istutions altered their course offerings to cater to their demands, and as they actively recruited students, overseas enrollments grew - providing a significant source of income for successfil institutions. in 1985, 60% of Universities had a committee to promote recruitment of foreign students, while 65% has a scheme of financial incentives to departments to make efforts to recruit. There are claims that "ghetto" courses have been created solely for fhese students. 38. Second, severe cuts in grants were announced of 8.5% over 3 years. The UGC presided over a 5% cut in student enrolments - on an ad-hoc basis based primarily on the quality of institutions and programs. (Polyechnics were still expanding rapidly, despite expenditre cuts of 15%). Funding was now on an overall expenditure basis, using more ad-hoc detminations based on 1980 expenditures, rather than on formulae. The crisis of fimding undermined the role that the UGC was supposed to play as a buffer. It became dictatorial in regulating enrolments and other areas of institutional management. The UGC grant fell from 70% of University funds in 1970 to only 57% in 1987, while research huids had risen from 13 to 18%, and overseas fees from 0 to 14%. 39. The reforms of 1988 replaced the UGC and NAB with the UFC and PCFC respectively, removed polytechnics from local authority control, abolished tenure, and suggested alteration in student financing policies. The bidding process was primarily intended as a mechanism to reduce unit costs and improve the efficiency of institutions which were seen as being rather inefficient and elitist. 70 40. The University Funding Council operates on a system of bidding for contracts rather than on grants. In the bidding process, institutions submit price bids at which they are willing to teach students. Institutions therefore have to compete for fumds on the basis of price. The funding bodies adjust the prices for quality, on the basis of external assessments. The bidding system is slightly different in the Polytechnics than in the universities. The intention in the University sectors was that institutions would submit prices for all their students. Thus, each year, there was no cerainty as to the minimum funding they would receive. This bidding process collapsed in the first year. 4t. The experience with the polytenics and colleges was more successfil. The PCFC guaranteed 95% of the previous year's budget to instituons, and required them to bid for the remaining 5% of overall resources. The PCFC awarded funds to those insdtutions with the lowest price bids. Institutions that could not reduce their costs had their funding cut back. 71 ANNEx 2. L1ST OF CONSULTANT REPORWT TITLE AUTHOR(S) Administrlo de Unirsidades Pdblicas: A Jacqu Schw Racionalidade da Ineficincia A Produtividade Interne das Insttuioe de Ensino Je-Jacques Paul Superior no Brasil Aspectos do Per61 do Candidato/nfcidiio do Haebrt Guaxini CilJau Programa de Cr6dito Educativo no Brasil Crdito Educativo no Brasui 0 Programa do }erbt Guaini Calhau MEC/Caixa Econ6mica Federal EvoluGlo e Imtitucionaliza35o do Ensino Superior Edson de Ollveta Nunes Privado no Brasik 1968-1990 Lasu Dant. Violets Mar Moner o Futuro da Educa;o Suprior no Brai Simon Sc O Problema Administrativo das Univeidades Pedro Lincoln Matts Federab Brasileiras: Questes-Cha o Processo Decisdrio e a Problemica do Teodoo Rogido Vold Gerenciamento das Unieides Feders Brasieiras Unrsidades Pdblicas e Pvadas: Sintese Teodoro Rog8h VaM Comparativs de Algus Ixcador de Universidades Comunita UU. Federais 11 AUl daed 19 and writm in (unless ( noed otlnw). 21 Alx availble in Eg 72 ANNEX 3. STSTISUCAL TABLES Table 1I DRAZL HIflOi EDUCATION Hihw Eduotm PazdmAZ bY . Laatltaadma TYP ad SCOW - I- 6.19 195. 1939 960 % 1975 % 1985 1989 % Undegpdue Edu.ii 9320 1,072,548 1,367.609 1.518.9W (97%) UnD_ua 59=,225 ( ) 465.930 (43%) 6719.77 816,024 (511) PublIc 298,934 (64%) 433,957 459.335 (59%) Privte 16,996 (161) 238,0 356,689 (41%) Noe-uulwaly HEs 33.977 (36%) 606,618 (7) 695.632 702.880 (49%) Fublic l.291 (18%) 12.723 125,079 (18%) PAve 495,327 (2%) M909 57.801 (8210 A HEXa Publi 51.915 (56%) 410.5 (3S1) 5.610 54.414 (39S1) Priv4st 41,287 (44%) 66323 (62%) 810,92 934.490 (611) Grduat Educaion (a) 37,726 51.956 (31%) Univeali 34.28 47,212 (91%) Public 29,327 40,389 86) Private 4,954 6.823 141%) Nowvn*ray HEl. 3,445 4.744 ,9%) Pubdd 2,019 2.781 f59%) Privoo 1,426 1.964 i41%) AUl HMiE PublP 31,346 43.170 .134) PItYSt. 6,380 8,6786 (1%) GCad TOa 93, 1.072.548 1.405,335 1.570,160 Pubac 58,026 627.514 6:% Prdvat 817,309 943.276 ,% Aal Ow PaU 10.51 03 01 EflZOll*L4as a Petmag. of Age Ceboat Ea,olmm as a Pao" o Smamias Mikd BwDZ 45% 58% Funk EnrUlla t% of totl Had Suma 1mlat (% o a) 5( N4& Clm AtOdsae O * (1 of ftb) 40% 351 Trmitin Rate fiam S.mdmy to lHa (1987) '' S 74% Namw Scume Is NEC. 19M ftgre far !!!duas odum6si' Mei aomgu 194 and 1975 fiam do = ind=ude emmUaus Tdd IL2 73 BAZML JUOIIER EDUCAAnON AKRUolMe by GoveminE Authori End Institiloal yn.190 9 1985, 198 1960 1975 loss 1989 % of to Fe&rwl 34,986 248,849 344,617 340,203 22 Universiio 236,59S 330,178 323,65 21 Non-aivvrities 12,2S4 14,439 16H 1 Stt 16,062 107,111 .60,067 211,946 13 Unesi"ti 58,196 117,692 154,386 10 Non-unveuiti 48,915 42,375 57,5 3 Midp 867 S4,265 83,342 75,434 S UniVstis 4,143 15,414 21,663 2 N0,122 67,928 53,771 3 PrIvate 41,287 662,323 817, 943,276 60 Univrsiis 166,96 242,974 363,512 23 Non-unverstie 495,327 S7435 57,70 37 AJp_o Tte 93,20 1,0724 1,405,335 1,570,80 10 Not: Fige 1960 ad 197S do not inclue graduat oimubot which ba leuh_2%. Sordu: NEC. 74 Stda Iasoa ad FW d sad Lod, 19 Fkld of Study Earoilbs spia of EnroU1meat !La__E!!! UnYndd_ _ UduBw Gtmduw No. S No. S No. S No. S _ 1 S.ill S _. 