A VWwfivu LATHE No 34 HGHER EDUCATON REFORM IN CHME., BRAZIL, AND VENE TOWARDS A REDEF1INIION OF 1IE ROLE OF THE STATE Laumnce Wol and Dougas ATbredst I Bw,ax Resces DWibifOX Tedakl Department Lati Amra ad th Caribea Region The Wodd Bank Noeeaber 25, 1992 r 'A Viewfrom LA47R' is a szes of occasionalflyersproduced by the Human Resources Division of Latin Amenca and the Canbbean TedumcalDepanrmcntof the World Bankfor the purpose of nimulating disacssion oong stffon key issfacing the ector. The views epassed here arwthose of the authors and should not be atibuted to the World Bank HIGHER EDUCATION REFORM IN CHILE, BRAZIL, AND VENEZUELA TOWARDS A REDEFINITION OF THE ROLE OF THE STATE Table of Contents Overview by Laurence Wolff and Douglas Albrecht I.1-16 Higher Education in Chile: Effects of the 1980 Reform by Jose Joaquin Brunner and GuillermoBriones 11.1-44 Higher Education in Brazil: Isues and Efforts at Reform by Laurence Wolff, Douglas Albrecht, and Alcyone Saliba m. 1-40 Higher Education in Venezuela: Issues and Options for Reform by Laurence Wolff and Jose Joaquin Brunner IV.1-18 Laurence Wolff and Alcyone Saliba ar staff members of the World Bank; Douglas Albrecht is a Consultantfor the World Bank; Jos6 Joaquin Brnner is Coordinator for Intemational Cooperation atthe Latin AmericanFaculty for Social Sciences(FLACSO)in Santiago,Chile; and Guillermo Brionesis a Professor at the Social Sciences Faculty of the University of Chile. The editors would like to thank Thomas Eisemon, Jamil Salmi, and Julian Schweitzerfor their helpful commentson drafts of this report. I OVERVIEW Table of Contents INTRODUCION ....................................... -1 HIGHER EDUCATION REFORM EFFORTS IN CHILE, BRAZIL, AND VENEZUELA ................................. I-1 Background ................................... -1 Chile's Reform Effort ................................ I- 2 Brazil: Starting the Reform Process ......... ............... I- 7 Venezuela: the Need for Reform ........... ............... I-10 CONCLUSIONS: THE NEW ROLE OF THE STATE ................ 1-12 I~~~~~~ INTRODUCIlON' This volume examinesthe issues related to higher educationreform in three Latin American countries. The countriesstudied-Chile,Brazil, and Venezuela-canbe consideredto be at three stages of reform: ten years after a radicalrestructuring and finance(Chile);initiatingimportant of institutions reforms (Brazil);and before embarkingon a reform program(Venezuela). This overviewsummarizes the reformprocessand issuesin thesethree countries,and identifies commonthemesapplicableto them and to the Latin Americanregionas a whole. H1GHEREDUCATIONREFORM EFFORTS IN CHILE, BRAZIL, AND VENEZUELA Backgund Higher educationenrollments in Chile, Brazil, and Venezuela,as in all of Latin America,have grownenormously in the last thirty years. In Chile, enrollments went from 25,000 in 1960to 250,000 in 1990, in Brazil enrollments increasedfrom 100,000to 1.6 million,and in Venezuelafrom 25,000 to 640,000over the sameperiod. In BrazilandVenezuela most of this growthtookplace throughthe early 1980'sand has considerably slowedin the last fiveyears. A muchincreasedproportion of theschoolagepopulationnowattendshighereducation institutions. Enrollmentin Brazil is equivalentto 12%of the 20 to 24 yearold age group, Chile enrolls 18.6%, and Venezuela 26%. It shouldbe notedthatthesefiguresare gross' enrollment figures. The net enrollment figures(e.g., enrollmentless repeaters)wouldbe muchlower, especiallyin Venezuela,whichhas very high levels of repetition and where students continue to be enrolled in higher education institutions for manyyears. In Brazil, more than 60% of enrollmentsare in private institutions. The growth of private educationwasa consequence of a governmentdecisionin the 1960'sto encourage privateeducationwhile at the same time preservingthe quality of public institutionsby enforcinghigh entrancestandards. Brazil's publicsystem can be furtherdividedin federal(22% of enrollment),state (13%, mainlyin Sao Paulo), and municipal(5%)institutions.In Chile, the 1980reformstransformed the systemfrom a small and exclusivelypublic system into a system in which half of enrollmentsare currently in private institutions. Colombiais anotherLatin Americancountrywith a similarlystrongprivate sector. In contrast to these systems where the private sector is, at least quantitatively,dominant, Venezuela'spublic system enrolls 78% of all students. Most other Latin Americanhigher education systemshave relied predominately on public provision. In Mexico,Argentina,and Peru for example, publicinstitutionsautomatically receivestudentsthat have graduatedfrom secondaryschoolsand enroll over 80% of all higher educationstudents. Venezuelahas an entranceexaminationwhich is highly but which is in effectan open accesssystemto the weakerpublic selectiveto a few of the best institutions institutions. With regard to financing,Venezuelarelies on public resourcesto finance 89% of the costs of and studentaccounts highereducation. In contrastin Chilecurrentlydirectpublicfundingof institutions for 309% on highereducation; of total expenditures undetermined an additional amount(upto 34%) comes from a variety of sourceswhich includepublicfundingthroughspecialgrants and transfers. In spiteof noted,theda in this Overview Unlessotherwise fromtheindividual comas cae sudies. 1-2 the large enrollments in private institutions, public funding in Brazil accounts for 75% of total expendituresin highereducation,a resultof a combination of high unit costs in public institutionsand prohibitionon chargingtuition in public institutions. a constitutional Table 1 providesa summaryof the situationin thesethree countries. Table 1 Enroment and Financingof Higher Educationin Chile, Venezuela,and Brzil Chile Venezuela iail (1990) (1990) (1989) Enrolment as a Perctagp of SchoolAge Population 19% 26% 12% Percentageof Totl Enrolment in Public Institutons 48% 78% 40% Percentageof Higher Education FinancingWhich is Public 30% " 89% 75% Percentageof Total PublicEducation BudgetGoing toHigher Educa±iom 22% 35% 23% UnitCostsof PublicHigherEducation Institutions USS1,700 USS1,625 USS7,930F StudentIFTEFacultyRato (publicinst) 10.8:1 16:1 8.8:1 a Plus an additionalundeteminedamt of gnts and transferece. b. Estim at US$6000not countingcosts of teachinghospitalsand pensi. ChUe'sRefonr ELfor Chile, in 1980, was a small, homogeneous and exclusively public system of higher education. Roughly 6% of the relevant age cohort was enrolled. Vrtualy all funding came from the state and the only institutional type was the multi-disciplinary university. The reforms implementedin December 1980aimed at a radicaloverhaulof the higher education system, to introduce market forces, to increase the efficiencyand responsivenessof institutionsto economicdemands. In addition, the system was to be expandedrapidly via the harnessingof private, rather than public resources. 2 , the reform Accordingto the Cerychand Sabattiermodelfor analyzinghighereducadonreforms had five outstanding characteristics. The degree of system change envisaged by the reform was high, 2 and MixedPerformance.The Imnlementation See Cerych, Ladislav& Sabattier,Paul, Gret Expoctations of Higher EducationReformsin Eurome, EuropeanInstituteof Educationand SocialPolicy,Tradham Books, 1986 I-3 affectingthe boundariesof the system itself as well as all institutions. Second, the scope of intended changeswas inclusive,affectingpracticallyall individualswithin each institutionand potential actors outsidethe system. Third, the depthof the reform itself led to the adoptionof radicalpolicies. Policy goals stronglydivergedfrom existingvaluesand practicesof highereducation.Fourth, the goals of the reform were clearlystated and had internalconsistency. And finally,changeswere brought about by centraledict, from top to bottomandthrough bureaucratic enforcement. The mainobjectivesof the 1980reformas articulatedat that time canbe summarizedas follows: To increaseaccessto higher educaton throughexpansionof privateeducaton. To meet this goal, only minimalrequirements for the creationof privateinstitutions. were established To diversif the insitutions In the highereducaton system. Threevertical tiers were established, based hierarchyof educational on a functional certificates: * Universities: expected programsleadingto licenciaturas to focuson long cycleundergraduate and to professionaltitles requiringa licenciadodegree. Only universitiescould establishgraduate programs. * Institutes(PIs): restrictedto four year programsleadingto professional Professional titles defined as not requiring the licenciatura. * TechnicalTraining Centers (trCs): restrictedto short circle (two year) vocationalprograms leadingto technicalcertificates. To reducethe instutionalpower ofthe twotrdifionalstate universities. Both the Universidad de Chile and the Universidad Tecnicadel Estadooater Universidad de Santiagode Chile) were forcedto give up their regionalcenters, whichled to the birth of twelve new state universitiesand to two state Pis. The breakingup of thesetwo universitieswasjustifiedas a wayof makingtheseinstitutions more manageable and as a meansof encouragingdiversity. Topartialy trawfer the costof sate-nanced inttudonsto the students and/or their families and force these institutionsto diversifytheir funding sources. Tuition was introducedinto all publicly funded institutions. The promotionof privatehigher educationwas another meansof shiftingthe burden of financeto private sources, since they are solely dependenton resourcesobtainedfrom tuition fees for recurrent expenditures. A Government financedstudent-loanschemewas introducedto be providedto needystudents in public institutions. To increase competiton among nsitutions, with the aim of promoting efficiency and quality. Institutionalsupport, which was confinedto the eight traditionaluniversitiesand the 14 new state- institutions,was supplemented throughinstitutionalself-financing: for research tuitionfees, competition funds, and contractfunding. A small amountof publicfundingwaslinkedto a "best-students formula, based on the numberof the best 20,000scorers in the lastyear's entranceexaminations enrolledby each institution.' The civil-servantstatus of academicpersonnelin state institutionswas eliminated,thus enabling universitiesto differentiatesalaries and competefor faculty staff. Tbe Governmentset up a Thisfigurewaslaterbroadened thebest27,500 cors. to include I4 publiclyfinancedNationalResearchFund (FONDECYT),whichmadeawardsto researcherson the basis of open competitionand peer reviewof proposals. No publicfundingwas visualized for new privateinstitutions,but for one exception:as of 1989, they can competefor asubsidy-carrying students (the 27,500 mbest and brightest"). Studentsenrolled in privateinstitutionshaveno accessto subsidizedloans. Finally, privategrantsto both private andstate universitieswere stimulatedthroughspecifictax deductions. Table 2 summarizesthe major institutionalchanges: Table2 Summaryof Higher EducationSystemBeforeand After Reform BEFORE 1980 AFTER 1980 1. One tier, one sector. low institutional 1. Three tiers, tuo sectors: high institutional differentiation. differentiation. 2. No tuition but aelective access accordingto 2. Tuition fewsare charged by all institutions. school performance and achievement in Selective accem is mainaed in Ihe mib- nationalstndlardizedacademictest sector t rcdvs institutionl public fumding.In the private institutions: open ace accordingto and/or familyincome. 3. State financing of HE on the basis of 3. Multiplecompetitivesourcesof funding. icm l fimding. 4. System oordination provided by State 4. Systemcoordination providedby makeu and authorityand institutional.oligarchies, policy regulations. Chile's higher educationsystemwas reformedunder specialand very harsh conditions,due to the existenceof an authoritarianMilitary Govermment. B.Clark's assertionon the feasibilityof authoritarian reforms in higher educationappliesto the Chileancase: 'Of course, in systemsunder authoritarianor totalitarianrule, the centralization of authorityand the centralconcentration of administrationallowmore manipulationfrom aboveand on a large scale'.' Institutionshad to endurethe changesenvisagedby the reformers without having the opportunityto intervene in the process nor to negotiate its outcomes. Authoritarianmindedreformers will probablysay that only becauseof this could system-widechanges be brought about and that reforms need not be assessedby the context in which they took place nor by the political conditionsthat lay behindor made thempossible. Reforms,they will argue, shouldonly be evaluated in terms of their actual outcomesand more general effectson society. In turn, those who experienced these changeswithoutanypossibilityof resistingor discussingtheir aimsand implementation procedureswill probablyfind it more difficultto have confidence in the blessingsof supposedlyefficient- Clark, Burton, *he OrganizationConqeion- in B.Clark (ed.), Persuctives on Higher Education.Eiabt DisciRlinarvand ComparativeViews, Universityof CaliforniaPress, 1987,p. 125 I-5 authoritarianrule. In short, the cost of reforms in Chile must also includethe shatteredhopes and the repressionsufferedby those who wouldnot yield. Tenyearslater the reformto a great extenthas achieved results its goals. Thepositivequantitative are as follows: * A unitarysystemof eightpublicuniversitieshas beenreplacedby a highlydifferentiated one of 310 institutions,includinguniversities,professionalinstitutes, and technical training centers. * The percentageof the school age populationenrolled in higher educationhas more than doubled. All the enrolmentincreasesoccurredin the private sector, which now has 52% of enrollment. * Real public expendituresin higher educationbetween 1980 and 1990 decreasedby over 30%. The percentageof the publicbudgetgoing to higher educationdecreasedfrom 38% in 1980to 22% in 1989, while the percentageof public educationexpendituresgoing to primary educationincreased from 39% to 50%. The real amounts going to primary education have increased by only 12%. During this period total enrollment in primary educationdecreasedby 89%, a resultof the demographic transitionand improvedflowrates. This means that, contraryto most other Latin American countries,unit expendituresin primary educationwent up duringthe decadeof the 1980's. At the sametime enrollment and fundingof pre-primaryeducationincreasedby about 70% and the percentageof the educationbudgetdevotedto pre-schooling increasedfrom 4% to 7.7%. 