941 655 22,57 44% 155.006 68% 1.5s 331 16% 7% 1ealth 3dm.. 174.945 12% 6.7 13% 29,5 13% 6% 13% 17% 10% 1Ix sl_ 1S8,652 10 11,004 21 22,629 10 1,240 25as 14% its .gIas 14,016 10% 7.131 14 .60 61 725 15% 101 10S Agiculo 42.873 3S 4.152 83 5.131 2S 677 14% 121 i16 tOW 1,518.904 100I S1,96 1005 227,037 100I 4.89 1001 i5s 9% SOWW: NEC. awoduuoaftoiint i baud a atuby CAPS. 75 TeWs IA BRAZaL HIGHER EDUCATION Underr&dgW LorImet by Govemina At Jit ut 4 1960-1_9 Yf-to-Yr Yr-to-Yr Yr-t-Yr Yr-to-yr Yr-to-Yr You Federa Chan Sta Cta Prive ChCage Totl Change 1960 34.986 16.062 867 4t,287 93.202 1961 38.795 10.9 1S,497 -3.5 2,040 135.3 42,.5d 3.1 98,892 6.1 1962 43,275 1.7 107,299 8.3 1963 47,428 9.6 124.214 15.8 1964 54,721 1S.4 1413 14.6 1945 65,380 20,170 3,436 6,795 22.1 15,5781 9.4 1966 72.455 10.8 21.751 7.8 4.236 23.3 81,667 22.3 180.109 15.6 1967 91.60 12.2 212,882 18.2 1968 124.496 35.9 278.295 30.7 1969 157,826 26.8 342.86 23.2 1970 214,865 36.1 425.478 24.1 1971 309.134 43.9 561.397 31.9 WM72 409,971 32.6 688,382 22.6 1973 184.312 80.60 35,162 472.721 15.3 772.800 12.3 1974 205.573 11.5 90,618 12.4 4,837 27.S 56.565 26.2 933 21.3 1975 248,849 21.1 107.111 18.2 54.265 21.0 662.323 11.0 1.072.548 14.4 1976 249,935 OA 99,679 -6.9 54.771 0.9 691,806 4.5 1.096,191 2.2 1977 253,255 1.3 103,691 4.0 S2,186 -4.7 704,754 1.9 1,113,86 1.6 1978 288,011 13.7 105.750 2.0 58.592 12.3 773.204 9.7 1,225.557 10.0 1979 290.868 1.0 107,794 1.9 63,641 8.6 849,496 9.9 1,311,799 7.0 1980 316,715 6.9 109.252 1.4 66,265 4.1 885,054 4.2 1.377.86 5.0 1981 313,217 -1.1 129.659 18.7 92,934 40.2 850982 -3.8 1.36.M92 0 7 1982 316.940 1.2 134,901 4.0 96,547 3.9 859,599 1.0 1,407,987 I1S 1983 340.118 7.3 147,197 9.1 69,374 -7.4 862.303 0.3 1,438,992 2.2 19U4 326.199 -4.1 156,013 6.0 89,667 0.3 827,660 -4.0 1.399.539 -2.7 1985 326.522 0.1 146,816 -S.9 83,342 -7.1 810.929 -2.0 1,367,609 -2.3 1986 325,756 -0.2 153,789 4.7 98,109 17.7 840.5" 3.7 1,418.218 3.7 1987 329.427 1.1 168,039 9.3 87,503 -10.8 8850 5A 1.470,5S9 3.7 1988 317,83 -3.S 190,736 13.5 76,784 -12.2 918,209 3.7 1506 2.2 1989 315,283 -0.8 193.697 1.6 75,434 -1.8 934.490 1.8 1.518,904 1.0 Sena .E.. 76 BRAZIL: EIGHER EDUCATION 60-83 83-89 1960 1983 % 1989 % Chane Chng Fodal 34,986 38% 4,118 24% 315,283 21% 872% -7% Ste 16,062 17% 147,197 10% 193,697 13% 816% 32% Muniipstl 867 1% 89,374 6% 75,434 S% 10208% -16% pirivae 41,287 44% 862,303 60% 934,490 62% 1989% 8% tOl 93,202 100% 1,438,992 100% 118,904 100% 1444% 6% Sore: MEC. 77 Table 11.6 BRAZIL HIGHER EDUCATION Hither Educaton Evoutin betee 1283 and 198 Uafrosa Othw2 Ht AD H11 1963 1988 Cam 1983 19 Chiap 1m8 1 Chg -e No. of I _uam Fedetl 35 35 0% 25 t9 -24% 60 54 -10% S_. 10 15 501% 9 72 4% 79 87 10M MUncipAl 2 2 0% 112 90 -201 114 92 -19% Privd. 20 31 5S% 595 607 21 615 638 4% Totl 67 83 24% 601 788 -2S s8 871 0% Teching f Fedea 40,276 42.029 41 2.698 2.319 -71 42.974 44.5 4% SootB 12,631 17,713 40% 3,703 4,090 10% 16,334 21.803 33S Munidpl 1,063 819 -23% 4,433 3.290 -26% 5,496 4.109 -25* Pulfte 15,086 19,241 28% 33,889 35,72 6% 48,975 S5,023 12% TOt 69,056 79,802 16% 44,723 45.681 21 113,779 125.483 t0% Emh_net Fedea 328.044 304.465 -71 12,074 13.366 IIS 340,118 317,831 -7% St"e 98,371 129,785 32% 48,826 60,95l 25% 147,197 190,736 30% Municpa 17.213 17,178 0% 72,141 59.606 -17% 89,374 76.784 -14% Priva 244.232 318.812 311 618.071 599.397 -3S 82303 918,209 * Total 687.860 770.240 12% 751,132 733.320 -2% 1,438.992 1503,0 4% Gmiu_ F!ebrd 40588 3S.742 -12% 2.29S 2.284 0% 42,U3 38.026 - Stat 12.469 16.597 33% 9,380 8,442 -10% 21,849 25,039 15% Municipd 2.948 2.320 -21% 13,547 9,610 -291 16.495 11,930 -:8% PrIM 36.609 44.809 221 126,823 103,036 -19% 163,432 147.845 -;Cs Total 92,614 99,468 7% 152,0U5 123.372 -191 244,659 222,840 -44 SoMM 16C. I [[[[i !{g IL I~~~~~§ I ta 9 * 0 ' S i " "P 'w I 0 -. -I~~~~~~ - a , - XiS~Ff 79 Tale DLS BRAZIL: HGEMR EDUCATION Public Educatido Epeditu by Gowvuu and Sebool Lewds, 1985 (a) Am.mt (b) Sbu Fedeal 38.2 100% pFity 14.1 37% Secondary 2.8 7% Highr 14.6 38% O3her 6.8 1M% Stte 42.5 100% Priam 31.7 75% S-e3nday 5.5 13% Higbe 5.3 12% Muniipal 11.2 100% Prinmy 8.3 75% Seconday 1.5 13% iger 1.4 12% Total Govenme 91.9 1o0% Primmy 54.1 59% Secolda 9.8 11% Hig 21.3 23% (a) All fige re estmate (b) Figure in billionf 1985 cazdas. Source: World Bak (1988) 80 Table U.g IRAZL: OEMER EDUCATION Minitr of Zducitio fluse 12jL-89 in CMontn 1939 PiesW year NCz$ usS Highe (bL*=-n) -(bMiuon) Edu=caaio 1986 $14.3 55.0 50% 1987 $16.7 $5.9 67% 19oe $17.4 56.1 63% 1989 517.7 56.2 63% Source: USlA and MEC. 81 Tame a.10 RAZL HlGHEl EDUCATION FemaR meom of Fodera UJadish S w bt Reby ! n, 199 s_!e Data . ~ ~ ~ ~ ~ eewg Tow ~~~~~~L-M TOWs Pfemu hoe. NI_ Nofdma Soees so9h Camsom ll il SWo (Mwmni Slaflut patmouth) Up to 3 13 35 14 30 9 IS 620 4-m0 !2 - ?2 i Pt a7 11-20 moah 7 25 23 29 15 20 700 21-30 moth 5 23 28 27 17 I1 389 31-50 moa 7 17 30 28 is 8 284 ov EO moath 4 18 23 39 16 5 182 Tot P e_ 9 28 21 29 13 100 TOWal idet in Samp 304 968 728 991 466 3,4S7 3.457 One m_imm alay a USSSt amt of dEa¢ta Mle So4c: M. am (1990) 82 Table Mi-u Page ' of 4 BRAZIL: IHGitR EDUCATON TIrAL PUC.MO PUICPR UNDII FURB PRIATE PHYSICALP AW Toal Am of Fa (equa asau) 57,14 91.962 16,175 24,370 189,651 AM of Labt S _ _pm (ujat atta) S,971 7,000 1,617 3,337 17.925 Am of LuMa (equ nuna) 5,000 "3 1,02 2,1700 9,660 Al" oftChamom (_bm _ MOM) ,22 7,000 8,01s 6,814 30,091 Na _atClausrmu 153 86 59 ? 391 No of Sat 8,445 6,083 2,525 6,96 24,0 UIUVXSIY DWALTANWT _ 6vi"2 30 16 46 No. 1-eb Suvd 216,130 216,130 Uti C C4. pe Na (U) 50.97 RecIt pF t VW (VII t0.75 CaclhJwe: by Sodos $94,611 546766 5541,377 (US$) by ov. lSbsldy tOW 5121,78 S446,766 sSA44 Days of Opumao- p yr 259 300 Totl olldI 2t6,94 756 76,019 95000 463,534 rine Acquin d o decadeo of Ws 6,6 18,908 27,785 40,067 95,727 ScWle um It Puddiab 363 210 1,0 341 2,094 No. of Reqiata Uaaau 13.756 Im85 7,057 22.398 AD4MDWmAIWK STAFf TOW Staff 757 484 209 5 1.709 As" Saf 727 480 209 WS 1,674 S,aff wvowsy Uuc_m 97 128 SO SG 331 taffwl Suoaty Edu 190 102 7n 73 440 Saff WI Phmuy EduEoa a Un 470 250 84 130 934 Toal A udvew Cow 522253 27,421 575,514 8116,005 AdmWm_W Cm pw Slaff 5294 5571 536 CoUnm AND SPACES No.ofCouneOffmd 10 29 20 19 68 1g4mcotm 5 7 12 a 32 CcauuOffwdl ist& Sbib 10 3 2 4 19 No. ofliIIbo Ceoas. 365 555 35% 371 No.ofLOWCosCiO_ 64% 45% 6SS 63% iO. ofiusers c ama I 0 | No. of DOeoa Comm 0 0 Caus. AddS Daiqg 1987.199 0 2 3 7 12 Coma D d el 19D7-1939 0 0 0 Subjestha d 8U8 1,109 1.9-7 No. of Sci1aOfr.d 1,728 163 210 840 2,4 TUDEIFT BODY DIITE fTO No of uaogduaA9da 1293 8,42 3,784 6,148 31,276 No.of GiduatSldaM 10 361 107 120 S98 No. of kDey4M Smad_ 7,215 6.110 1551 ,600 16,535 No. of Nit*4ide Sodom 5,708 2,25 2,233 4.548 14,741 Nul in HiCmotnC 4,932 4,383 447 1,585 11,347 NO. in LOW Cm Cw 7,99 4.0M 3,337 4,51 19,929 SauiaeWkbLasa,nato10Ssdsm 43 1 22 is 129 Sco WhhM lMando40lwsn_ 8a" 10o 54 384 1,391 Potst, Of O&uti Sou. 84.005 7s.00S 83 Tal 11-1 Page 2 of 4 BRAZIL: HIGHER EDUCATION lTfAL TEACMINO SrAFI (198 No. fTbas 905 636 24 334 2,116 No. of AJWe Te 0? oil 206 32 1,953 No. ofTeeses md$q fe PhD. In 31 too 105 25 20 340 No. otTeoma madyhfr I&M., Ovm 4 6 10 2 22 NewTeshasRhd 232 76 30 487 T.AG Ta_IIM& 181 89 102 10 332 FuIby- PFA Tiau Em" 32 32 Feuky - a Tier 40 m is 96 100 25 306 Facuh_y- fuit Tho; 20 Rm 232 129 23 3S FePult- PoT rumMoody Su 393 89 265 1.3S Fcai In AMaeUv Pea 16 19 4 1 40 Fcut in Adminiwo Pam an Tea g 106 106 35 4 313 FamIl Tean Ody 677 381 124 224 1,406 Fmuy Teachmig a Doive Resmch 104 112 54 33 305 Nu_dN of Ron Teahg: 40 bomn Conu 14bn h.A. ".A. NA. NA. Numbe of omN Teahing 20 hoCaus 12 bh NA. tJA. NA. N#.A Teachg Exsadb" Ses,S" 542.10 MO.= 5218963 53427,010 Sibha Full oesr SdM s1,t35 $1,700 Sl.101 51.600 HUukAawat PsufaSdee $1.210 $1.134 S0 $1.070 FACULTY WITI GRADUATE SCIOOLIN Full 33 19 16 57 125 Aseaa 59 24 14 WNA 97 Asisag 49 35 a 0 97 T_AAhi AlaIa 0 S 1 4 10 FACULTY WrTI rOUt GRADUATE SCHOOLING Fuh Si 43 4 131 229 Aceociate 78 42 15 WA 135 Aaim 450 231 34 33 748 TeAINuAiti _It5 162 120 15 41 RESXAICH & PULICeATO OasoWReitPi 11 4S 9 19 Books&CbIuoOf Boob S 10 15 30 Aaticld n bahN loomjnue 36 0 16 2 54 A Xit In MIsP Jow._ 0 0 Thme 24 0 24 Oab! 57 10 15 7 s9 SALAAY OF FACJTV WIE ECCLW DEIATO Fu 51.599 1599 Aaociamclata S18 51.382 A _iai .38 51,218 Teachin A ina 51.071 51 SALARY OF 40 HOUR FACULTY Fun $1,7S5 $1.074 $1,99 $1,406 Aschae £1,574 5973 1,182 51,243 AulEag $1.405 s7 5105 51,080 TeinAmin $nl 1,210 587S 5893 84 T.WUU Page 3 of 4 DRAZIL- JUONE 3DUCA7KO? TOTAL PUC4MO FUM UNUI FM "VATS SALARY OF 2011MM1 FACULTY PA Ss"36 0 56 -~ 546 5591 5539 A-ai SW $519 "S9 S _Adog A#9om U6 549 536 SALARY OF NOUUY. FACULTY PA S7. 51.92 58.74 51.22 A _oda 57.18 S720 738 7. A-INM b6.14 $5.97 56.9 56.60 Tombigs A _ima S 53 54.4) 55.48 5A.0 TOTAL COMYST ARM OF SUw ( AM ~~~~~~~~~~~~~538.046 553,04 =Gk 'teS*g 53,737,0 5290,047 51,4534 1,1645,158 3j_jg 53,*44616 510,08 466 311.52s momw" & Sam lbd_a $7,186,21 5 4,790 53,1671 54,10057 TOTAL 514,437,974 52,9U1,911 ,0S54 522.46,719 WIkOLLN STY ASRA OP IIUDY AMU 124 libmeGt$*bg 3,735 237 1,3 i _asI 1,198 792 360 Rm _in a 1Sa .s1o. 8,00 2,73 4.455 TOTAL 12,93 3,69t 6,146 C06T MM ITUDUI NV ARMA OF STUDY AmS $710 5710 1.1...auGT.sbukg 51,0)4 $I 5124 51,094 5523 swam"a 522 $1 57312 81,297 1,2 Nunis_&.1ISIu 5696| S1 5714 8709 55 _h=mewn & Samlmms St $14 "a 8 31 AVtAGI S1401.41 1.00 S345.S S1 A1 60.22 cOm PU Mom By co13OF3 STY Edo"= $1,201 51,201.6 Dsdmy 52743 $132 52,127.2 ,lascm.is 5.01 7S7 560.75 Law $1,019 S30 $720 VW 59 "l_8 1,25I 5"31 S5040.95 .sy N4 5717 5810o $716.94 AUW 69 5169.01 &#iI. 5$1,151 51,753.00 5w11 A:AM S l ,404 $971 smS7137 $985.76 85 Page 4 of 4 Table ln-it BRAL: HIGHR EDUTCAIION Summa of Came Stdio Of Two on~rm1ad NWo in MM Wmv-ae Pia unibmwole TerAL PUC4do PUC4PR UNDIJ UREa PRIVATE ACCESS (VI ENTRMNCB UAMR4ATION No. of Opaiqe 1.650 2.570 1.008 2,040 7268 No.of Candidas 12469 9.626 1.957 S,470 293=2 Rdo of Codidwaespsrop.a112 7.44 3.7S 1.94 2.68 4.0 Opekg. in REio cot Cmwsu. 530 1,260 209 620 2669 Candidetaa SWt Hih cot Coui S.421 5,649 465 1,833 13368 bami.S CadlaawpJu brig Cog Cm.. 9.'5 4.46 2 2.96 5.0 OP.Sag. in LwCoe Cowin 1,070 1,310 7I9 1,410 4599 C40didn (W OW CAMcouna6,848 3,977 1,492 2,6.) 1595 Ro" fw LOW Coe COW 6.40 3.06 1.87 2.36 3.3 No. of UnOW oponkp0 0 IS0 119 299 NUlMM OF OPENINS sy coURtSg Edumsatla 65 OS Dsatigay 68 140 ?38 Eoaogi. £00 200 300 lAW 120 210 165 200 695 Nuiniog GO 70 44 174 Pedagg 430 66 160 49 AgrSOwy 55 5 Medicine 70 70 Idadasingl 70 70 NUMR OF CANDIDATES Y COtRSE Educatla 86 86 Demiiwy 2,203 1157 3360 Ectoom" 586 325 911 LAW 1,725 2188 781 124 5941 N=In 26 662 43 1069 Pedaoy 128 59 167 3S4 MeidIcn 1522 1322 M _daie II II Biolog 44 44 RATY OF CANDIDATES BY COURS8 Edusailos 1.32 1.32 Deasy 32.40 82 M1S1 Ecnosuines 5.86 1.63 3S04 Law 14.3 £0.42 4.7) 6.24 1.5 Numn* 6.07 9.46 0.93 6.14 PedWag .91 0.89 1.04 0.97 AnMM 1.53 3.55 MedI 21.74 21.74 Maihmts 1.9 1.J9 Sabu3 0.88 0.88 Soma: VabIl (1990) 86 rage 1 of 3 __ Iv** B RU III IIpS f tEigj ___ __- - st 0 #! ,gg a§ufi§R RfS { I ~- S 9Rf J l l |£i!s# _w_ I" R^ * 8| giftSglFl tgi { |~~~~~ .~ ** b=mt II" - -~ -- l i £ a~~~~~~~~~~~~~~a 5j j -- -- ~s" {' {§u§ a aauaa sj$ e a ~ ~~~~~~ a L s # lR !R 1lc|Zt g ~~~~~ ~~~ ~~19" S t*!la rsjl 0 0 __ - - § ig $fig|- tOle^ onX _ - " 2t$t lg t§^ tigXt Table 11-42 BRAZII IUGCHER EDUCATION Private Non-University hilher Educallon In the Slate of Rio de Janeiro, Case Studies Summamry CIEAC SEPNL SECI RXG SJMA FUSUE FAA SoI FE IUD AM AMO AWA cm Aw,pI OTALIEACHS 3 25 too US, 431 n 34 S41 n 246 246 3 211 3 WACHaW EoRDD EOs 5 2 n 31 9 4 20 111 0 IU$ 61 04 4 31 Su IEACHESW r ttl 24 13 84 32 27VI 29 449 I4 7 S U12 2 US14 PARTflW TnU 13s la3 de 7i 57 6 5 Ft)UUl= lACEMS is 4* SO 5 tot 4 2 HOUR-BAS53A035 3o 20 us 211 460 3 13 244 136 63 37 SlEACER^S W EREBE US 9 4 V% 27% 23% Is3 0% us 12% as 35% Xs 23 145 2S IS FiLUbIM e% 5% 0e 0% 0% 3% a% o1 As Is 3o Is TEACHERS FM QOUWME a 0 573 aS? a 35 0S la in 0t o44 U71 snJDSIWIAC 35 46 24 1t UI 33 34 33 23 a8 2S 2 tS AE (O MET4U 30 3.16 1" 3.4S 'AM 50 3 1at* 3.56 13*415 14.144 10.06 WMOF LPaSOO0 2) a6 4a a 3 44 I" 2S 90 a0 S 14 (*3A1IS 13 a5 3.