3 * Whereasin 1980100%of supportto highereducation wasin the form of directinstitudonal support, only 56% is now in this form. The new fundingmechanisms now include a per capita paymenton the basis of studentsrecruited with high entranceexaminationscores, 16%; studentloanscheme, 15%; and scienceresearchgrants, 8%. i Studenttuitionand loan repaymentnowaccountfor 28% of overall expenditures of public institutions. Studentson averagepay aboutUS$700per year in tuitionto publicinstitutions. * Qualityin the public instituionshas not suffered. The percentageof teachersin the public sector with graduate degreeshas increasedfrom 19% to 24%. The elite public sector universitiesaccountfor 77% of scientificpublicadonsin the countryand Chilecontinuesto lead Latin Americain per capitaproductionof scientificresearch. The less than positiveresults of the reformincludethe following: * Accessto highereducationcontinuesto be inequitable changedsince but has not significantly 1980. Currentlythe lowesttwo quintilesof the studentpopulationaccountfor no morethan 11% of highereducationenrollment. Lower incomestudentsare more likely to enroll in 5 Dataon the overall educationsystemcomesfromtheWorldBankStaffAppral Report,"Chile,Pimy Improvement Education Project,September1991. 1-6 public than in private institutions. One reason for this is that loans are availableonly for public institutions. * Iternal efficiencyof the systemhas declined. It nowtakes about9.2 years to produceone graduatefrom a public institution,comparedto 7.1 years at the end of the 1970s. * FTE studentteacherratios in publicinstitutionsare 10.8:1 and havedecreasedslightlysince 1981. In short publicfundingdeclineswere offsetby cost recoveryand revenuediversification in public institutions. The qualityof instructionand researchat public institutions,whichwas historicallyhigh by Latin Americai standards, has not suffered. The reform measuresclearly prevented quality declines which mighthave occurredif the publicsystemhad beenallowedto expandrapidly. Despitesignificant cost recovery(combinedwith studentsupportschemes- scholarships and loans), the system as a whole has not becomemore inequitable. A portionof the publicfundssaved havegone to primary education.' Higher educationin Chile still faces a numberof unresolvedproblems. One problem is that of relativelylow intermal efficiency,in terms of high dropout and repetitionrates, at many of the public institutions. Despitecompetition, there is no evidencethat staffare better deployedin publicinstitutions. There are high defaultrates in the studentloanschemeand it is still not availableto private institutions. Further reform in Chile may need to focus on changingthe funding formulato explicitlyencourage increased efficiencyin terms of graduationrates and faculty utilization,as well as to strengthenand reformulatethe loan program. A final issue of concern, which is currentlybeing addressed, is the low quality of many new private institutionsand programs. In the firstplacethese insteutionsare excessivelysmall. Of 40 private universities,only five enroUl 1000 studens or more. These five accountfor 68% of the sector's total enrollment. Most of the other private universitieshave less than 500 students. Private university teachingbodies sufferfrom lack of objective,merit-basedcriteria for the recruitmentof academicstaff, lack of a core group of full-timeacademicpersonnel,high turnoverrates of faculty, the absenceof an academiccareer structuredaccordingto the needs of the institutions,inadequateworkingconditions,and the lack of library, laboratoryfacilitiesand equipmentsupportfor academicwork. These institutionsare also inadequatelyaccredited. The proceduresup to 1990involvedprior approvalof the new institution's teachingprogramsby a publicsectorinstitudonfreelychosenby the privateinsttution. After satisfactory completion of a trial period under the aegis of the public institution,the new private institudon automatically gained its full autonomy. However,some of the examining public institutionswere weak, there were no public assurances that academicstandards are being achieved,and the *commercial- arrangementbetweenthe two institutionsdid not encourageobjectivity. Under a new law passed in 1990, a public, autonomous body-the Higher Councilof Education (HCE)-has been set up, with two major functions: to accreditnew privateuniversitiesand PIs; and to approve the core national curriculumfor primary and secondary education. To encourage improved I It was never stated that an objectiveof the reformwu to shift fund to low'ereducatonal levels. In fiat the initialproposalsaumed tat muppotfor higher duchation would sgnificantly crae in terms through a vastly expandedloan scieme and 'bedt oh-Am formulafunding. At the same time the Governmentdid explicitly ate tht swpportfor basic services,includingthe lower levels of education,should increase. 1-7 quality, the accreditationproceduresset up under the 1990 law provide a licensing system for the establishment of new private universitiesand PIs centeredaroundthe approvalof both an institutional project and the career-programs that willbe offered. After a minimumperiodof six years, the HCE can declare the full autonomyof a private institution. The long term objectiveis to encouragequality improvement of privateinstitutionsthrougha systematic, objectivereviewprocess. A recent strategyto increase qualityhas been to permit private instituions to competefor public funding under the wbest students' formula. However,as yet privateinstitutions are not eligiblefor studentloans, which,properly maged, couldbe anotherform of indirectsubsidyencouraging improvedquality. BrudLStardng the Refom Process Reformof highereducation in Brazilstartsfrom a differentpoint than in Chileand with a different emphasis. In the first place in 1968Brazil's Government made an explicitdecisionto encouragethe growthof private education and to restrictaccessto publicinstitutions,whichwere to be the high quality end of the system. This decisionis reminiscent of Chile'sdecisionin 1980. Sincethat time enrollment has increasedrapidly in private education,which now accountsfor 60% of enrollment. Like Chile, Brazil has benefittedenormouslyfrom this decisionthrough being able to provide increasedaccess to highereducationwithoutincreasesin public funding. Brazil's 1968 reform, however, did not pay attentionto issues related to fnancing of public insdtudons,and to autonomy,accountability, governance,and equity. The result is that Brazil's public educationsystem, especiallythe federalinstitutions,is excessively expensive. In fact the unit costs in Brazil's federalhighereducation system,estimated at USS7930, 7 are amongthe highestin the developing world, a result of low studentstaff ratios and high administrativecosts. This figureis fivetimeshigher than the costs of most private institutionsin Brazil, twice as high as the costs of the two best private institutions,and significantlyhigherthan the two insdtutions run by the State of Sao Paulo, which are arguablythe highestqualityuniversitiesin Brazil. Enrollment has been staticsince in federalinstitutions 1980 while faculty and non teaching staff have increased significantly,at the same time that these institutionshave under-utilizedphysicalfacilities. The result is that, despitethe fact that only 40% of enrollmentis in private institutionswhich receivevery little publicsubsidies,75% of the financingof higher educationis public. Furthermorefree tuitionin all publicinstitutionshas exacerbated the always troubling issuesof inequityin publicexpendituresin higher education,especiallyconsideringthat the socio-economicstatus of students in the public institutionsis on average higher than in the private institutions. FinallyBrazil's highereducation basedon the conceptof the large multi- systemis still excessively purpose university linzking teaching, research, and extension, in spite of the fact that only a few institutionsare able to successfullymeet these objectives. Brazil has only a fledglingsystemof short course post secondarytechniciantraining. Brazil's publiceducationsystemhas neitherautonomy,sincethe Government excessivelymanages and mandates most higher educationundertakings,nor accountability,since, with the exceptionof graduate education, there are no systems of financialor qualitativeassessment. Funding of federal institutionsis on a negotiated,incremental basis, withno built-inincentives for improvedefficiency.The federalgovernmentsets all personnelpolicy, includingrecruitment,promotion,and salarylevels, much ' About US$6000if the costs of universityhospituls and of pensionsare excluded. I-8 of whichis encodedin the alawof isonomy' passedin 1987. While rectors are officiallyselectedby the Ministerof Education,between 1985and 1991generallyonly those candidates with the highestvotes in an open electionwere selected. The rectorsoftenhave little power and inadequateexpertiseto manage their institutions. The Federal Councilof Education(CFE) establishesminimumcurriculumfor all courses, which covers94% of the total of creditsneededto obtaina diploma. The CFE, chargedwith authorizingnew private institutions,makes such authorizationson the basis of unclear and subjectivecriteria including 'labor market demand.' Regulations are suchthat diplomasare the entrypointsto most professions,thus encouragingexcessiveacredentialism.' Brazil's large private sector is of great importanceand varies from high quality institutionsto diplomamills'. Howeverpublic policycurrently discouragesquality improvementof private institutionsthrough tuition controls, regulationof professions,and inadequate programs of assessmentand certification. By 1990budgetarypressuresbrought aboutby the overall compression in governmentspending, an increasingpublic impressionthat large public expenditureswere not resulting in an adequatereturn in terms of the qualityand quantityof graduates,a growingrealizationof the impossibility of improving qualityunder the current system,and continuous strikes by facultyand students,had put Braziian higher educationin crisis. A new team of reformersin the Ministryof Education(MEC)took officein late 1991. This team wasunusualin the Braziliancontextbecauseof its high degreeof technicalcompetence. The teuam began to put forth a comprehensive packageof reforms, designedto make Brazil's higher educationsystemmoreefficient,effectiveand equitable,to improvethe qualityof output, and at the same time to reduce unit costs of federal higher education. The reform especiallysought to change the relationshipsof Governmentto public and private institutions,through establishingnew systemsof autonomy,accountability, evaluation,certification,and financingof both public and private institutions. As in Chile, the reforms soughtto changethe financingrole of the state so as to increase equity and efficiency. With regardto the financingof federalhighereducation,the reformersin the Mimistry of Education (MEC) had as their long term objectivethe provisionof financialautonomyto federal higher education institutions, as a means of ensuring that institutionsoperate efficientlyand effectively. The main instrumentof financingwouldbe a fundingformula explicitlyencouraginghigher studentteacher and studentstaff ratios, reducedstudentdrepout, morerapid completion of degree studiesby students,more highly qualifiedstaff, and an initial decrease in unit costs of 25% to be achievedthrough increasing enrollment in public institutions. The funding formula, summarizedin Annex 1, Chapter m, is a sophisticated adaptationof the formulaused in the Netherlands.Formulafundingwouldbreakthe culture where accountabilityis absent, since institutionsthemselves would have to make.trade-offs among elementssuch as salaries, maintenanceand equipment. The formula would be a dynamic instrument, changingover time. Institutionswould be subject to audits to ensure the reliabilityof the staistical reporting which forms the basisof the formula. Through early August, 1992, MECused its formulato allocateportionsof the discretionary funds directly under its control. In 1993, it plannedto allocate all of these funds (about 10% of the higher educationbudget) on the basis of the formula. Based on the formula, by mid 1992, a number of institutionshad already movedto increaseenrollments(manythroughnight classesthat facilitateaccess for lower incomestudents)and to increasestaff qualifications. 