% tl3 3.50 244 44 1.4m 0 410 5.2 32 AVG CAPACIITO OOaIS Ss a 4 nS US a to s 1 to 00 a OOS3810 19M 30210 ,.052 31.40 32244 52.723 .64 uSS 11.6$ 336 5311 63.300 22.48 9OOKSLwz5 4SINCE l 7I5 .40 5.0J7 VER0CAAS 25t 31 278 217 7K 3.46 634 2UN 29 2465 I3 SSl 1 wiSaURSCJ1ON 3 3 22 34 5 22 3437 iS 0, STAF 46 3In 12 16 so is 43 44 as 1 332 910 SltUWMSTAFP S2 44 n 34 2 22 70 A 26 34 eXED(RumrE WvOOEfCAUOOSY (a) TEAOHRSSAlARAY 74.30 33.036 2,05 Lao 39X4 3463 2to" n 3 4 Om UN MM SAPwSllRY 240 on 47S 50* 813 4m 00 " 6 11 m 2 33" 4S21S smxn $3. is 06 ff3 tlS 239 tl0 e4s .7 3450 lQ!_4 . t0 i I S3 1 i2 23s 1.10S 21S too 2 "tl 000MABILM on 32n 75 a3 32 3.4 140 00 04 0 to 6 3474 IEAC MAERLUIS 144 0 St33 a of 5n 1a 32 U ff .0 cosuno. * aEPAIR 454 1 4 1o a US 4.039 MAII4AS! 36 f 4 221 St s o. m nD OthER0 .3 t2 Om 2 2 I3S TOAL 34.0 AM 3 Im1 mm 2363 3.0) 7.47 3t, n I3,3 1116 244.20t S TEACH SALARI us 65S %S as 21 24% 4% 4 2&% 40% 3" 14% S SIAff SALARY 15% 23 13% % us 30 us 40 us "a 21% 32% us 5 PERSOOO 71 S "S I S f ts 54% as J"S as^ §*S nz PX44 cosT PER SniENr 64.7 too4 4.6 17S 465 s t 5 2t4o 3.60 34 .643 up 3W COSTI IN 133 b) t25 Loa 210 1 3.67 on . o5f t- "S i.0 t(fsnlu3 1 s) 3M 3.2 Lou 944 0a4 I'm 1.3044 so L,43 34 2.302 Table 11-12 BRAZILb HIG1ER EDUCATION Private NonUnlversity ligher Eduction I. the State of Rio de Janero, Case Studies Snumary cRWC son SW FOG SUA FtREM FAA SW FMRP m ASO AMA CM Aepfe EXfM4M=f FSCALAC1IWT (i) nAct of 16 $ on 374 Ant 7"a UIXa&1 ) 40 U 271 7 *o 1% 2 ti 4 IIlA1 1 S *~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~s 20 IUS 40 7 ARMEMtIM U15 %M41O. AVIIALIO 78 03 12 3313 IS 17 97 RUAUIMT :9a so s SEtS~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~m ^ m In a oa 4n U 1to4 to s w auRVC *3 35 23 *24141 4 TOAL 129.0 X 71 too 0 ll 34 78 M3 W 1 soceaPFU () iumow * FEeS 310 3mm too 134)9 X34 3171 33.2 70S 2n4 STUDStLOAI6 4AU I.s 5* 744 1. 31? U71 t,l7 1.703 GO alO_N a 3 7 4.S S on a UA- " 0 .438 SW 4S 0 SERlVICE FROVISOS m1 4833 38 3.37 1.788 O IHER 1.05 3*3n 0 3.38 17. 8 7TAL t9 21141 X 225 9.3 Mm A1" 7.21 A411 2W 197 1?.n S 111039% 41 7 2%a us%21 am1 No "S no sm TM Is 6nc * onu 1 me, as as us 4* am *1 3% 01% on as to) U.aIha .i,mmm 6pm (b) -1*4 OWMOW- Box- (e)UMe"_ (4)S== CMA I (ae )9m uZoSff3Me 89 Tab6l ERAZIL: HIGHER EDUCATION -unit Ex iUre in ubli HiE Edation. Sectd Coimre Expedie GNP per Unit Expenditurn AS Coutry Yer pr Studet Capita % of GNP per caita lapan 198S $S,968 $11,300 53% UK Univerdies 1986 $12,950 $8,870 146% Polytecnics 1986 $6,160 $8,870 69% USA 1984 $8,724 $16,690 52% Philippines 198. $3,492 $590 592% Spain 198S $906 $4,290 21.1% Austaia 1987 $6,126 $11,100 s5.2% Brazil Using Official Excange Rate 1988 $7,930 $2,160 367% Using Parallel Exchange Ra 1988 $4,760 $2,160 220% Soces: OECD 1990; phip:as: Jam, 1988; Brazil: Pu & Wolyneck (1989) and Jacque Schwartmn (1990). Table 111.2, Past A BRAZIL: HWGHER EDUCATION Stlut Teewm Ratlos in Fede Uneste, I98 STUDENTfrEACHER FACULTY STUDENTS RATIO PART FULL TOTAL HEAD- NAME OF INSTITUTION TIME TIME FACULTY F.T.E. (a) UNDERtORAD GRAD COUNT F.TE.L I UnbezselicdedBsulia 200 769 969 869 a.4.7 51t 9 t0 2 UalversldaddoAmnoms 125 S88 713 651 7.306 10 I1 3 Unlealdaedo RIode aelro 311 246 557 402 2,191 19 4 6 4 UnIvcrsldalodeRlo Gand. 49 404 453 429 4.071 24 9 t0 S Umivesaldade FHr de ll 608 1.401 2,009 1,705 16,091 511 a 10 6 Ummsd de Fe&W lda sxrU 350 2,388 2.738 2,563 17,758 8 7 7 7 Universlads FedealdeAlagoes 242 653 895 774 6,S36 7 8 8 UniverdtadeFaderg deGolas 357 965 1,322 1,144 7,262 t37 6 6 9 Universlads Fedeuu de Jb de Fan 864 578 742 660 6,126 20 a 9 10 Univerudae Fderl de Min" heral 629 1.898 2,527 2.213 15,925 1,659 7 8 11 Udivesslacde Ferl de Oo Preto 49 238 287 263 2,162 29 a 8 12 Ulwus&dde Federa de Pdots 122 626 748 687 5.047 161 7 a 13 Uuiveraldae Fed4a de Penamboco 542 1,3S2 1,894 1,623 12.938 899 7 9 14 UIJverhldadee F_d detaMdaia 55 53 108 81 1.332 12 17 IS UalvWadFedXeradeSam Cmarl 649 1,267 1,916 1,592 12,023 694 7 a C 16 Univeraldf. FederaIde Sam Mwis 77 1,083 1,160 1,122 8,764 249 8 S It UnW &d dPeda1lde8aoCako I 1 409 420 41S 2.438 316 7 7 18 UalvmakladcFederaldSaSelpo 142 340 482 411 5,S23 10 I1 13 19 Univerldae Fedea do Ubedandls SI 833 884 859 5,903 34 7 7 20 UaiverslddFedsadoVia 12 600 612 606 4.214 619 8 8 21 UnlVda Fodel deAer 22 231 253 242 1,988 8 1 22 UnIveriald.Fede&*ldoCam 232 1.115 1.347 1,231 13,261 S36 10 11 23 UniversldadeoFderu)do splEorkmnta 174 884 1,058 971 9,043 Ss 9 9 24 UnIverildadsFe, eudl de arambso 173 587 760 674 6.443 8 10 25 UnIvenaile Fedado b Mat Gms. 104 929 1.093 1,041 6277 6 6 26 UFvesdmdsFederado Ma.OGromodo SW 169 466 635 SSI 4,912 8 9 27 Umlwuldsd.FederaldePa 255 1.017 1,3I2 1,160 14,092 166 11 12 28 Untveldsts, Fodl d 1Pam 86 38 124 1I 1.014 784 IS 22 29 Univa"tdadsFdetado lPs 525 512 1.037 775 8,766 8 I1 30 UoJavevlFaded adRio deianelo 770 2,106 3,S76 3,191 22.769 5,03S 8 9 31 U daverale F.deb do Rio Grand. do Nost 191 1,605 1.796 1,701 15,372 202 9 9 32 UalveesladeFadel doRlo Garde do Sul 869 1.325 2,194 1,760 16,368 1,379 8 t0 33 UdivenkdasFederal Flublen 72S 1,9SS 2,680 2,318 11,709 786 5 5 34 tUivenrldde Fealal Ral de Penumbuco g0 333 434 384 4,186 130 10 II 3. Univeridade Federal Rurl do Rlo de Janeo 61 416 477 447 3,412 253 8 8 Toal 9.232 30.970 * 40.202 35.586 291,644 16,079 a 9 (a) Asuming one full-time appointment equals t-o pert-time ones. Source: MEC's Sino"e Esdtlsticn do Enalno Superior 1988 Tale MH.2, P RAZWIL HIGHER EDUCATION S_admt Tm6w Fades in Sai ad Mu pal Univarsdtil, 19J3 FACULTY STUDENTS STUDENTITEACHER RATIO PART FULL TOTAL NAME OP INSTfTUTION TIME TIME FACULTY F.T.E. (a) UNDERGRAD GRAD HEADCOUNT F.T.E. STATE I Unlveuldad0doEatedualdaflbba 130 192 322 257 3.534 1 14 2 Unhacamdae do Eds do Rlode lmndra 1.338 891 2,229 1.S60 15,6St 303 7 10 3 Ualweld Eftadual Pullta Juli Mesqut FPUbs 282 2.395 2,67? 