1-9 MEC also introducedseveral laws that would make higher educationinstitutions"autonomous public institutions'not subjectto normalcivil serviceregulations. Parallelto this, MEC was seekingto permit all federal institutionsto establishprivatefoundationsunder their control as a means of giving them furtherflexibility to seek new sourcesof fundingand to provideadditionalservicesto government and industry. Over the long run, MEC was seekingto replacethe law of 'isonomy", which providesa single salary and promotionpolicy for all higher education teachingstaff, as well as civil serviceprotection, with a new systemunder which teachingand non-teaching staffwould be employeesof the institutions where they work. In principle,this wouldallowinsdtudonsto developdifferentiated programsand to manage their funds more effectively. Becauseof possible oppositionfrom the key groups affected, and time consuming. As an intermediate passageof legislationof this sort wouldbe difficult measure, a law which establishesa floor for salarieswhichhighereducationinstitutionscouldsupplement with their own funds wassubmittedto Congress. On this basis MEC hoped to enable institutions to be free to establishtheir ownset of salaryincentives. Passageof laws on financialautonomyand on isonomy wouldbe essentialfor the fundingformulato have its full effecton institutionaldecisionmaking. MEC also initiatedwork on a long term plan for evaluationand certification of undergraduate higher educationinsdtutionsand programs,to be mainlycarried out by peer review committees. The main objectivesof the programwould be to providepublicly availableinformationon the quality of public and private institutions,so as to help inform students' decisionson attendance,as well as to as a meansof encouragingimprovements partiallytie public fundingto the resultsof these evaluations, in quality. MEC wouldbeginwith evaluations of medicaleducaton followedby teachereducation. MEC also sought to reform the studentloan program. In contrst to Chile, the loan scheme is available solely to students attendingprivate institutions. Similar to Chile, it has sufferedfrom a combination of high subsidiesand defaultsand has had the effectof supportingthe lowestqualityprivate institutions.The objectiveof the reformwas to ensurethat federalfundscouldassistthe largestnumbers of studentsat the lowestcost to Government and to utilizeloans as a cost effectivemeansof encouraging increased quality in the private sector, through setting quality-based criteria for loan eligibility. The publicsubsidywouldbe reducedand repayment wouldbe on an income contingentbasis. In May 1992 Congresspassed a law making MEC responsiblefor establishing the regulationsfor the studentloan program,and mandating a minimumof USSS0 millionper year for the program. This was a major step forward, since it meant that the loan programcouldbe integratedinto nationalhigher education policy. MEC planned to introduce legislationto deregulate a large number of professions, with the exception of the traditional areas of health services, law, structural engineering, teaching and accountancy. Deregulationwould help to end the pervasive 'credentialism of higher education. Professionalassociationswere expectedto opposevigorouslythese proposals. MEC was also planning to introducelegislationleadingto indirectelectionof rectorsthrough electionby the universitycouncil. Many of the laws and regulationsdescribedabovewouldencourageinstitutionsto diversifytheir sources of income. For example,financialautonomy wouldfree institutionsfrom excessivelystrictcivil service regulations. Current regulations,for example,do not permit institutionsto depositfunds into interestbearingaccounts,whichis fundamental in a high inflationeconomy.The establishment of private university-associatedfoundations wouldalso encouragediversification of income. Permittinginstitutions to use their own funds to top off teachers' salarieswouldalso functionas an incentivefor universities to seek additionalsources of revenue. Implementation of the fundingformulawould make institutions I-10 more awareof the trade-offsin expenditure categoriesand wouldencouragethemto reducesubsidiesin the provision of food and other ancillaryservices. The Governmentwas also consideringa matching grants programto further encourageprivatesourcesof fundingfor public institutions. Finally,over the period 1991-92,MEC reformersfloatednrumerous ideas on waysand meansof asking studentsin public institutionsto pay for a portionof their education,either now or in the future. These included a 'parent' tax, a graduate income tax, and encouraginginstitutions to charge for miscellaneous servicessuch as diplomasor parking. Anotheridea, currentlybeing implemented in one state, wasto establisha fund whichbetter off studentswouldhave to contnbuteto, but which wouldthen be used to support needystudentsattendingthe same institution. Given the political uncertainties,and in order not to jeopardizethe implementation of the other reforms, the reformers at MEC were not immediatelyseekinga constitutionalamendmentwhich would permit the chargingof tuition in public institutions. Theyexpectedto seek such changesin a year's time, whenthe entire constitutionwouldbe revised. The reformprogramthus put forwardwas significant.Its focus wasto improvethe efficiency and effectivenessof the current public interventionin higher education. The strategy was to provide institutionswith greaterautonomyand to introducenew mechanismof accountability,especially through implementation of a newfundingformulabut alsothroughrelatedprogramsof evaluationand assessment. The fundingformulawouldpromoteefficiency, whilethe studentloanprogramwouldbe used to promote equityas well as qualityin privateinstitutions. In mid 1992, Congressas well as the informedpublic, were becomingincreasinglyconvincedof the importanceof higher educationreform, and the prospects for implementing many of the reform proposalswere good. Unfortnately the resignationin early August1992of the Ministerof Education and his team, for reasonsunrelatedto the reformeffort,temporarilystoppedthe reformprocess. Venwla.* theNeedforRefonr As noted above, Venezuelahas seen a vast expansionof the higher educationenrollmentsfrom 1960to 1990, and now enrolls26% of the schoolage population. 80% of its enrollmentis public,and tuition is free. While there is a nationalentranceexamination,there is essentiallyopen access to the lower qualitypublic institutionswhile accessto the three or four higher quality institutionsis available only to those studentswho score higheston the examinations. Besides this vast expansion,Venezuela has had a numberof other successes. To a larger extent than other systems, Venezuelahas channeledmore expansioninto lower cost, shorter length technical institutions. 800 graduate education programs enrolling 15,000 students have been established. CONICIT,the sciencefunding agency,appearsto be a well operatedand functioningagencyproviding research grants on a competitive basis and the Government is acting to increase its investmentin R&D significantly. Nonetheless,as in Brazil, there is nowa generalperceptionof decliningqualityin the publichigher educationsystem, an apparentresult of a demoralizedacademicstaff, decliningsalaries,poor and overly politicizedmanagement,inadequateinvestmentin libraries and equipment,and turmoil in the form of strikes and protests by students and faculty. I-ll takenin the 1960's, 70's, and early To a great extentthe currentproblemsare a resultof decisions 80's, whenthe Government vastlyexpandedthe publichigher education system. As the large amounts of fundsfrom the oil boomgraduallydisappeared,theonly way Government couldhope to achievethese two goals simultaneously was to increase the amountof public funds going into the system to the detrimentof lower levels of educationand other socialprograms. As a result Venezuela now devotes 35% of its educationbudgetto higher education,a figurewhich, alongwith Costa Rica, is the highest in Latin America. In short Venezuela took exactlythe oppositepolicy decision,with regardto public fundingof highereducation,that Braziltook in the late 1960'sand Chiletook in the early 1980's. The Governmentnow finds itself unable to support a large public system, little or no tuition, and the semblanceof quality in a few of its public institutions. While unit costs in the public systemare comparable to public systemin other Latin American countries,only 26% of all enteringstudentseventually graduate. Overallthe publicsystemrequireson average 16 student-yearsper graduate, comparedto about six student years for private institution graduates,and nineyears for Chileanpublicinstitutions. Thusthe costsper graduateare far higherthan they should be becauseof the inefficiency of the system. Low internal efficiencyis caused by two factors: a relativelyopen admissions polices, and weak universityadministrationwhich is unable to establishor enforcerules on studentattendanceand failure. The combination of relativelyopen accessto public institutions,provisionof free tuition, and subsidized foodand transportmeansthat a broadersocio-economic groupin Venezuela comparedto Chile or Brazilis able to attendhighereducation,especially the publicinstitutions.However,highereducation continuesto serve a relativelyelitegroup, in part becauseof inadequatequalityat the lower levelsof the educationsystem. About 30% of higher educationstudents are in the worker and marginal category, comparedto 80% in the generalpopulation. The resultis that, in Venezuela as in Brazil and elsewhere, the populationas a whole is payingfor the full costs of a relativelyprivilegedclass whichwill benefit monetarilyfrom higher education. Free public educationalso encouragesstudentsto continuetheir studies, thus contributng to the inefficiencyof the publicsystem. Venezuela'spublic system also suffersfrom a poor budgetingsystem which fiurtherencourages inefficiency.Like Brazil, its budgetaryallocations arebasedon ad-hoc judgments,proportionalincreases and political bargaining.Other elementsin the financingsystem which are counter-productive include, as in Brazil, the requirement that salariesare the samefor all teachersthroughoutthe system, as wellas a system of early retirementwhich results in universitieslosing their best staff at the time of highest productivity. In 1991 and 1992, the Governmentbeganto discusson a tentativebasis a numberof possible reforms. In particular, there were initial discussions on developinga formulabasis for funding which might have a modest impacton efficiency. The proposedformulawas designedto encouragethe full complementof studentteacher contacthours, as well as mandatespecificamountsto research and to library and other support services. There was also a growingtendencyto restrict the intake of new students at public institutions. There was as yet no significantpublic discussionon cost recoveryin public institutions. A more radicalreform agendafor Venezuela, similarto that of Chile or Brazil, shouldfocus first and foremoston the changesin the financingof higher education. Efforts shouldbe made to increase the private financingof higher educationand reduce the excessivelyhigh burden on public funds. The most straightforward reform wouldsimplybe to chargetuition fees to all studentsand then to expanda 1-12 loanand scholarshipschemefor the needieststudents. Venezuela has a numberof fledglingstudentloan schemeswhich couldbe built on. However,in Venezuela as in mostof Latin America,the long tradition of free public educationmakes cost recovery in public institutionsdifficultin the short run although nonetheless necessaryin the long run. A less immediatelypainfuloptionfor increasingprivate fimancing of higher educationwouldbe to restrict enrollmentin the publicsystem and thus encouragethe growth of private education. This is already happeningin Venezuelabecauseof public financialconstraints. Throughthis policythe public systemwouldincreasingly focus on higher quality,graduateand research oriented education;and the current loan schemeswould be strengthenedto help support the neediest studentsattendingprivate institutions. The secondpart to a reform agenda,similarto that contemplated in Brazil, wouldbe to changethe way government fundspublicinstitutions, throughimplementing a fundingformuladesignedto encourage improved quality and efficiency,enhance faculty performance and encourage self-assessment. This formula could be based on how the Governmentestimatesthe systemshould cost rather than on how higher educationinstitutionswould like to spend such funds. In particular, in the Venezuela case, formulafundingshouldincludea parameterbased on the numberof graduatesof each institution,rather than enrollment,so as to createa strong incentivefor improvedinternalefficiency. Other criteriacould be based on performanceand efficiencyin the use of resources, and safeguardscouldhelp ensure that institutionsdo no arbitrarily award degrees to get increased funding. In addition institutionscould competeon an open basisfor someportionof fundingwhichcouldbe heldbackand awardedon the basis of innovativeor particularlycost-effective programs. The result of increasedinternalefficiencycould be a muchlower cost per graduateas well as increasedsocial and private returnto investment in higher education. Along with fundingmechanism based on formulas,contrctual arrangements,studentfees and other similar provisions, Venezuelawould need to expand its utilization of institutionalself- assessmentproceduresand the use of externa evaluations and accreditation. CONCLUSIONS:THE NEW ROLE OF THE STATE The reforms undertaken in Chile and under discussion in Brazil seek to increase diversity, strengthenautonomyof public institutions,increaseaccountabilityof all higher educationinstitutions, of all institutions. The objective is to increase access and equity, and encouragequality improvement develophighereducationsystemsbetter able to meet the demandsof the twentyfirst century. To meetthese goalsthe reformsradicallychangethe relationship of the state with highereducation institutions. The role of the state as a financingagent is reduced and additionalprivate resources are mobilized. To mobilize private sources, tuidon charges are establishedor increasedto cover a large portionof the teachingcostsof undergraduate educationin publicinsdtutions,and publicenrollmentsare restricted so as to encouragethe growth of the private sector. Increased private funding results in incentivesfor efficiencyand helps provide labor market signals to institutionsregarding appropriate courses and methodsof instruction. It also protectsthem from the vagariesof excessivedependence on the state. The financialrelationshipbetweenthe state and higher educationinstitutionsis also changed. In the past Governmentsdirectly funded institutionsand negotiatedincrementson the basis of political bargaining. In the reformed system, governmentsupport of institutionsis provided on the basis of formulas designedto encourageaction by institutionsto meet mandatedgoals of increased efficiency. quality, or equity. Governmentsupport also increasinglyfocusseson the areas withinhigher education such as research and graduate education,which have externalitiesin terms of encouragingeconomic 1-13 development, and on institutionsin backward or underdeveloped regions. Financialsupportis provided increasinglythrough open competition based on peer review, especiallyfor research. In additiona significantproportion of governmentfunds are provideddirectly to students or on a student basis. Supportmechanisms includescholarships (or vouchers)on the one hand, and loanprogramsto encourage individualsto makeinvestments in theirhumancapitaland enablingthemto pay for that investment when they are reaping its benefits. Subsidizing studentsthrough scholarshipsand loans rather than directly subsidizinginstitutionsmeans that Governments place greater choice of institutionsin the hands of students,with possibleencouragement of increasedcost efficiencyof higher educationsystems. With regard to governanceand oversight,Governmentsin the past strongly intervenedin the operation of institutions,directlymandatingexpenditures,enrollments,courses, staffingpatterns and salaries. The new role of the state is to operateat a distance,establishing the rules by which a diverse set of institutionsoperate and relateto the overallobjectivesof the higher educationsystem, including settingclearrules governingthe establishment of new institutions,whichgivethemthe autonomyto make judgmentsabouthow to use their resources to efficiently undertake their instructionaland researchtasks. At the sametime