2.536 16.051 6 6 4 Un adadsEiadual d2 ralbe 78 419 497 458 8.232 17 Is S UAlla Eaitdua de Ceanpla 366 1.609 1,975 1,792 6.63S 2.981 3 4 6 U al".i_adualedFe & deSantana 54 257 311 284 1.332 11 12 7 UnivaeI Fdoduad o LdoLada 354 797 1,151 974 10,053 9 10 S UnlvridaEdoiude Mutbap 193 670 863 767 8.119 9 g 9 UnIvaiabAs Eatadul do Pena Gram 286 199 485 342 6.096 13 12 10 UnlveuaddEadualdloC ra 418 S17 935 726 10,24S 1 14 I IU1vwaddRc 1nM1 do Ca0l 83 0 53 42 2,518 31 62 12 U.ivaeuId.d EnAldeMosabso 191 220 411 316 2,230 S 7 13 Ualvedad_E*dado Sd *ga deBl 269 0 269 135 1,S23 6 It 14 U Qarsibado Flom D_w do BEt de Sas CutlA 206 228 434 331 3,414 a to IS UVldakad l Saooaalo 1,641 3.359 5,000 4.180 32,082 6 8 Sutotal 5.06 11,753 17.559 14.656 127,197 3.214 I 9 MUNICIPAL .I UlbvddodoTde batu 525 6 S31 269 1,S39 22 43 2 Uuvtd.da Regwu de nimuan 240 48 283 161 S.09 20 34 Subta 765 54 J19 437 17,178 21 39 (a) AatiinS oft IUH-Ur appomn equas two ped-tim oca. Somtce: MEC' Siapsn Eata do Enalno, Sulkw 1988. Tblo 111.2 Pad C BRAZIL: HIGHER EDUCATION Sthded Teacbu Rati. In Pdate Ugnmtl 198 STUDENITACHER FACULTY STUDENTS RATIO PART FULL TOTAL HEAD- NAME OF INSUIUTION TIME TIM FACULTY F.T.E. (a) UNDERGRAD GRAD COUNT F.T.E. CATHOLIC I PouduilacUulvwadadeCaollcade Campa 1,186 131 1,317 724 19.307 too IS 27 2 Po.tlflclUal_CdadeCntlklcdaMlacralb 720 93 813 453 13.380 16 30 3 PMi'cUaWsaldabCoca doPlme 574 42 616 329 8.549 14 26 4 PloUc UnvegIedabCatAcade SaoPw)o 1.058 433 t,491 962 15.103 2.594 10 16 5 Peaia UnIeldad Cdolce do Riode J_ando 367 470 837 654 6.305 1.29 8 10 6 lftc_UCavusad.CatbdbaRioGrandSuI 1.649 129 1,778 954 22,461 467 13 24 7 Uov.bdadC.tdcadeP*mplls 232 41 273 157 3,401 12 22 8 Uulwabdd ohca de Savda 61S 0 615 308 I.19 19 38 9 UnIvaddada ica do GAa 397 237 634 436 12.347 19 28 10 Uniadade Calaicad Pdote 312 28 340 It4 4,743 l4 26 11 Unhwaldodd.CMoIcadePuaabaco 522 31 553 292 14.542 26 50 12 UnVedadahde CitAeIk de Saao 39S 11 406 209 5.232 13 25 Subtal 2.4t0 266 2.746 1.506 41,236 ISO IS 27 OTHER RELIIOUS I Unbalvedo do SaVado Coceo 113 36 149 93 2,906 20 31 2 Ualvweldods LutanadeflStd 188 43 231 137 3.019 13 22 3 U Jal"d tbmodtitadoPlclca 309 88 397 243 8.212 144 21 34 S 4 Ual_vkd de Seo Frmbenho 607 25 632 329 9,316 1S 28 Sutota 1.217 192 1.409 801 23,453 144 17 29 OTIER I Univerat Benz Cuba. 365 66 431 249 10.970 25 44 2 Uaivweldd de QuCwado Sol 499 43 542 293 3,390 1S 29 3 UalvesaldadadeFortalka 551 107 658 383 12.197 19 32 4 Valverldads do Gua-&m 284 Is 301 160 6.790 22 42 5 Uaddabdde IN 98 102 200 151 4.m 22 29 . 6 UdIvadddds,de Maugh 191 20 211 116 6.987 33 O0 7 UalveiWdd ds Md du Cnm 833 0 833 417 16.238 20 39 8 Unbeudadede rP Faa o 444 59 503 281 7.024 14 25 9 UIavenlbde,deRlb lRaPwredo 153 128 231 205 6.349 23 31 10 unuiwssdade do oe hullbt 303 S0 353 202 7,645 22 38 I UalvcnldadedoValeodoRiodosSlnow 690 67 7S7 412 21.218 28 52 12 UalvealddeGem Fllho 1.324 40 1,364 702 20,396 271 IS 29 13 Uftivefidade macKenzi 669 0 669 335 12.068 18 36 34 Ualverkiade Soma Caediha due Bancwes 2n ItS 462 324 8,223 1a 25 tS tJas,g,s,der Sen& tI,sula Sa80 I1 591 303 9.357 16 31 Sut.i.siaI 7 2t1 899 8.159 43.29 IS8.280 271 19 IS T.,taJ lot Piovaic Un,ve,gt,c_ 11.?J1 2.046 15,7?8 8.912 270.891 2.775 17 30 (a) Assumtng that one full-Itbc oppowimntnt cqucl tIo part-timenc Oe. Tabe 111.3 BRAZIL: HWIER EDUCATION Fedeal UIverste, Expeadftu per Student In 1988 (1aUS$) TOtWl Undergradut Got TOW E'weditr Expendir lmea Iuardlment EnraUmet per Studla Uiveruldade do BmSSiS 70,444.463 8,417 SIt 8,928 7,90 Univeruidede do AXma 41,784,341 7.306 7,306 5.719 Univeralvdde do Rio deo Jlmeo 28,094.234 2,191 19 2,210 12.712 Univenklde do Rio Grands 25.099,608 4,071 24 4.09S 6.129 UWvm*d* ederal do Bahi 130,827,169 16.091 Sig 16.602 7.80 Univasidade Federd do Parate 152.609.612 17,7S8 S58 1S,616 1.198 Univenaldde Federa do Alagoas 42.079,779 6,536 6.536 6.438 Umverndde Federal do Goias 65,23S.123 7,262 137 7.399 8,317 Universiddoe Federal de Jut do Form 36,776.467 6,126 20 6.146 5.984 Univeruadode Fekde do Mln Gemts 161,406.240 15.925 1,659 17,584 9.179 Uoahrudadc Fedeal de Ouro Ptdo 19,10S,13S 2.162 29 2,191 J.720 Univerldmde; Federal do Pdtos 43,172,353 S,047 161 5,208 8.290 Univeraidade Federa do PemanbuDco 122,544,094 12,938 899 13.837 8.856 Univeudade Federd do Rondonia 5,276,987 1.332 1,332 3,962 Univesadade Federal do Sant Cwile 99,859.127 12,023 694 12,717 7.8S2 Univerakfade Federa do Santa Moso 76,192,453 8.764 249 9.013 S.454 Uvnerdd Federal deo Sao CaI 24.811,908 2.438 316 2.754 9,009 Unmeldaded Fedal d Sergipe 24.S22.SSI S.523 10 S.533 4,432 Umiveawdad Federd do Ubeldadia 62,317,236 5,908 34 5.942 10.488 Ud"nvegude Federa de VIa.. 65.704.517 4.214 619 4,333 13,DS9 Uan)sidsd Feder 4d Acre 16,574.646 1.9U 1,988 8,337 Univesidade Federa do CeAra 101,239,931 13,261 536 13,797 7.338 U.lveraidade Federal do Edrito Swo 60.761.59 9,043 so 9,101 6,676 Unlv"Made, Fedal do MDard 39.527.736 6.443 6.443 6,135 Unlvatlmda Feerd do Mat. Ga 55,007l96 6,277 6.277 8,763 Ueaudda& Federal do Asdt. Ger do Sd 39.499.1U3 4.912 4,912 8,041 Unlvaldad Federal do pa 77,24.564 14.092 166 14,258 5.417 Umlvead Fedeal do Par n/a n/a n/a n/a UJnlalde Fedea do Pulm 57,071,544 8,766 5,766 6%I Univea Fedeal" do Rio deJanelso 249,090.119 22,769 S,03S 27,804 8.959 Uniesade Federl do Rio bands do None 106,091.926 15,372 202 15,574 6.812 Univmiade Fedea do Rio Graed do Sul IS2,086,308 16,368 1,379 17,747 8.S70 UnvsalOd Fedal Fluminm 139,633,147 