Governments developindirectmechanisms to ensure accountability.Theseincludenot only fundingmechanisms whichtie resourcesto the quantityand qualityof differentactivitiesand which also make awardson the basis of open competition, but also strong mechanisms to openlyand publicly assess and evaluateinstitutionsand programs. Chart 1 summarizesprogress and proposalsin Chile, Brazil, and Venezuelain relation to the above. Chilehas gone the farthest. It has diversifiedthe sourcesof funding,throughincreasedprivate sourcesof financing,and has devisedfundingmechanisms which explicitlyencouragenationalgoalsof efficiency,quality, equity, and relevance. The Chilean Governmenthas moved away from direct involvement at thesametimeprovidinga coherentframework institutions, in the affairsof individual with clear roles and responsibilities and with a variety of incentives linkedin a systemof publicaccountability for imProvedquality and efficieny. Chile sil needs to developfunding mechansms to encourage internalefficiencyin publiclyfunded institutionsand to improvequality in the private sector through programs of accreditationand assessment. It also needs to strengthenand redirect the loan program. Nonetheless,becauseof its reformprogramChile's highereducationsystemis arguablythe most stable and effectivein the region. Brazil's proposedreformprogramis at least as ambitiousas Chile's was. It goesfurtherthan Chile by proposing to set up a sophisticated funding formula for public institutionsexplicitlydesignedto encourageincreasedefficiency and to lower unit costs, by proposingincomecontingent loans and to use the loan schemeto encouragequalityin privateinstitutions,and by settingout to deregulateprofessions. Similarlyto Chilethe Brazilianreformseeks to link financingreformto expansionof accessto higher education(in this case, to the publicsystem). Whilethere havebeensetbacks,criticalareas of Brazilian societyappear to be reachinga consensuson the need for reformsof this type. The key step will be to reinforcethe technicalcapacityof MEC and to reduceits traditionalpoliticalpatronagerole. Venezuelahas a more difficulttask ahead than Brazil in implementing reform. It begins with a public dominant,tuition free system, enrollinga very high proportionof the school age cohort, leaving little leewayin terms of increasingaccessas a meansof ameliorating any cost recoverymeasures. Chart I. [MGHER EDUCATION REFORM IN CHILE, BRAZIL AND VENEZUELA Summary of Polides POLICY AREtA CMILE. :-AZIL VENEZELA Before Reform Alter Reform BeforeReform Proposed Refonrm BeforeReform Financing Sour-es Tuitionin Public None 28%of expenditure. None Proposed, nofiguregiven None Institutions Enrollment in Private Institutions None 52%of enmrment 60%of enollment No change expected 22%of enrollment Fnaing Meas DirectInsdtutional Negotiated Baend qn perstudent Negotiated Ponnula-based to encourage Negotiated Support estimates _ and efficiency quality Lan Schemes None Largeoveaxge, but lfrge In operation, but highly Reduced subsidy, expanded Minimum covergc,highly deraults subsidized coverage, to beused to subsidized encourage qualityin privte institutions Research Support Non. Open competition and Directinstitutional support Nochange Directinstitutional support and peerbased awards andaboopen competition alsoopen competition Diversty LOW High Highbut few short coures Nochange High in technology Atoomy of Publi LOw High: institutions ully Low government controls High: institutions would Some autonomy in budgeting but instItutions manage theirown laries,budget kvels, and determinesalaries,working teacher salaries, condkions of budgets enroUments conditionsandenrollments work,andenrollments amr centray dctermined Acuntabiliy of Public Little Evaluationsystem to be None Funding formula wil reward Little InstItutions sd up. Somc funding cfficiency and additional based on qualiy of funding will bemade enteringstudents available on thebasis of evaluations institutional Qualty Improvement of None Accreditation system None. Tuitioncontrols Loan chemeandevaluation None, except for inhial PrivateInstitutions operating andbeing andloan cheme system to encourage qualiy recognition improved discoragipr qualiqt Credentlas Professions highly deregulation Somc Professions highly Deregulation proposed highlyregulated Professions regulated reported regulted -1-15 Nonetheless evenin Venezuela the long term prognosisis hopeful. Leadersin Venezuela,as well as in Chile and to some extent in Brazil have shown considerable talents in implementing radical economic reforms. The task now is to direct thesesame skillstowardsthe socialsectors. The pathto highereducation reformliesthroughdialogueand negotiation,ratherthanGovernment decreeand imposition,as wasthe casein Chilein the early 1980's. To achievereformon this basiswill requirea firmgroundingin highereducation factsand figures,and a nationaldebateon highereducation reform, throughthe newspapers, radio and television,conferences and reports,designedto mobilizekey opinionleaders. This is alreadyhappeningon a large scale in Brazil and Chile. Internationalsupport maywellbe helpfulhere as leadersof othereducational institutions informand describetheir experiences. The key stake-holders-students, teachers,businessmen, politicians,willneedto be convinced that reform is to their long term benefit;and importantpotentialstake-holders who havebeensilent up to now,such as primaryand secondaryschoolteachersand parentswhotheoretically might benefitfrom cost savings in higher education,wilLneedto be broughtintothe picture. Reform efforts will need to be structuredso that all parties feel that they gain something. For exampletuition fee increaseswillneedto accompanied by at leastpartiallyoffsetting programs, suchas well managed, expandedstudentloan and scholarshipprograms, as well as increasedaccess to high education,and a large portionof tuition revenueat least initiallycouldremain withinhighereducation institutions. New fundingformulaswill needto includeexplicitfinancialrewardsto the most efficient and productive institutions. Overallreforms in highereducationsuch as those descnrbed abovewouldresult in gains for all elementsof society. Government couldgainthroughreducingits financialburden and redirectingfunds on the basis of equity and efficiency. The academiccommunitycould gain through improvementin conditionsof work and throughrewardsfor quality. Studentscouldgain through increasedaccessto a varietyof highereducation institutionsthroughexpanded loanschemes.Finallysocietyas a wholewould gain throughmore agile and differentiated institutionsable to respondto changingeconomicand social needs. As keyopinionleadersbecomeincreasingly awareof thesepotentialgains, the issuein Venezuela and Brazil, as perhapsin all of Latin America,will no longerbe whetherreformwill come, but when. HIGHER EDUCATION IN CHILE EFFECTS OF THE 1980 REFORM Jose Joaqu(n Brunner and Guillermo Briones Table of Contents INTRODUCTION....................................... 11-1 THE OBJECTIVES OF THE 1980REFORM ...................... II- 1 of the Refoms ..................... The Scopeand Objectives Il- 2 SYSTEMLEVELEFFECTSOF THE 1980REFORM ................ Il- 5 From low to high systemdifferentiation ...................... 11-5 From a state subsidized selective system to open access according to achievement and income ........................ -6 From incrementalbudgetsto a competitivefundingmechanism .... .... I- 7 From state-corporative systemcoordination to market-policyregulations... 1-7 INSTITUTIONAL AND OPERATIONAL EFFECIS OF THE 1980REFORM . I- 8 Institutions . ....................................... 1- 8 Enrollment........................................ 1-12 Funding .......................................... 11 -13 Conclusions .. 1.................................... 1I-16 EFFECTS OF THE 1980 REFORM ON QUALITY, EFFICIENCY AND EQUITY .1....................................... .- 16 Quality .......................................... II-16 Efficiency......................................... II-23 Equity . 11-28 ........................................... REMAING ISSUESAND ACTIONS.......................... II-31 Further Reform ............ ......................... H-32 APPENDIXTABLES..................................... 11-35 l~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ E1-1 INTRODUCTION' Arguablythe mostradicalreformof highereducation wasimplemented in Chilebeginningin 1980. Basic featuresof ChileanHigher Education(HE) were drasticallyalteredby the 1980Reform, most notablythat in order to rapidly expandaccess, market oriented strategieswere appliedto the higher education system. An important featurewasto increasethe mobilization of privateresourcesthroughcost recoveryand the rapid promotion of private institutions. Giventhe pressureon manycountriesto adopt similar reform programs, this paper examines the impact of the reforms ten years after their implementation. This paper is dividedas follows.In the first part, the systemprior to reformis outlined,alongwith a synopticrecapitulation of the Reform's primaryobjectives. Second,the majorsystemlevel effectsof the Reform wDIbe examined, particularly in terms of the distribution of: establishments, their size, tier and sector affiliation, and regional distribution; enrollment growth and its distribution by tier, sector and fieldof study; and finallychanges in HEfiding, publicexpenditure and institutional financing. A third sectionwill examinethe issuesof quality,efficiencyand equityas a consequence of the 1980Reform. In this place a more detailedanalysis of outcomesis intendedbased on two principalsources:(i) official information and secondarydata as wellas the analysisof resultsobtainedfrom previousstudies;and (ii) new information on specificquality,efficiencyand equityissuesproducedon the basis of data analysis and a brief case studythat wasconductedfor this studyduringthe springterm of 1991. A final section brieflyoutlinesoutstanding issuesand plannedreforms in the upcomingyears. THE OBJECTIVES OF THE 1980REFORM Prior to 1980,Chile'sHigherEducation systemwasrelativelysmall,homogeneous and exclusively public.2 The system consistedof one tier and one sector, comprising8 universities: 2 state and 6 privateuniversities(3 Catholicand 3 non-confessional universities), all fundedby the publictreasury. Henceforth, these 8 institutionswill be referredas the 'traditional or 'old universities,followingthe local convention. Within each institution, the undergraduate level was clearlypredominant. In 1980, only I % of total enrollmentcorresponded to the postgraduatelevel (Table 1). All higher education institutions were legally recognized as self-govering bodies with no administrativedependence from Government. Universitieswere entitledto freely initiate schoolsand careers and extendprofessional titles and academicdegrees. But in 1973,shortlyafter comingto power, the Military Governmentdirectly intervenedin universityaffairs, imposingtight control over the universitiesand invalidatingtheir self-governmentpowers. In addition,faculty associationsand student a Paper prpared for the World Bank with the pl support of the RegionalProject on Higher Education Policies in LatinAmerican which is funded by the Ford Foundation,FacultadLatinoamericana de CienciasSociales(FLACSO), Josd JoaquinBrunnerand GuillermoBriones,Chile, January 1992. 2 For a more detailedoverviewsee Bnmner, JosEJoquin, Informo be la EducacionSuperor en Chio. FLACSO,Santiagode Chile, 1986 II-2 bodieswereprohibited.Universities whatwaseuphemistically underwent labelleda 'depuration"process. Highereducationinstitutionsreceivedpublicsupporton the basis of incrementalfunding. That is, the annualbudgets at institutionswere based on previousbudget allocationsand a distributionformula looselyarrangedaccordingto institutionalsize, enrollmentsand additionalcosts originatedby research and postgraduate programmes.In addition,no generalhighereducationlaw existed. But new universities had to be establishedwith the Government'sunderstanding, thus securingpublic fundingand the legal validityof their educationalcertificates. Free of charge HE was the norm but selective admissionwas enforced on the basis of the compound result of students scores in a national standardizedacademic test (Prueba de Aptud Acadimica, P.A.A.) and their secondaryschool performance. Table 1 Stistics of HigherEducation Summary in 1980 Numberof HE institutions 8 Enrolmat in 1980: 118,978 Stt univarsities - 63% uiiveraities Private 37% (withpublicfunding Postgradut enroment (PhD. and MA) 1,200 The Scope and Objectivesof thc Refonns The reforms implemented in December1980aimed at a radical overhallof the higher education system, to introduce market forces, to increase the efficiencyand responsivenessof institutionsto economicdemands. In addition,the system was to be expandedrapidly via the harnessingof private, rather than public resources. The reform was to be radicalin nature. Accordingto the Cerych& Sabattiermodel for analyzing HE reforms', there were five outstandingcharacteristics.The degreeof systemchangeenvisagedby the Reformwas high, affectingthe boundariesof the system itself as well as all institutions. Second,the scope of intendedchangeswas inclusive,affectingpracticallyall individualswithineach institutionand potentialactors outside the system. Third, the depthof the Reformitself led to the adoptionof radical policies. Policygoals stronglydivergedfrom existingvalues and practicesof HE. Fourth, the goals of the Reformwere clearly stated and had internal consistency.And finally, changeswere brought about by central edict, from top to bottom and throughbureaucraticenforcement. B.Clark's assertionon the feasibilityof authoritarianreforms in HE perfectly apply to the Chilean case: 'Of course, in systems 3 See Cerych, Iadislav & Sabattier,Paul, Great Ex2eafions andMixed EMfgn=ceT I lmetation of Higber EducationReformsin Europe, EuropeanInstiute of Educstionand SocialPolicy, Trendam Books, 1986 II-3 of of authorityand the centralconcentration under authoritarianor totalitarianrule, the centralization administrationallowmore manipulation from aboveand on a large scale'.' A brief caveatis here in order. Chile's HE systemwas reformedunder special and very harsh conditions,due to the existence MilitaryGovernment.Institutions of an authoritarian had to endurethe changesenvisagedby the reformerswithouthavingthe opportunityto intervenein the processnor to negotiateits outcomes. Authoritarian mindedreformerswillprobablysay that only becauseof this could system-wide changesbe broughtaboutand thatreformsneednot be assessed by the contextin whichthey took place nor by the political conditionsthat lay behindor made them possible. Reforms, they will argue, shouldonly be evaluatedin termsof their actualoutcomesand more general effectson society. In turn, thosewho experienced thesechangeswithoutanypossibilityof resistingor discussingtheir aims and implemenmtion procedures willprobablyfind it moredifficultto reflectuponthese changesandhave less confidence in the blessingsof supposedly rule. In short, one shouldnot forget efficient-authoritarian that under these special conditions,the cost of reforms must also includethe shatteredhopes and the repressionsufferedby thosewould not yield. Togetherwith the legislativeandpolicymeasuresthat were adoptedto directthe future of HE, the Military Government disclosedthe main purposesof the 1980Reform. Its primary objectivescan be summarized 5 as follows: To open-upaccessto HE. Deregulation was announcedas a wayof promoting'private initiative" in the organization of new institutions. Onlyminimli requirements for the creationof were established privateinstitutions.It wasstatedthat, from this momenton, expansionof HE enrollment shouldbe taken care of by private institutions. To dive,Yi the instiutions in the HE system. Three vertical tiers were established, based on a certificates: functionalhierarchyof educational UUniversities:expectedto focuson long cycleundergraduate programmesleadingto licenciaturas titles requiringa licenciado degree. Onlyuniversitiescan initiatepostgraduate and to professional programmes. * Professional Institutes (PIs): -are restrictedto four year programs leadingto professionaltitles definedas not requiringthe licenciatura. * programs TechnicalTraining Centers (ITCs): arerestrictedto shortcircle(twoyear) vocational leadingto technicalcertificates. 