11,709 786 12,495 11.17S U omelda Federa Rrd do Pnombuco 33,057.555 4,186 130 4.316 7.6S9 Ueadae& Federal Rurd do Rio de Janeiro 38,136,249 3,412 253 3,665 10.406 Sou,ce. MLC Therc ae modifictrec4a wh Paul (:990) because of different echange rae and enrolmen figurce sed. Table 111-4 URAILt HIQR EDUCATION Sumory of Came Stdee oef Foderal Litelver.tloe TOTAL UFII UFSC FEDERAL PRIVATE PHSICAL PLANT Total Art of F..illt1.. (auaa astwr) 028,50 *2,919 851,66" 19,651 Area of Labratory Spac (equate mete) 61.6 18,56 78,101 17,92S Area of Library (sqare eate)rs 12,146 6,540 17,647 9,66 Are of Clseereem (wivsr, motore) 6",981 81,60 9,481 86,69 Humor ot Cloeu 99" 11 1,812 191 ftier of Chairs 27,16 19,4i8 47,406 24,84 O4IVERSITV RESMXtT Eployeo IS 08 2c9 46 me. oals Sorved 1,157,2W0 16,467 1,867,747 210,186 Unit Coat per Meel (US$) 1.14 86.98 Receipts pt Meal (US) 0.6 .27 ContributlJos: by Students 3l,5" 8129,241 3411,997 1U1,377 (US#) by Gov. Subsidy 6281,78 6712,249 8944,930 so Total 1628,641 1832,487 11,456,611 156,644 Days of Op eatlea per yer gm 216 LIRARIES Tota1 edidags 660,5 107,016 690,622 463,584 Title. Acquire In the doce of Of's 17,167 U6,m 67,16t N,727 Pterodical. Cerretly Swbaeribed 4,011 2,179 6,361 2,4 No. ot Raltered Users 27,764 19,102 47,176 22.s8 ADMINISIRATIVE STAFF TOTAL STAFF 4,647 8,1906 7643 1,709 ACTIVE STAFF 4,186 1,176 7,356 1,674 STAFF IJ UNIVERSITY ECATIONI l 766 1,1i29 al1 STAFF W/ SECODARY EDUCATION 1,208 668 a,9 440 STAFF 8/ PRIOI DUCTION On LoSS sea 1,747 2,846 994 TOTAL ADMINISTRATIVE COSTS 35,202,5o 18,985022 ADUINISIUATIVE COST PU STAFF 81,189 s. COUSES AS SPACES ND. OF GRADUATE CURSES OFFERD 84 84 a DAYTIME COURSE$ *4 84 p HIIT-TIME COUSES 6 I 1 12 CURSOS OFF11110, IN 0 SHIFTS 8 7 I1 1S NO. OF "ION COST COURSES si 66S5 NO. OF LOW COST COURSES 433 44X NO. OF MASTER'S COUISES 46 21 1 1 t 940. OF (OCIRAL COURSES 14 7 21 * sinS(S ^0{ 001w,s tw rt9t 9 e.& o aIas 2 COLJAQS AMES1 NDkIC 1-1969 9 6 12 o &OAMt) 051D 1.3 1,938 1 218 1,921 Pau to 'ius (011HI 2.101 0.46 0,168 2.041 Table 111-4 DRAZILt Hhl& EUCAT10N S.wry of Ca. Stdies of Federel Uninareltlee TOTAL URIC WFSC FE0RAL PRItATE sIWoEN BODY PISTRJUION NO. OF h31DECRADUATE S7tETS 16,921 11,466 27,437 81,270 NO. OF GRAWATE STUENTS 1,786 1,1,1 2,310 69 NO. OF DAY-TI STWENTS 15,144 9,427 24,6n1 16.135 NO. OF NIIHT-TIME STDENTS 677 2,889 2,916 14,741 NO. IN WON COST CURSE 0,753 S ,on 14,061 11,347 moo. It LOW COST COWSES 7,269 6."5 12."6 i9,o, SESSIONS WITH LESS TEN 1o STUDENTS 644 1,478 2,117 129 SESSIONS WITH MM THiN FORTY STUDETS 573 ass 9 1,91 PEJCENTAGE Of GRADUATING STUDENTS ".30 ?B.86x TEACHING STAFF (19.9) no. of Teachar 8,029 1,780 4,006 2,116 lb. of Active Teacher 2,711 1,612 4,323 1,953 lN. of Techer. stdinlg for Ph.D. In Brasil 198 140 248 840 No. of Techer. studylin for Ph.D., Over.... 124 75 10o 22 N.. Teachoer Hied 128 1 143 407 Teachera Torminated 124 186 392 85 Faculty Full Tim.; Exclusive 2,183 1,876 8,511 32 Faculty Fell Ties; 40 Hors 321 120 449 B Faculty Part lTiel 23 Hours 674 274 643 86 Faculty - Fert lier; nrly 0 1,8J6 Faculty In Ailulstrativo to." . 6 4 Faculty sla Admtitralve Post. and Tleaing 6C 81 Faculty Teachlin Dnly 572 572 1,4 Faculty Te"hlog *ad Performing es etrch 2,816 2,815 s humbeo of HIurs Teach las 49 hour. Contract 08/14 br. Oubet f Hours Toeahinag 23 hoors Contrfct 80/14 bhr Teaing Expenditures 60,416,2" 84,629,8 $11,0u, 81,627,6w hIgbht Fuell Prfoteser Satlry 82,95 32,498 Nl_het AselUtat Pretofsr Salary 81,801 81,94 FACULTY WITH AUtATE SCHOOLING Full 18 145 218 126 £sioclots, 1,m 716 1,76 9 Assistant 886 289 Sig 97 t Te"chln Assisant 81 ? t FACLNTY WITHWT GRADUATE SCHOOLIN Ful I Is at 40 220 A::ocist. 416 466 o0 135 Assisteat 731 n~~~~~~~~~5 766 143 Tesching AealstMi 824 124 443 462 Table IIL-4 9RAZILs NIR EDtUCATION Sumnry of Cam Studios of Fedaral Universitle TOTAL UFUG UFSC FEDEtAL PRIVATE RESEARCH & PULICATIONS ONCOINO RESEARCH PROJEC'S 2640 597 ,141 U4 OOOKS A CIAPTERS OF BOOKS 22 22 39 MTICLES IN NATIONAL JOMALS S61 359 661 64 ARTICLES IN FRIO JOtNALS 101 7L61 6 nhSES 240 2I4 24 OTHERS 902 sea 1,486 69 SALARY OF FACULTY WITH EXCLUMIV DEDICATION Full "2,66S? 2,0310 12,04 $1,6" Asaolte 11,791 81,439 81,016 11,382 Aus1ateat 81,298 81,979 11,162 81,286 Tacthiog Assistat 1084 70e 1812 11,877 SALARY OF 40 NM FACULTY Full 11,m8 S1,47S 61,610 11,408 Aseoctsto 11,194 81,096 21,146 61,248 Assistant 8612 811o 19701 811,5 Teachlng A.lsta.t 8656 6649 1698 1893 SALARY OF 20 HMOW FACULTY Fel I 8961 8711 t871 1610 Associste 1597 1598 69F 8539 t s.st.at 1481 1437 1484 3469 lesching Aesistant 1320 8328 1827 86 SALARY OF HO3LY FACILTYl Full 18.22 Associate 17.45 Assistant M.4 Teaching Asslot 16.8 TOTAL COSTS BY AREA OF SlY 711) ARTS 07,t8,0S7 ST,192,9CT 888,4 SCIENCE A#ND TEOOOLOY 987,97,6,38 137,075,688 11,946,160 IlO-MICAL 812,68,619 162,636,619 81,68B,626 RJMANITIES A SOCIAL SCIENCES 841,8161,00 4111,7 14,100876 PRIMARy & SECONDARY EDUCATION 811,771,36? 811,771's TOTAL 110,291, 11N,2fr1,m 822,430,719 Table III-4 IRAZIL: H4Ot ECATION Sumir of Case Studieo of fedor. Umiveroltlee _______- ------------------__ _ -___ __ __ __ __ __ _ ----__--__ _ __- ____________________ TOTAL UFRC UFSC FEDERAL PRIVATE …-.-____,__ _ _____ _ __--______ _ ____ ______________----- -_____- _________------__ ___ -- __________ ENROLLMT SY AREA OF STWD ARTS 461 SCIENCE AND TECHNOLOGY 4,47 810-MEDICAL SS;bs HUMANITIES A SOCIAL SCIES 7,ON7 PRImARY s SECONDARY EDAIn 1,37 TOTAL 10,101 COST PER STUDENT 0BY M OF STUY ARTS 816,714 816,774 5716 SCIENCE AND TECHNOLG 80,281 sW1,6S 59,447 8833 B10-MEDICAL 810,"4 18t, 61i,966 31,231 HUMANITIES A SOCIAL SCIENCES 85,966 87,8 ",854 8681 PRIMARy a SECONARy EDlUWCATIo 68,602 380692 AVERAGE t398,88B.06 810,480.89 819.193.4 15879.22 COST PER STtNT ST COAXRSE OF S1UI1V EDUCATION 668690 80,997.64 11,201.86 GENTISIRY 811,114 816,0$ 813,069.76 32,127.25 ECONOMICS 96,402 06,491.07 3896.76 LAW 86,220 87,135-6 36,681.45 8729.38 NURSING 811,765 9lll565.5 11. U 85 PEDAGQVO9 81176 01,76a.90 f10.94 ACRONOY 5869.6i MEDICINE @11,844 811,844.90 31,768.09 AINEMATICS 87,2w 87,296.09 81,222.96 SIOLCOYV- AVERG: 67,606 311,071 89,877.07 6986.76 ACCESS (VIA ENTRACE EXAMATION) NO. OF OPENINGS 3,280 2,7a0 6ea6 7268 NO. OF CANDIDATES 27,108 10,711 43819 29322 RAItO OF CANDIDATES PER OPENING 8.a5 6.08 7. 4.6 OPENINGS IN HIGH COST COURSES 1,09 1,84S 3146 2660 CANIDATES FOR HIGH COST COURSES 16,741 8,186 24877 1368( RATIO OF CANIDATES/OPENINCS FOR NICH COST COURSES 8.76 6.70 7.9 6.1 OPENINGS IM LOW COST COURSES 1,408 1,406 2891 4696 CANIOAtJS FOR LOW COSt COLPSLS II,."1 1,676 18942 16964 0AI1O Of CADZOAI1SI/PININGZS FOR LO COST C("A$LS 1.66 6.80 6.6 8.6 w # ts u>LLLO OtAIzWS $I aS 7 t 299 Table 111-4 UAILs NIIMEI EuCATION S_inry of Ca* Studies of F.d.tal Uaiwvr.1tle TOTAL UF30 UFSC ERAL PRIVAIE NU1WR OF OPENINGS SY COUtSE EDUCATION 12t 129 a6 DENTISlRU 120 66 266 2no ECONOMICS o f6 I" LAW 26 146 436 696 NURSDIN 68 66 146 174 PEDAGO 166 anI 36 AGROOMY 66 MEDIeINE Ine U0 73 MATNEMATICS 6o 76 BIOLOGY 6 5 IBER OF CANIIDATES BY COUSE EDUCATION ass S3O e OETISTRY 2,382 1179 3661 336 ECONOMICS S28 623 s1t LAW 3,402 t661 363 6941 mESIMa 875 244 619 1669 PEDAGO 422 422 354 AGROOMY 196 MEOICINE 1716 1716 1s22 MATHEAnCS es Os II BIOLOGY 6 44 RATI OF CANDIES 9I COURSE EDUCATION 3.19 3.19 1.32 DENTIST 19.96 14.74 17.61 16.16 ECO CS 6.72 6.72 8.o4 LAW 18.09 13.44 13.21 9.65 .mmSIMQ 4.69 4.07 4.42 6.14 PEDAGOG 4.22 4.22 9.97 A _ONOM 8.6S MEDICINE 1T.16 17.19 21.74 ATHMATICS 1.42 1.42 1.69 BIOLOGY 6.66 Source: Veal (196)r 99 Table MLs BRAZL: HIGHER EDUCATION Compriso of Federal and Private Univerities Quality Efficency Measures and Cost, 1989 _ _ __ Federal (a) Pxivato (b) Efficiency Measure Square Mote per Student 28 6 Number of Chair per Stunt 1.6 0.8 Occupany Rate of Chais (%) 32 66 Yearly Subsidy per Stldent per Meal 31 0 Numbr of Libray Books per Studen 23 1S Rato f Stude to Administativo Staff 3.8 18.7 Percent of Student in Grduate Program 9 2 Percet of Studets in Day-Time Cous 81 51 Percont of Staents in High Cost Courses 48 36 Perc&tt of Class with less than 10 Students 7 0.4 Percent of Classe with mor than 40 sudens 3 4 Stdents per Teachers 7 16.3 Students per Teacher F.T.E. 6.9 30.1 PoercAt of Total Faculty Full Time 82 16 Percet of Total Faculty Part Time 18 19 Percent of Total Faculty Hired an Hourly Basis 0 65 Percent of Faculty with Graduate Deogres 55 16 Books and Chapters Written per Year as a Ratio per Facultp 18 0.03 Cost per Studont r' USS 8,51S 986 (a) Based on case studies, undertakn by Vahl (1990), of two federal universites: UFMG and UFSC (b) Based on case studies, undertaken by Vahl, of PUC-MG, PUC-PR, UNUUI, and FURS. Table V*1 BRAZIL: NIJNER EDUCATIOi 100 SUmdent Lgoa Scem. Artmdg th Wsrld Avnrage Percenit of Repoaytnt Administering Purpote of Loan Year students DatE Comitry (Loa Organination) Mechanism iswtitutien Suort Value Bgin with loom ye LATIN AlEIa,AN CARIBBEA Argentina CINCE) M ortgage Lon Autonomous Body Living Garbados CSRLF) Mortgage Loan AutonrAoua Body Tution and Living 511,000 1976 12U 19 Bolivia (CIOEP) Mortgage Loan Living Brazi (CEP) mortgage Loan Coerc1il Banks Tuition 5400 1974 2S5 1989 Chile Graated Universities Tuition 1981 19U8 Colombia CICETEX) Mortgage Loan Autonwmous Tod Tuition end Living 528 1953 6 1911S Costa Rica (COAE) Mortgage Loan Co_rcial ak*s Tuition and Llving 1977 1983 Dominican Repubtic fCE)Mortage Lea AutowA lady Living Ecuador (IRCE) Nortgage Loan AutonmuA Body Living 33 El Salvador CEducredito)ortgae" Loon Autonmos Body Honduras Cdxrodito) Mortgage Loa Autonomos "B Tuition -nd Living S2,700 1976 1S 1991 Ja ca (MLe) Mortoge Loan AutonAug B Tuition ad Llving 1405 197 20 195S Meico Nortgage Loan Co mmreislanks micaragu (Edacredito) Mortge Loa Auton s Body Panw (IFARNU) Mortgage Lon Autnonu god 19` 19 Peru (IMABC) Mortgage Low Autonomous " Trinidad CSRLF) Mortge" Loan Aut tms lody Tuition and Living 1972 Veneauel CEducredite) Mortgoa Loa Othtr Tuition an Living 00 1967 13 1991 CFGUA) Mortgage Low universities tuition aNd Living #2,200 1975 1X 1991 (BAMAP) Mortgage Lw' Comercial Banks Tuition ad Living $700 1S 1t1 ASIA China Mortgage Loan Universities Tuition Nd Living 1957 30S 1919 India Mortgage Loan Other Tuition and Living 585 1963 1t 1919 Irdonesia * mortgage Loan Universities and Tuition and Living s5S0 1982 33 1986 Cinrcial Banks Korea Mortgage Lon Comrcial Banks t1ition and Living 1975 Mataysia Mortgage Loan Cmewrcial Banks Living S1,300 195 Philippines Mortgage Loan Tuition 1976 13 Pakistan Mortgage Lw' Commercial Banks Tuition ad Living 1974 Sri Lanka Mortgage Lon Comrcial Banks 1964 NICOL! EAST. NORTH AFRICA Egpt Mortae Loan Autoom Body 53 190 Commercial lanks a3 1980 Israel Mortgage Loan Commecal Banks Tuition an Living 123 1983 Jordan Mortgaqe Loan Morocco Mortgage Loan Circial Banks Tuition 13 1990 SUB-ANARAN AfRICA Gha Incom Contingent Goermnt Dept. Living 8200 1989 683 1990 Kenya Mortge4e Loa Comrcial Bank Living $84s 17 1003 1990 Nigeria* Mortg"e Loan Autwrom Body Tuition Rwanda Burunidi* Malawi Mortgage Loan Living 180 1988 501 1989 Twzania* Mortgage Loan Living INDUSTRIA ONRI!S Australia Incm Contingent Government Dept. Tuition 51,730 1989 8IX 10 Canada (Quebec) Mortpge Loan Commercial Banks Tuition and living 52,800 1963 59% 1990 Oemerk Mortgage Loa Coinrcial Banks Living 53,700 1975 198S Finland "ertpgag Loan Living 52,200 193 198? France -a.tgage Loan Governt Dept. Living 12 Geny Mortgage Loam Livfng 1,500 1974 303 1987 hong Kong Mortgae Loan Gorvint Dept. Tuition and Living 51,050 1969 263 1919 Notherlads Nortgage Loan Living $200 1919 Norway Mortgage Loan Autons* Body Livirn 54,000 8OX 1986 Japan Mortoga Loon Autonmus ody Tuition and Livine 