4 Clark,Burton, *The OrganiutionConception- in B.Clark(ed.), Pmrectives on HigherEducation.Eight and CoMparative DisciDlinarv Views, Universityof CaliforniaPress, 1987,p. 125 f is contained The 1980Reformlegislation LeyN° 3.541de 1980;DecretosconFuerr basicallyin: Docreto de Ley Nes 1,2,3, de 1980;Decetos con Fuer7 de Ley N°s 4,5,22 y 24 de 1981.For a detaileddiscussionsee Bnmner,JoseJoaquin,'La Educaci6n Superioren Chile:1960-1990. y Polftik* Q;LACSO, Evolucion manuscript) Jose Joaqin, Politic Pdblicasde Educ cin Superioren CUIdumnte el! 1990,and Cox, Cristiin & Brunmner, RegimenMilitar Geneaci6n,Ejecuci6n,Resultados- n0 ) 1 11-4 To reducethe instutionalpower ofthe two taditional state universities. Both the Universidad de Chile and the UniversidadTecnica del Estado OaterUniversidad de Santiagode Chile) were forced to give up their regional centerswhich -through a process of institutional amalgamation- gave birth to 12 new state universitiesand to two statePIs. From the Government's point of view, the contraction of these two universitieswasjustified as a way of rationalizing them and makingthem manageable. To paflaly trnsfer the cost of state-financedistitutions to the studentsand/or their families (cost recovery)and to force these institutionsto diversifytheir funding sources. In addition to cost recoveryat publiclyfunded institutions,the emphasison promotingprivatehigher education also shifted the burden of financeto private sources.New private institutions are dependenton resourcesobtained from tuitionfees. (in fact, incomederivedfrom tuitionfees coverspart of the capitalinvestmentin these institutionsalso). To increase competitionamong instdions, with the aim of promotingefficiencyand quality, incremental budgeting to institutions was replaced by a more competitive funding mechanism. Institutionalsupport(aportefiscal directo)was confinedto the 8 traditionaluniversitiesand the 14 new state-institutions.In all these establishments core fundingwouldnow have to be supplemented through institutionalself-financing: tuitionfees, competition for researchfunds, and contractfunding. A small amountof public fundingwas linkedto a 'best-studentsformula (aportefiscal indirecto), based on the number of the best 20,000 scorers in last years P.A.A. enrolled by each institution!' In addition, a Government financed student-loan scheme (crdditofiscal unlwrsitarlo, later designatedpagaff wwirstario) was introduced. Lastly, the civil-servant status of academicpersonnelpertaining to state institutionswas eliminated,thus enablinguniversitiesto differentiatesalaries and competefor faculty staff. In addition, a publicly financed National Research Fund (Fondo Nacdonalde Desarrollo entfficoy Tecnol6gico, FONDECYI) was set up. Researcherswere calleduponto contendfor funds in a yearly competitionunder peer-reviewcontrol. No public fundingwas visualizedfor new private institutions, but for one exception:as of year 1989, they can competefor "subsidy-carrying students" (the 27,500 best and brightest"). Students enrolled in (new)private institutionshaveno accessto subsidized loans. Finally, private gra1n to both private and state universitieswere stimulatedthrough specifictax deductions. Table 2 provides a summaryof these changes. Ihis figure was later broadenedto include the bea 27,500 scos. 11-5 Table 2 of HigherEducation Summary SystemBeforeand After Reform BEFORE 1980 AFTER 1980 1. One tier, one sector low institutional 1. Three tiers, two sectors: high institutional differontiation differentiation 2. No tuitionbut electiveacces accordingto 2. Tuition fees are chaged by all institutions. school performance and achievement in Selectiveacess is maintained in the subsector nationalstandardized test academic that rceives institutionalpublic fimding. In open accessaccording the private institutions: to and/or familyincome 3. State financing of HE on the basis of 3. sourcesof fimding Multiplecompetitive incrementalfimding 4. Systm coordination provided by State 4. Systemcoordination providedby marketsand authorityand institutional oligarchies, policy regulations SYSTiEM LEVEL EFFECTS OF THE 1980 REFORM At the systemlevel, the main outcomes of thepolicyand legislativemeasuresadoptedcanbe seen as a series of structuralshifts that have take place duringthe last 10 years. Fromlow to high system differendatio A relatively small scale 'HE system, with a total enrollmentof approximately119,000students distnrbuted in 8 universitieswith a commonfundingmechanism and a similar development pattern -modelled accordingto the ideal of all becoming'comprehensive universities' coveringa wide array of disciplineand/or profession-based faculties(facldrades)and both researchand teachingfunctions,has developedinto a systemwith a total enrollmentof about 250,000 studentsdistributedin 310 different establishments. lnstitutions wildly vary in size, function and core values, and each one adopts a developmentpattern adjusted to its mission-definition and market orientation. Thus institutional homogeneity has beenreplacedby systemheterogeneity. In 1990,Chile's HE systemincludedtwo major sectors (i.e., a public sector of 20 universitiesand 2 PIs with public institutionalfunding and a pbia sectorconsisting of new institutionswith no publicfunding,distnbutedalongthe threeverticaltiers).The growth in the numberof establishments by tier and sector is shown in Table 3. 111-6 Table 3 of HE institutions, 1980-1990 Chile:Development Institutions 1980 1985 1990 Universities 8 21 60 Withpublic fiunding 8 18 20 New private,nopublic funding - 3 40 Institutes Professional - 25 82 With public fumding - 6 2 New private, no public funding - 19 80 TechnicalTrniDngCenters - 102 168 With public funding New private, no public funding - 102 168 Total 8 148 310 Source: Divisi6nde EducacionSuperior,MINEDUC,1991 From a ste subsidizedseledive systemto open access accordingto achievementand income Chile is the first Latin Americancountryto have introduceda comprehensive and stringent cost recovery scheme for all HE students through tuition fees or the deferred payment of state subsidized loans. Whereas before 1980 students were admittedwithout charge to one of the 8 old universities accordingto their school performanceand achievement in the P.A.A. (nationalstandardizedacademic test introducedin 1967), after the 1980Reformall HE establishments, both public and private, charge fees that are expectedto cover the approximate direct cost of the givenprogram adjustedaccordingto market reputationof each institution. In practicalterms, the 22 publiclysupportedinstitutionscontinue to sustain selectiveadmissionprocedures.In addition, studentsare asked to pay tuition fees but those who can prove economicincapacityhave access to state subsidizedloans. 7 Studentsenrolled in the private sector will be admitted regardless of their school performance and achievementon the standardizedacademic test so long as they pay the full price charged by each institution.' The Governmentdoes not controlthe amountof fees chargedby the institutionsnor does it intervenein the process of fixingyearly adjustments. From incremental budgets to a competitivefunding mechanism The total amountassignedto the 8 old universities,expressedas a share of GNP, rose from 0.72% in 1960to 1.289% in 1970 and to 2.11% in 1972. After the adventof the Military Governmentand up 7 Only a ftaction of those studnts who request loans will get themn See Sanfuentes, Andr6s, 'Compormiento Universitario y Polfticasde Financiamiento,in C.1 mann (ed.), Financianientode la Eductci6n Superior Antecedentesv Deffos. Foro de la Educaci6nSuperior, CEP, Santiagode Chile, 1990. ' of increasing But a few of the new private institutionsare nowbecomingmor selectiveas a consequence demandand the competitionfor market reputation. II-7 to 1980,the same allocativemechanism was maintained,but with decreasingamounts. In 1980, HE fundingtotalled1.05%of GNP. Until 1980,Government allocations to the eighttraditionaluniversities were madeon the basisof incremental funding,that is, the previousyear's budgetplus someincrement. Suchbudgetsdid not necessarily relateto the activitiesat universitiesor their costs, nor were institutions givenincentivesfor efficiencyor quality. Afterthe 1980Reform,Government continuedto fall and were drasticallyrearranged. allocations corefundingwasrestrictedto the22 publicsectorinstitutions. Fromthat momenton institutional Overall, incrementalfundingwas replacedby four differentfundingmechanisms: * Publicinstitutional funding(aponefiscaldirecto),whichis confined universities to the 8 traditional and the 14 newstate institutions. But incrementalfundingas a wayof allocatinginstitutionalcore- fundingwas substituted by a formulaconsisting in diminishingpublicresourcesduring a set period of time and, after that, the levellingof that amountover time. * In addition,an increasingamountof publicresourceswastied to a best-studentsformula",based on the numberof the 27,500best scoresin lastyears P.A.A. enrolledby each institution(aprorte fiscal indirecto). * wereaskedto chargetuitionfees. A Government Publiclyfundedinstitutions financedstudent-loan schemewas introducedto supportcost-recovery (criditofiscal uniwvitario). * A separate Fund (Fondo Nadonal de Ciencia y Tecaologfa, FONDECYT)was establishedto financeresearchprojects chosenafter a peer reviewprocess on the basisof a yearly competition open to all researchers. The fundingof new private institutionswas made dependenton resourcesobtainedfrom tuition fees. HE institutions to seek supplementary were encouraged fundingbothfrom publicand privatesources throughvariousmeans such as conractual funding,incomefor servicesand from privatephilanthropy. to market-policy Fromstate-orporative system coordination regultions Traditionally,system coordinationwas providedby state authorityand institutionaloligarchies. In fact, as has been statedby B.Clark, nationalsystemsnot strictlyorganizedas state systemsare likely to dependheavilyon the manywaysthat academicoligarchycanlinkpersons, groups, and institutions. Wheninstitutionsare fundedmainlyby Government,academicsnormallyseek the privilegedautonomy of a direct and unfetteredlump-sumgrant from the nationaltreasuryto the individualinstitution(...) with past commitmentsand budgets guaranteeingand adequatesum'.' This was precisely the way ChileanHE worked and producedsystem coordination until the 1980Reform. The 8 old universities benefitedfrom privilegedautonomy - which came to an end in its self-governing featuresafter the military intervention. In addition, they had receivedpermanentGovernmentsupport through yearly lump-sumgrants, and producedtheir own self-administered ways and meansof coordination, the most importantbeing the Consejo de Rectores de las UniversidadesChilenas (Councilof Rectors)chaired by the Rectorof the oldestinstitution. E Clark,Burton,p.140-41 II-8 The 1980Reformradicallyalteredthese arrangements. Systemcoordination was to be provided mainly by a social choice context 'in which there are no inclusive goals, and decisions are made independently by autonomous organizations"'°whichwere calledupon to competefor students,faculty and resources. Market-orientation becomethe catch word that was presumedto best capture the practical philosophyof the 1980 Reform, both for friends and foes alike. Additionally, system coordination was to come from regulationsprovidedby policy orientationsand the allocationof public money. Thus institutionalleverageand the academic-oligarchic power of professors, deans and rectors were both curtailedby the Reform,with the expectationof changingthe system's self-centerdness (its monopolistic features,as was claimedby the MilitaryGovernment)and forcingits openingto society's demands as expressed through markets and diversified funding opportunities. In this context, "competitionis seen as leadingto value for moneyfor the consumerand a diversityof productsto suit individualdemands. In broad terms, the publicsectoris seen as wasteful,inefficient,and unproductive, whilethe privatesector is seen as efficient,effectiveand responsiveto the rapid changesthat are needed in the modemworld'."' INSTrITTONAL AND OPERATIONALEFFECIS OF THE 1980 REFORM This sectionexamineschangesin the main functionalvariablesof the ChileanHE systemduring the last decadeas a resultof the 1980Reform. Modifications,mainlyof a quantitative nature, produced in the followingsystem components will be analyzed: institutions, their size, tier and sectoraffiliation, and regionaldistribution; enrollment growth and its distributionby tier, sector and field of study; and finally changes in HE funding, public expenditure and institonal fnding. The following section will examinethe impacton quality(includingteachingstaff and research),efficiencyand equity. Insitutions As a result of the reforms, the numberof higher educationinsdtutionshas rapidlyproliferated. Both the breaking up of the two largest public institutionsand the encouragementof new private education have significantlyincreasedthe supplyof higher education. A group of 12 new public (state) universitiesand 4 PIs was createdout of the "rationalization'imposedon the Universidad de Chile and the UniversidadTecnica del Estado,lhe 2 old State-universities.' Most of thesenew Stateinstitutions are locatedin major cities outsideSantiago,and their creationas autonomous institutionswas supported by the local communities. '° bid, p.l137 " Walford, Geoffrey, -ChangingReationship betweenGoverment and Higher Educationin Britain', in G.NeaveandF.Van Vught(eds.), PrometheusBound.TheChanginzRelationshin Betwee Govemment and Hither Educationin WesternEurove. PergamonPre, Oxford, 1991,p.169 12 Only two state-PIsremn The other two were icoported into preexisting universities. The Insitute Profesionalde Chilldnwas assmilae by the Univeraidaddel BEoBhoand the InstitutoProfesionl de Valdiviaby the UniversidadAustdl de Chile. n1-9 The foundingof new private institutionswas encouraged through a permissivelicensingsystem. In fact, private universitiescould be establishedwith the only obligationof offering at least three professional programsleadingto differentprofessionalcertificatesselectedout of twelvedefinedby law as previouslyrequiringthe obtentionof a licenciatura.Y To ensuresome qualitystandards,privateuniversitiesand PIs were requiredto submitto a special examination procedure. Once officiallyauthorizedby the Ministry of Education, the new private institutionmust sign an examination agreementwith either one of the traditionaluniversitiesor one of the new state institutions. This agreementtakesthe form of a private, fee-for-service,contractwithout any intervention by any publicbodynor furtherGovernment control.In accordance withthe examination agreement new privateinstitutions mustsubmittheir teaching programmes for approvalby the examining institution. During a periodof 5 to 10 years studentsof newprivate institutionsas well as the first five cohortsof graduatingstudentsmust take their final examinations with a committeecomposed by faculty members from both the new and the examininginstitution. After satisfactorycompletionof the examination period, the new privateinstitutionwill automaticallyobtainthe statusof a full autonomous university or PI. No additionalaccreditationand/or evaluationprocedures were established,the underlyingideabeingthat qualityand efficiency were to followfrom competitionfor students,teachers and researchers,and for public and privatefunds. By Governmental decision,the two old state-universities gave birth (eventually)to 14 additional (state) establishments. In addition, a total of 276 new private HE establishments have been officially recognized during the years 1980-1990. In effect,the shape of ChileanHE has changeddramatically, as shown in Diagram1. New institutionswere establishedmore rapidlyin Tiers II and HI (i.e., in the non-university levels of HE), in part becausethe foundingof new privateuniversitieswas submittedup to 1987to one additional prerequisite,involving a prior authorizationby the Minlsterio del Interior, that is, the Ministryin chargeof internalaffairs and security. This restriction, and not the sociallymore demanding expectations involvedin the settingup of this typeof institutions, explainsthe slowemergence of privateuniversitiesduring the first half of the last decadeand the abruptinstitutionalexplosion after 1987 (See above,Table 1). D This specificationwas later changedto the obligationof conferringone professionalcertificateout of a listof 15definedby law as previously requiringa Ucendatura. Thelatter correspondto the followingcaree law, economics, psychology, medicine, odontology,engineering,forestry engineering, agronomy, chemistry and pharmacy,veterinary,biochemistry,architecture, primaryeducation teacher, secondaryeducation teacher, special education teacher. II-lo Diagram 1 ChangedConfiguration of the ChileanHE System, 1980-1990: Tier and SectorDistributionof Establishments YEAR1980 YEAR 1990 SECTORand SUBSECTORS SECTORSand SUBSECTORS ( ) Number of insti- ( ) Number of institutions tutions PUBLIC PRIVATE 1.5. (20) 1.1 11.2 11.3 1.1 11.2 11.3 1.4 (2) (3) (3) (2) (3) (3) (40) TIER II II Universi- (12) ties 1.6. (22) TIER II 2.1. 2.2. Professional (2) (80) Institutes 3.1. TIER III (168) Technical Training Centers 1.1. Old state universities(2) and their offsprings:(12) new state universities) 1.2. Old private catholicuniversitieswith public funding (3) 1.3. Old non-confessional private universitieswith public funding(3) 1.4. New private universities(40) 1.5. Sum total of universitieswith publicfunding (20) 1.6. Sum total of HE establishments with publicfunding(22) 2.1. PIs with public funding(2) 2.2. New private PIs (80) 3.1. New private ITCs (156) 1-11 The rapid increasein the numberof HE institutions over a relativelyshortperiodof time had four major effects: * The ChileanHE systemhas turnedprivatedominantin the non-university levels and now is dual public/privateat the university level. Post-secondary, HE is now a (private)market non-university concernwith a wide supplyof differentstudy opportunities. * Establishments havegrownmore regionally dispersedthus enlargingaccessopportunities over the whole of the country,particularlyin the non-university levels. Nonetheless,overalldistribution of enrollmentbetweenthe Metropolitan Regionand the rest of the countryhas changedin the directionof a still higherconcentrationin the capital city, which in 1983 accountedfor 49% of total enrollmentand in 1990for 51.2%. * Supply of HE is now providedby a large numberof small size private institutionsand a few medium-size establishments. AppendixTable 1 showsthe average enrollment by establishment in the different tiers and sectors. In the private universitysector, the three largest of the new establishments accountfor 56.3% of total sector's enrollment. * The rapid expansion of higher educationcapacity,however,may be exceedingthe demand. The latter phenomenon is most intenseat the non-university levelsbut can also be perceivedin the privateuniversitysector. Thus, whilein 1980the ratio of new openingsto first year enrollments in the 8 old universitieswas 100.0%in 1990the ratio for the whole system is 59.0%. That is, there are moreplacesavailablethanthere are first yearstudentsenterringthe system.But for each tier and sector, ratios differ: publicuniversities97.4%; private universities 69.6%; publicPIs, 96.5%; privatePIs 50.5% and ITCs 45.6%." Table 4 Chile: HE Supplyand Demand, 1983-1990 Variable 1983 1985 1990 Terminalenrollmn a the secondary level 122,210 137,795 132,018 Numberof studentswho take the PAA(1) 119,245 121,168 114,343 Newopenins suppliedby the HE system 72,765 116,564 162,794 1st year enrollments 64,037 71,766 96,089 Ratio of 1st year enrolls./ new openings 88.0 61.6 59.0 Source: Divisi6nde Educaci6nSuperior,MlNEDUC,1991 (1) Considen both High Schoolgraduatesfrom last years cohort and schoolleavensfrom previousyears who take the PAA . 14 SeeYanez,Jost, 'Antecedentes econ6micos del distema y financieros de educaci6nsuperior' (manuscript, 1991), Table I 1I-12 EnroUment HE enrollmentshave more than doubledduring the last decade. The entire expansionhas been provided by private sector institutions,mostly at the non-universitylevels. University enrollment increasedby 12.6%during the decade,exclusively by new privateuniversities. Table5 Chile:HE Enrolment by Tiers and Sectors 1980 1985 1990 Universities 116,962 118,079 131,702 With Public Funing 116,962 113,128 112,193 New Private, No Public Funding - 4,951 19,509 Profossinal Instiutes - 32,636 40,006 With PublicFunding - 18,071 6,472 New Private, No PublicFunding - 14,56S 33,S34 Tech Trining Centers - 50,425 77,774 With PublicFunding - - - New Private, No PublicFunding - 50,425 77,774 Total 116,962 201,140 249,482 Source: Divisi6nde Educaci6nSuperior, MINEDUC,1991 In 1990, 52.4% of total enrollmentwas providedby private institutionswith no public funding. But at the universitylevel, enrollmentin publiclyfunded institutionsclearly outweighsprivate sector enrollment. From a comparative point of view, the gross enrollmentratio in the ChileanHE system of 18.6% for the correspondingage group (20-24years) is similarto other Ladn Americancountriesand as high as some European countries. Chile's HE system is still in the process of massification,despite the andthe doublingof enrollment proliferationof establishments duringthe lastdecade.However,giventhat demand for places is for the most part being satisfied, further massificationwill not be possible until secondaryschool coverageis increased. Enrollmentdistributionby fieldof studyat theundergraduate level greatlyvariesbetweentiersand sectors. Table 3 presents the overall distributionat the systemlevel and for the different tiers. II-13 Table 6 Chile:Studentsby Field of StudyAccordingto of HE Institutions,1990 Tier Classification FIELD OF STUDY University PIs TTCs Agricultue 8.5 6.3 6.5 MedicalScienceandHealth 9.7 - 3.7 Engineering and Reltd. Tech. 28.6 16.1 29.9 NaturalSciencesand Mathematics 4.4 0.5 0.7 Social Sciences 11.3 17.3 2.8 CLommercwialand Bus. AdrnirL 7.0 24.6 46.9 LAw 6.4 - 0.9 Humanities 7.1 21.3 Educationand TeacherTraining 12.3 20.6 1.9 Arts and Architecture 4.7 11.0 5.3 Source:Divisi6nde Educaci6nSuperior,MINEDUC,1991 Publicinstitutionsenrollstudentsin all fieldsof study,but tend to concentrate on engineeringand related technologies,educationand teacher training, the social sciences, and commercial& business administration.During the last decade, there have been some major shifts in enrollmentdistribution withinthis subsector.On the onehand, publicinstitutions havewitnesseda declinein the area of teacher training, that waspartiallyabsorbedby Tier n institutions (PIs), and in the area of medicalsciencesand health related studies. On the other hand, these institutionshave expandedprograms in agricultural studies, naturalsciencesand mathematics, and law. Private HE institutionspossessa very different enrollmentdistributionprofile, specializingin courses that are relativelylow cost In Tier I institutions,law, social sciences and commercialand businessadministration careersaccountfor more than 80% of total enrollment. In the caseof PIs, the social scienceprogramsand teachertrainingaccountfor more thanhalf total enrollment. In the case of 7rCs, almost50 per cent of enrollmentis in the socialsciencesand a third in the technologicalfield. Funding Probablythe singlemost far reachingeffectof the 1980Reformhas been its impacton financing. The 1980Reformenvisagedtwo major changesrelatedto funding. First, there wouldbe an emphasis on mobilizingnon-government, particularlystudent,resourcesto financehigher education. This was of private institutionsto achievedvia tuition fees at publiclyfunded institutionsand the establishment expandaccess. Second, the allocationof publicresourceswouldbe on the basis of criteria that would stimulatequalityand efficiency. made the followingprojectionof the In accordancewith these changes,the MilitaryGovernment percentageallocationof publicfunds to HE (Table7). 11-14 Table 7 Chile: Projected Public Expenditurc in BE, 1980 onward Resources 1980 1981 1982 1983 1984 1985 1986 Institutioal support 100 100 90 75 60 S0 50 Best students relatod formula - - 10 25 40 S0 S0 Student loan scheme - 7 15 23 30 40 50 Total 100 107 115 123 130 140 150 Source: Decreto con Fuemz de Lay NO 4, 1981 Thus not only a change in the funding mechanismsof the publicly supported institutionswas envisagedbut also a net increaseof the total amountdisbursedby the public treasury, over a period of 50% more (in real terms)of the 6 years, after whichstate fundingwouldstabilizeat a level representing amount apportionedin 1980. Althoughthe proposedchangeswere rapidlyadoptedthe budgetprojectionsdid not materalize. In fact, Chile's economyexperienced a severe crisis in 1982,and duringthe followingyears fiscal policy and public expenditurewere highly restrictive. As a result public expenditurein HE followeda very differentpath than had been projectedin 1981 (Table8). On their part, public institutionshad to cope with diminishingStateresources and with a broadeningspectrumof new financingopportunities. Table 8 Chile: ActualPublicExpenditue in BE, 1980-1990 Resources 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 Institutionalsupport 100.0 96.0 86.2 59.6 S8.7 53.2 47.9 43.7 41.6 37.5 33.4 Best studentsrelated formulae 0.0 0.0 11.0 11.0 9.4 8.4 7.8 7.2 7.8 11.0 10.8 Studentloan ceme 0.0 7.0 14.7 18.3 22.7 18.9 17.7 16.2 15.6 12.5 9.4 FONDECYT 0.0 0.0 0.3 0.2 0.4 0.4 1.0 1.8 3.0 5.5 5.6 Developmet funds 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 4.1 1.9 0.0 Total 100 103 112.2 89.2 91.2 80.9 74.4 68.9 72.1 68.4 59.2 Source: C.Le .a (ed.). Financiamiento v Datos. CoJecci6n de ls Educaci6nSuneror Antecedentes Foro de ls Educaci6nSuperior, CEP, Santiagode Chile, 1990,p.92 Thus, while direct public allocationswere diminishing,the public-sectorestablishmentsincreased their income from other public and private sources. The net effecs of these changesare presented in Table 9, which shows the actual level and compositionof HE funding in Chile consideringthe whole system (i.e., public and private) and the various fundingsources. 11-15 Table 9 in HE by Sources, 1990 Chile:EstimatedTotal NationalExpenditure Source USS (million) Public sources Publictreasuryallocations a) to publicsector institutions' 125.6 b) to new privateinstitutions2 2.6 FONDECYT(assigned to HE) 11.8 Subtotal 140.0 Private sounrcs Privatephilanthropy a) goingto publicinstitutions 4.9 b) goingto new privateinstitutions 1.5 Tuitionfees a) Paid in publicinstitutions2 81.6 b) Paid in privateinstitutions" 77.4 Subtotal 165.4 Publiclmvatesources 5 Variousother incomes 159.4 Subtotal 159.4 Total 464.8 Source: Ministerio de Hacienda, Direcci6n de Prepueso, Balances and Pre-upuestarios, MINEDUC,Direccionde EducacionSuperior,1991 US$ 304,90 pesos. Inludesapoiiefiscal direcao,Vonefcisal indirecto, pagari urnuario and recpraci6n de pr&stamo por criddaoflcal. The apone directo rmpresents aound 62%, the aporte indirecto16%, the pagar6universitario 15% and the rest is recuperacidn de prdnwno por crAdito. 