12,500 19 1987 Singapore Mortpge Loan Government Dept. Tuition and Living 39X 1990 Cainrcifl Banks Sweden Income Contingeont Autonmous Body Living 55,828 United Kingdom Mortgage Loan Autonomous Body Living 1750 1990 73 19 USA Mortgage Loun Commrcial Banks Tuition and Living 82,176 1964 283 1987 Planks imply Informstion was not aailable. Programs n Irnons, lsrel, Nigris, Tanania an lurundi hav bwen bndoed. Souce: Albrecht, 199I 101 Table V.2 BRAL HIGHE EDUCATION Susiie gng Governmenmt Loess. go 23StdtLon M maxiam or Hi' C t Nominal Resl Projected Grant to loss with loss with Interest Interest Repry nt students defoult default and year Country rate rate Period per cent dmcinistr- Es:imatiS of loan (1) (2) (3) (6) (7) (8) MORTGAGE LOANS Coltai I 1 11.0% a -10.61 8 n 76 c 87X 1973 Administrotive ZX Colonbi II 24.01 3.01 b 5 m 38 c 471 1965 Administrotive 21 Sweden 4.31 3.0X 20 611 621 701 1988 Administrative 1% Indone is 6.01 -2.31 b 10 571 611 71X 1985 oefoult ¶01 Ad 21 USA (GSL) 8.O a 3.81 b 10 29S 411 53S 1986 Administrative 2X Hong Kong O.01 -6.31 b S 431 43X 471 1985 Administrative 2X UK 6.01 0.01 7 261 301 411 1989 Default 51, Ado 1X Norway 11.51 a 5.6X 20 331 331 481 1986 Adminfstrative 11 Demrk 8.01 a 1.6X 10 521 56a 621 1986 AdMinistrative 1X Finland 6.5X a -0.61 10 451 461 521 1986 Default 21, Ado 1X Orsafl I ¶5.01 -35.01 b 5 91X 94X 98X 1983 Default 30X, Ad 2X Brazil It 318.0X -14.91 8 62X 651 711 1969 Default 101X AdM 1 Jaimica 1 6.0X -10.71 9 741 841 c 92S 1957 Administrative 2X Jamicea It 12.0X -5.61 9 56 621 c 701 1988 Default 20%, Ada 21 Sarbodos 8.01 4.11 12 132 18X 331 1988 DefOtlt 51, Ads 21 Kenya 2.01 -6.9X 10 701 941 e 1031 1989 AdinIstrative 2Z Quebec 10.01 a 5.21 10 311 311 371 1989 Administrative 1X Chile varies 1.01 b 10 48 691 c 821 1989 Administrative 2X Japan 0.01 -1.4X 20 501 51X 601 1987 Administrative 1X Veneauela 4.01 -231 20 931 981 1081 1991 Administrative 3X Nondwras 12.01 3% 8 S12 531 731 1991 Administrative S5 INCME CONTINGENT LOANS Australia varies 0.01 17 481 521 571 1990 Eviain 3, ads 0.51 Sweden varies 1.01 10 282 301 331 1990 Evasion 31, aft 0.51 General notes: All subsidy calculations use a real opportunity cost of capftal according to the governeent rate of borrowing or estimates used by the World Sank. Loans are assumed to be paid in equol installments over a four year perid, adjusted in size each year to keep up with inftation. Given the amailability of relevant data, Swedish Ircommcont1ngent calculation Is based on Australia's age earning profile information; (1) Cotntries with I and II refer to situations where the loon progrm uniderwent refom. (2) Nominal interest rate refers only to the rate during repyment. la refer to town that use a diffreent rate during the disburseent and grace period. 'b, denotes those program with interest rates Which are indexed. (3) Real interest rates use Purchasing Power Pority foroutl, where inflation Is based on the everage of the 1980-1988 peorod as reported in the World Dvelop ont Report, except in instances noted Wisee a 5 year averge of inflation was calculated from the data date. (4) The repayment length is th. maxima prescribed in the lon, except for the two fnrmo contingent loans chore it is the repayi.nt length that Is Iptlied by the aweras incom profile of a gradate. This does MI nelude grace periods. (5) The hidden grant percentage is caleulated as a discosnted cash flow of the student's accr mt, and therefore excluaft defoult and administrative costs. (6) The goverrment loss dus to default subtracts the X of default frm each year of the repasymnt stream. *el denotes sere these figures have been estisated. For Colocbia, Jamice, Chile and Kenya the figure used is loans in arrers. (7) The loss with default aid administrative costs subtracts an snat afdinistrative cost related to outstanding debt each Vear. (8) rear is date from which loon inforation was collected, nd ffrom ich inflation calculations were oad. Source: ALbrecht, 1991 Table V-3. O0_bsdim of _mob an fmnta for Hiwgh. Educalon With d Wibtut d m Raogun le;_n (Expendlrwe m ku 1990 US,t Fedeal State Imciplt Totat P$vate Other tot.l Syste-ide lnstits Instits Instits P"tic Universlties Private Private total Enrollments (so Reform) 318.000 191,000 7,000 55,000 506,000 703,000 1 210,000 1,795,000 Enrollments (artiatl eform) 318,000 191,000 77,000 S85,000 S06,000 703,000 1,210,000 1,795.000 Enrollments (Partitl Refom a) 477,000 238,000 77,000 92000 403,000 600,000 1003.000 1,795,000 Enrollauts (Partial leform b) 318,000 191,000 77,000 S85.000 S06.000 703,000 1,210,000 1,795,000 Expenditures (no Refom) 2,50 760 230 3,S30 1,010 600 1,610 S.140 Expenditures (Partial ReforM) 1,430 760 10 2,340 1.210 720 1,9O 4 270 Expenditures (Partial Reform a) 2,450 760 230 3,s30 970 610 1,560 5,110 Expenditures (Partial Reform b) 1,630 740 230 2,600 1.200 820 200 460 percents Federal State nIcipat Total Private Other Total System-wide Imntits Instits Instits P"lic Universities Private Priwvte Total Enrolltnts (No Reform) 1T.7X 10.62 4.32 32.6 28.2x 39.22 6.41 100.02 Enroltlments (Radical Reform) 17.71 10." 4.32 32.6 28.2X 39.22 61.42 100.02 Enrollments (Partial etform a) 26.42 13.32 4.32 44.11 22.52 33.42 S5.91 100.01 Enrollments (Partial Reform b) 17.72 10.62 4.32 32.62 28.22 39.22 67.48 100.02 Expenditures t(o Refom) 47.72 14.82 4.52 68.72 19.62 11.72 31.32 100.0X Expenditures (Radical Reform) 33.51 17.82 3.52 54.82 28.32 16.92 45.21 100.02 Expenditures (Partial Reform a) 47.92 14.92 4.52 69.12 19.02 11.9X 30.92 100.0S Expedtures (Partial Reform b) 35.32 16. S.01 S6.32 26.02 17.72 43.72 100.02 Expenditures at institutions r nt s of ptlic en private contributions. They do not reflect the source of the Income to finace expenditures. Asseas current fee levels In private universities averae US2,000 atl aro $85O In other private.