2 Share of aportefiscal indirectogoingto private universities. ' Estimatebasedon Appendix Table 10. 4 Estimate based on average cost of tuition fee for private sector institutionsaccording to tier (US$1000, US$800,US$400). Estimatebasedon AppendixTable 10. Includes:wnoade bienesy seriwos, renta de invetiones, yentade acaivos,transferencias del sectorprivadoy de entidadespabiUcas, endedudaniento, recaperacidn de otros prstanos, yei especiales y oowr ingresos. In 1990, total nationalexpenditurein HE representsthe equivalentof approximately1.67% of Chile's GNP. Around30% is directlydisbursed by thepublictreasury,36% comesdirectlyfromprivate mainlythroughthepaymentof tuitionfees, and the rest (34%)is mixedfundingoriginating contributions, II-16 from services,variousspecificpublicallocations,borrowing,ventade activos,internationalcooperation and private philantropy. Studentfees in 1990averagedaboutUSS700. AppendixTable 10 illustrates the sourcesof funding at the universities. Condusions Overall,the 1980reformsenableda doubling of enrollments in highereducation,primarilythrough a rapid increasein the numberof private, non-universityinstitutions.To achievethis expansion,a large portionof the financingburden of highereducationwas shiftedfrom the state, to the students and their families. Marketforces came into play in the allocationof state funds and competitionfor fee paying students. The nextsectionwill attemptto addressthe impactof such marketforces and expansion of the systemon the-quality,efficiencyand equityof the system. EFFECTS OF THE 1980 REFORM ON QUALITY, EFFICIENCY AND EQUlTY QU4aty Quality assessmentsin HE are difficultto come by in general. It is still harder to evaluatethe impact on quality of broad and in-depthchangesas those brought aboutby the 1980 Reform. In the absence of reliable quality measures, analysismust rest on circumstantialevidence and on indirect assessments.In the subsequentsectionsthe followingquality-relatedaspectswillbe analyzed: * ImPactof enrollmentextensionon public sectorinstitutions * Student's institutionand career choices 3 Faculty staff - Institutionalresearchcapacities * Provisionof postgraduatestudies * in the various sectors and sub-sectors Institutionalreputationof establishments On the whole, the 1980Reformdoes not seem to have negativelyaffected public institutions in terms of the qualitativeaspectsof their performance. Moreover, it might be arguedthat the policies adoptedpreventedthese institutionsfrom experiencing the more commonnegativeeffectsof enrollment massification. On the other hand, nothing seems to indicate that dramatic improvementsin quality resultedfrom the 1980Reform. Using studentchoiceas a proxyfor quality, one can beginto makejudgementsabout 'perceived' qualityshifts withinthe systemas a whole. While such indicatorsmay reflecthow one institutionstands relativeto another, such informationrevealslittle abouthow the qualityof the systemas a whole has been affectedby the reforms. The demandfor accessto the elite, selective,universityportion of the system has remainedfairly constantsince the reforms. Whereasin 1980the ratio of new openingsto effective demand was 0.54%, it was 0.64% in 1990. The relative stabilityof demandover time suggeststhat II-17 potentialcandidates with less than fair chancesto be admittedin the moreselectivelevel of institutions are now optingfor private universities and institutions. Moreover,candidates whoused to select second choiceprogramsin the old and more prestigious universitiesbecausethey were not admittedto one of the moreselectivecareers,nowhavethe optionof stilltakingtheirpreferredcourseof studyin a private university.In short, nothingseems to warrantthe contentionthat public institutionsare perceivedby students as havingdeterioratedin quality and would thereforebe redirectingtheir demandto private sectorinstitutions.It is muchmore likelythat demandis steadyin the more selectiveinstitutional sector becausesupply of openingsis not changingin this sector whileat the same time selectiveadmission procedureshavebeen maintained,whereasnew privateinstitutionshavegenerateda significantamount of new HE opportunities, particularlyin those professionalcoursesof study that commandthe highest prestigeand expectations of private returns. On the other hand, an importantnumberof young adults- with non-elitesecondarytrainingor comingfrom the vocational branch of secondaryschools-are now choosingto enter PIs and TTCs.Y An importantindicatorof quality withinthe new system is the institutionalchoice by the best students. In fact, the new fundingsystem rewardsthose institutionsable to attract the best students. Indirectly,quality can be associatedwith the enrollmentdistributionof these 27,500 subsidy-carrying students. In the years 1990 and 1991,public institutionscaptured87,8% and 88,9% of the total sum allotted for this purpose. Appendixtable 2 presents the distributionof these funds betweenpublic institutionsand comparestheir respectiveshare withtheir participationin total first year enrollment. The only conclusion to be drawnup to here is that after 10 years of the launchingof the Reform, public institutions,particularlythe 8 traditionaluniversitiesand a group withinthose derivedfrom the reorganization of the 2 traditionalstateuniversities tendto fare betterthanprivate institutionson selected items that can be used to indirectlyassessquality of undergraduate teaching. This can not come as a surprise, consideringthe very recent creationof privateuniversitiesand the well establishedreputation of the older and publiclysupportedinstitutions.Moresurprising- at leastby Latin Americanstandards- - is that a system-wide Reformset in motionand implemented in a period duringwhich state spending on HE was decliningdid not demolishthe public sectorinstitutions,which in generalhave managedto maintaintheir standingand reputation. One indicatorof quality of the system as a whole is the level of staff professionalization.The percentageof faculty holdinggraduatedegrees(PH.D and MA degrees)increasedin the public sector institutionsfrom 19.3% to 23.3% during the period 1981-1988,which again can be taken as indirect evidenceof qualityimprovement,particularlyif one considersthat student/teacher ratios, alreadylow, were at the same time further decreasing(Table 10). " For a detailedanalysis, see Muga Alfonso and Rojas, Fenando. 'Andlisis dcl Prooeso Regular de Admisi6na las UniversidadesChilenas, Perfodo 1982 a 1989', in Lmnitre, Maria Jose (ed.), IA Educacion Superioren Chile: Un Sistemaen Transici6n. Colecci6nForo de la EducacidnSuperior,CPU, Santiagode Chile, 1990 11-18 Table 10 Chile:Public SectorInstitutions, Total Numberof Academic Staff, Percentageof Postgrduate DegreeHoldersand Student/Teacher Ratios, 1981-1987 TotalNOof S of Postgrad Year acadenic staff degree holders SrI ratio 1981 12,833 19.6 8.01 1985 16,683 22.8 7.64 1987 17,655 24.0 7.10 -Source: Consejode Rectores,AnuariosEstadfsticos, Santiagode Chile, 1982a 1987 In public sector institutions,about half of the academicstaff is employedon a full time basis, while in the private sector universitiesno more than an estimatedaverage of 3% to 5% of the total academicstaff is hired in the same condition,the latterfrequentlybeing engagedin authoritypositions withinthe institution.Estimatedstudent/teacher ratiosin the leadingnew privateuniversitiesarenot very different from those found in the public universities,with an average of 11/1 in the two major private universities.16 Appendix tables 5 and 6 examine in more detail staff professionalization at different institutions. Another indicator of quality is the research output of the higher educationsystem. Such an indicator,however, only reflects what is happeningat the elite institutionsin which research activities registeredscientificpublications, are concentrated. In terms of internationally the majorstate university (Universidadde Chile) producesaround50% of thetotaluniversityoutput,followedby the Universidad Cat6lica de Chile (25%), and thetwo leadingpublicregionaluniversities(UniversidadAustral de Chile and Universidad de Concepci6n)eachwith around 8% of the total output."' Amongstthose researchers actively engaged in the presentationof proposals to the yearly competitionof FONDECYT (2,137 researchers), during the period 1982-1989,42% belong to the Universidad de Chile, 24% to the Universidad Cat61icade Chile and around6%, in each case, to the Universidad Austral de Chlle and the Universidad de Concepci6n. The other four old universitiesaccountfor an additional11% of the active researchers. The remaining 10% is distributed between researchers of the 14 new state institutions." Table II shows the percentage participationof the various HE institutions in the distributionof projects and resourcesassignedby FONDECYTover the 1982-1990 period. t In these two universities, full time teachers are 5% and 10% of total faculty staff, respecively. Administrativepersonnelstands in a ratio of around0,25 to one faculty,and there are around 45 students per one administrtive staff member,without considering junior staff (auxiliares). 17 Manuel,op it. See Krmuskopf, " Cientflica See Saavedm,Francisoy Vergar, Patricio,RecursosHumanosea Lvestig&ci6n v Tecnol6ica. Sggirtici;scifn en FONDECYT,CONICYT,SantiaSode Chile, 1989 1-19 Table 11 Chile: Percentage Participation of the HE Institutions in the Distribution of Projects and Resources Assigned by FONDECYT, over the 1982-1990 Period Approved % of allocated Institution projects resourca Number% U.de Chile 1,038 40.1 40.2 U.Cat6lica de Chile 635 24.5 23.0 U.de Concepcidn 156 6.0 6.9 U.Austral de Chile 139 5.4 6.8 U.de Santiago de Chile 106 4.1 4.3 U.Cat6lica de Valparaso 90 3.5 3.6 UTFederico Sta.Mada 79 3.1 2.9 U.Cat6lica del Norte 16 0.6 0.5 New state universities 161 6.2 5.9 Other HE instiutions 17 0.7 0.8 Other institutions and individual researchers 152 5.8 5.1 TOTAL 2,589 100.0 100.0 del FondoNacionalde DesarrolloCientlficoy TOcnolItico.Peifodo Source: BsD n F.Saavedr, EEstadsticas 1982-1991-,CONICYT,Santiagode Chile, 1991 Considersyears 1982to 1991,raourc alrady allocatedor set aside for approvedprojects. Researchproductivity meassuredin accordancewith the science-publicationdataprovidedby the Institutefor ScientificInformationshowsa regularincreaseduring the latest comparableyears and puts Chile steadilyamongstthe 5 most productiveLatin Americancountries(Table 12). Table U Chile and Selected Latin American Countries: Number of Scientific Publications, 1986-1989 Country 1986 1987 1988 1989 Argentina 1,674 1,688 1,648 1,718 Brzil 2,001 2,083 2,193 2,556 Chile 815 799 841 901 Mexico 1,022 1,139 1,061 1,270 Venezuela 382 375 362 412 Source: CONICYT,Op.cit. In 1988,82.3% of Chileaninternationaly publications registeredscientific originatedwithinthe HE system. The 2 oldest universities,that is, the Universidad de Chile and the Pontificia Universidad 11-20 Catolica de Chile, contributedrespectivelywith 45.6% and 24.4% of all HE institutionsoriginated publications.In turn, the 8 traditionalstate supporteduniversitiesaccount in 1988 for 94.2% of the scientificpublicationsproducedby the HE system." Althoughthe Chilean researchcommunityis small and beavily concentratedin a few of the old universities,and public expenditure on R & D is just about average in the Latin American region, its performanceand comparativequality have been assessedpositively. A 1988 report prepared for the NationalScienceFoundation,the NationalAeronautics and SpaceAdministration and the Departmentof Energyof the UnitedStates,declaresthat althoughChile is the leastpopulousof the five major scientific countries in Latin America, it had twice as many papers as fourth-rankedVenezuelain 1985. 'On a population-weighted basis, Chilewasmuchthemost productiveof the five, evendiscounting international activity in astronomy.In astronomy,the geosciences,and agronomy,Chile was active in a higher proportionof rapidlydevelopingareas than average'. The report concludesthat Chile's profile in the internationalliterature is more like that of a small developednation than one that is attemptingnew industrializationefforts'. Appendixtable 4 illustratesthat this scientificoutput has occurred with a decliningpublic expenditurefor R&D. These and other favourableassessmentsof Chile's university-based scientific performanceare challengedby some recent evaluativestudies. For example,Zanelliand GarcIa argue that a country's positionwithinthe Latin Americanscientificcontextis not relevant, due to the meager overall regional scienceand technologyproductivity, which is 20 to 60 timeslower thanthat of developedcountriessuch as Belgium,Australiaor the Scandinavian countries. Moreover, Zanelliand Garcfashow that Chile is losing positioncomparedto more dynamicdevelopingcountriessuch as Nigeria,which already in 1985 was more productivethan Chile in termsof main streamscienceproduction,and other countriessuch as Taiwan, HongKong, SaudiArabia and Korea, which are rapidlyincreasingscientificoutput.' 'he provision of graduate studies can also be taken as an indirect indicator of university developmentand insdtutionalcapacity. According to Chileanlaw, only universitiescan conferdoctoral and masters' degrees.Presently,thesedegreesare grantedsolelyby publicsectorinstitutions,whilemost of 4th level enrollmentcan be found in the old universities.Total postgraduateenrollmenthas increased by 83.1% during the last decade, but still representsonly above2% of undergraduateenrollment. 123 master's programs and 21 doctoralprogramswere registeredin 1988(Table 13). t See Kmuskopf, Manuel, *Viaionde la Investigaci6nen Chile a travds de algunos Indicado Epistemomdtricos'(manuscript,n/d) D andTechnology Science PolicyProgram, 'NewDiectionsforU.S.A-.a SRIIntentionallWashington, American in Science Cooperation FinalReport,1988 and Technology', 23 See Zaneali, C., IA Jorge y Garcia,Martha v la Universidad,FIACSO, Santiao Ciencia.la Tocnoloefa de Chile,Contribuciones N.66, 1990 II-21 Table [3 Chile: Graduate in PublicSectorUniversities,1981-1990 Enrollment Year Number % of undergraduate enrollment 1981 1,432 1.4 1985 2,835 2.2 1990 2,622 2.2 Source: Consejode RectoresUniversidades Chiles, Boletfn Estadfstico, 1990. that possess As can be expected,almostall graduateenrollmentis in the group of old universities at least some researchcapacities. During the last ten years (1981-1991),a total of 3,231 peopleobtainedtheir degreesat the fourth level, over a total of 23,500studentsenrolledoverthe sameperiod. Computed on a yearlybasis, it took an average2350studentsto graduate323; a ratio of 7.2 studentsenrolledper graduate. Doctoraldegrees conferredduringtheperiodamountto around5% of thetotal degreesawarded. Newprivateuniversities have recentlyestablishedtheir first few MA programmes,up to now only in the social sciencesand education. The figures thus far analyzedmostlyrefer to publicsectoruniversities. Overallthey indicatethe existenceof a core group of "qualityuniversities', with a more or less developedresearch base and teachingprogrammesdistributedacrossvariousfieldsboth at the undergraduate and postgraduatelevels. Theseuniversities-amongst whichthe Universidadesde Chile, Cat6lica de Chile, de Concepci6nand Austral de Chile preeminently qualify-alsopossessa well established reputationbasedon their capacity competefor researchfunds and producemost part of nationally to attractthe best students, successfully originatedscientific publications,grant most of the more prestigiousprofessionaland postgraduate degrees, and on their abilityto maintainand increaseover time the numberof graduateholders within a predominantly tull-timebodyof faculty. Appendix table 3 presentsa summaryof some quality-related data for the 22 publicsectorinstitutions. The remaining groupof publicinstitutions,involving14 (new) state establishments and 4 old universities(one state universityand 3 private state-fundeduniversities) are more heterogeneous. Private sectorHE institutions,whichin 1990comprise40 universities,82 PIs and 168 ITCs can not be assessedthrough indirectquality indicatorsbecauseof the unavailabilityof public information. No systematicevaluationstudieshave been carried out on these institutions. Furthermore,their short periodof existencealso inhibitsqualityassessments.In the caseof newprivateuniversitiesthe following argumentsreferring to academic-quality issueshavebeen debated.' n SeeApablaza,V. andLavados,H., La Educaci6n v Persoectivas. SuocriorPrivadaen Chile. Antecedentes CPU, Santiagode Chile, 1988.Also, Lemaitre,MariaJosd (coord.),Informesobrela Educaci6nSuperioren Chile. 198. CPU, Santiao de Chile, 1988. AndBrunner,Jos6Joaqufn,InformesobreIs EducacifnSunerioren Chile. FLACSO, Santiagode Chile, 1986 11-22 Privateuniversitiesare said to be queuingbehindmost publicuniversitieshavingto acceptsecond- best students. The stereotypeis that new private universitieswould be restrictedonly to studentswho do not performwell in the P.A.A. but have sufficientmoneyto pay their way into HE. In fact, as was shown earlier,the best performersin the P.A.A. tend to go to traditionalpublic sectoruniversities;only a small fractiongo to privatesectorestablishments.Studentswantingto enterthe privatesectormust pay the full amountof the demandedfees since state fundedstudent-loans and scholarshipsare restrictedto public-sector students. In short, up to now private universities-withthe exceptionof a few programs- only attractsecond-best studentswho can pay the full cost of their studies. This distribution,as will be shownlater, reflectsthemore intensesocioeconomic selectivityof privateuniversities.On the otherhand, it mightalso contributeto explainthehigh dropout rates of theseuniversities. Accordingto the available information,inter-annualdrop out rates in the three oldestprivate universitiesvary over 1984 to 1987 between 20.6% and 19.5% in the case of the UniversidadDiego Portales,between 36.7% and 30.9% in the caseof the UniversidadCentral,and between28.59% and 30.2% (onlyyears 1984and 1985)in the case of the UniversidadGabrielaMistral. On the whole, then, the 1980 reform does not seem to have negativelyaffected the qualitative aspectsof the performanceof publicinstitutions. Moreover,it mightbe arguedthat the policiesadopted that year preventedthat these institutionsexperiencedthe more commonnegativeeffects of enrollment massification.A coregroupof four publicuniversities-thosethat originatemost research- canprobably be consideredas reasonablygood institutionswithin the Latin American context. A more detailed analysisof these institutionswouldprobablyshow a very unequaldevelopment of its differentsections and levels. For example,in most of them the socialsciences-bothin terms of researchand teaching- tend to be below level. In all of them, also, graduate studiesare still an area in need of development, both quantitatively and qualitatively.It is muchharderto assessthe quality of theremaininguniversities. With the informationat hand it is not possibleto draw qualitativedistinctionsbetweenthese different groupsof institutions,althoughthereseemsto exist,on the whole, somecorrelationbetweenthe sizeand complexityof these institutionsand their perceivedquality. Newprivate universitieshave adopted, up to now, a very differentdevelopment pattern comparedto state-supported institutions,in part precisely becauseof this differencein funding. Newprivate universitiesare solely teachingestablishments, with a high enrollmentconcentration in a few low costprograms,practicallyno full-timeteachers,with open access" of studentstied to their familyincome,and only minimalregulationsin terms of publicassurance of their qualityperformance,thoughthe lattertrait has been partiallycorrectedthroughthe introduction, in 1990,of a publicaccreditation system. Withregard to qualityevaluationof PIs and TrCs nothingcan be said at this stage due to a completelack of information. Efficiency Efficiency can be broken down into intermaland external efficiency. The study of internal efficiency in turn has two distinct but closely related aspects. Intra-system(and sectoral) efficiency considersdecision-making at the system level and allocationof resources amongthe various sectors and institutions. The second aspect of internal efficiencyis intra-institutional efficiency, which considers D Econ6micas See Sanfientes, Andrds, Polfticas an ls Universidad.ILADES-Georgetown University, Santiagode Cbile, 1990, p.16 9 II-23 decision-makingat the institutional level and the allocationof resourceswithinthe institutions.' The followinganalysisfocuseson internalefficiency. Intra-systemefficiencyis said to be lowin mostLatin AmericanHE systemsdue to the widespread employment of publicincremental fundingprocedures,' whichdo not provide incentivesfor improving quality and performanceon the part of institutions.On the contrary, it induces counter-productive organizationalbehaviors which result in over-dependenceon the state and lack of institutional accountability. As part of the reform,the governmentradicallyalteredthe meansof allocatingstate resourcesto institutions. Enrollmentdoubledwhilepubic allocationdecreased. The entire cost of HE expansion during the 80's was carried by the students enteringthe private sector institutionsand their families. Moreover,studentsenteringpublicsectorinstitutions also sharedthe cost of their training, contributing in 1990 with US$32for each US$100awardedby the Government to supportteachingin the various public sector establishments.21 Overall efficiencyin the use of public resources allocated to HE thereforedramaticallyimproved,withoutseverelyaffecting the qualityof public sectorinstitutions. Studentteacherand studentadministrative personnel ratiosgreatlydifferbetweenpublicsectorHE institutions. Considering nominalratiosfirst, differencesamountingto almostfive times existbetween the two universitieswith the lower (Universidad de Chile) and the highest (Universidad del Bio Bio) studentteacherratios. Not countingthe Universidad de Chile, the averagenominalstudentteacherratio in public institutionsis 9:1. If the Universidad de Chile is included,the averageratio dropsto around 7:1, sinceits ratio is 5:1. In termsof FTE teachers,ratiosdifferby a factorof four betweenthe lowest and highest, with an averageof 11:1 (see appendixtable 7). Overstaffing thereforeexiststo a similardegreein otherLatin Americanuniversities. For example, in Brazil, FTE student/faculty ratios both for Federal and Stateuniversities(1988)is 9:1, whichstill is about two times higher than the Universidad de Chile. lhe state universityof Sao Paulo (USP), the largest and consideredto be the best institution of HE in Brazil and one of the moreprestigiousin the Latin Americanregion, has a studentteacherratio of 10:1. UNICAMP,with the highestpercentageof enrollmentat graduate level (319%) has a studentteacherratio of 7:1. Also, Chile's average ratio for public sector institutionscomparesunfavorablywith other country's public sector ratios. Thus for example,in the USApublic 4 year institutionsshow a studentteacherratio of 17:1 in 1988. In brief, althoughsimilarto the nominalratiosfound in otherLatin Americancountries, Chile's averageratio is significantly lower than ratios found in North American,European,Asian and Oceaniccountries,with the exceptionof Norway, Portugaland Japan. The numberof studentsto administrative personnel' is > For these standarddistinctions and theirapplication to IAtinAmericanHE, sec Winkler,Donald,Hiibhe in LatinAmerica. Education andEauitv,World Issuesof Efficiency BankDiscussion 1990 Paper77,Washingtom, Z See Brunner,Jost Joaquin,Evaluaci6nv Financiiamiento de la Educacion Sugneor en AmericaLatina. Basesyara un Nuevo Contrato,CPU, Documentode Trabajo62/91, Santiagode Chile, 1991 X In fact, the actual figure of student'scontribution should be stil higher if the sum assignedto loan is discountedfrom the overallpublic allocaton and computed as deferrodexpenditure by the individuals. 27 Paradocentes,administrativos and auxiliares nI-24 also low, with an averageof 6.1:1, a figurethat is better than the very low ratios in FederalBrazilian universities(3.1 and 4.1 in the UFMGand UFSC) and lower than the ratiosof two BrazilianCatholic universities(11.1 and 14.1 in PUC-Rioand PUC-SP,respectively). Internal efficiencyof publicsector institutionsneedsto be assessedalso in terms of productivity. A study carried out in 1984 shows that on-time graduationrates in 12 university-onlybased career programsfluctuatebetween23% and 74% percent. Nine out of the 12 programshad less that 50% on time graduationrate (Table 14).3 Table 14 On-TimeGraduationRates by Field circa 1985 - Careea LAW 34 Agronomy 29 Architecture 47 Biochemistry 43 Dentstry 74 Enginering 28 Economics 39 Forstry 23 Vebeinary 47 Medicine 71 Puychology 73 Phamacy 30 Taking year 1985 (or around) as base, a comparativeefficiencyindexfor 8 Latin AmericanHE systemshas been computedso as to allowfor cross nationalcomparisons insideLatin Americaand with EuropeanHE systms. As shown in Table 15, total enrollments across a selectionof 8 Latin American systems were 4,218,769 and the graduateoutput497,621. In effect, these 8 systemsgraduate 11.8 per cent of their total enrollments.If, all other things being equal, this average output is taken to be the norm and that it constitutesan actual-rather than desirable-measureof efficiency,it is then possible to see how far effectiveoutputfor 1985compareswith what might be expectedif all HE systemshad an outputof 11.8%.1 In the case of Chile, the efficiencyindexis below the norm. Surprisinglyenough, the Latin Americansituationdoes not differ substantiallyfrom the situation of the nine university systems that were analyzedby G.Neave.In fact, expectedoutput in Europewas slightlybelow the LatinAmericannorm, at 11.3%. Andthe efficiencyindexvariedfrom 51.2 in Austria to 204.4 in the case of Great Britain. SeeJ.Ruz, I.Navarro & C.Aguilar, Cuademnos Conseiode Rectores.NO22, 1984 ' The efficiencyindex representsgraduateoutputx 100dividedby expeted output.The me formulaha been used by G.Neavefor 9 Europeanuniversitysystems.See Noeve, Guy. 'Editorial', G. Neave (ad.) Eurpa UniversitvSystems,Part I, CRE-Information, 3rd Quarter, 1986,p. 18 11-25 Table 15 Selected Latin American countries: Efficiency levels in graduate output, around 1985 Country Total enrollments Gmduate output Expected output Efficiency index Bnzil 1,479,397 2S3,553 174,495 145.3 Colombia 391,490 56,052 46,172 121.4 Costa Rica 63,771 4,908 7,522 65.2 Chile 197,437 20,256 23,288 87.0 Mexico 1,207,709 120,829 142,449 84.8 Peru 443,640 9,706 52,327 18.5 Uruguay 87,707 3,488 10,345 33.7 Venezuela 347,618 29,402 41,002 71.7 Source: Based on UNESCOF Statistical Yearbook, 1988 International comparisons only give very gross relative efficiency measurements because of the differences in systems, their internal differentiation and the length of studies in the various tiers and sectors. The above contrast, for example, could be disputed because it does not take into account the different mix of graduate output of the various systems according to levels (and therefore length) of careers. Using the same efficiency index to compare institutions within Chile yields results that are listed in Appendix Table 8. Total enrollments across the 22 establishments were 121,876 and graduate output 12,624. Overall, the public sector institutions graduate 10.4 per cent of their total enrollments. Graduate output is lower than expected in the case of 13 institutions, 5 of which belong to the group of the 8 old universities. The Universidad Catolica de Chile has the highest efficiency index across the old universities and the Universidad de Talca in the case of the new regional state-universides. Allowing for differences in the duration of the various careers, on time graduation for the whole sector is on average 38.8 per cent. The high input-output ratios in the public sector confirm the low levels of internal efficiency amongst these institutions. In theory, most university based programs take 5 years. But computed input- output ratios show that on average th figure per graduate is 9.2 years. Although this figure indicates a reatively high level of inefficiency, when compared to the still higher input-output ratios in other Latin American countries, as Venezuela for example, the Chilean public sector institutions stand out -in comparative terms- as 'fair performers'. In Venezuela only the Universidad Sim6n Bolivar, which has by far the best input-output ratio, favorably compares with the average ratio of the Chilean public sector institutions .° Over time, there has been a noticeable rise in input-output ratios over the last two decades (see Appendix Table 9 for the ratios for 24 institutions). At the end of the 1970's, the average input-output ratio for the 8 existing HE institutions was 7.1 years. Today, this figure is 9.2 student-years per graduate at system level and it stands at 10.1 years for the old universities (Table 16). 3° In the case of Univeradad Sim6n Bolivarthis figure is 9.13 years. See Reum, Feando, Mg Feasibility of Introducine Loan Scbemes to Finance Higher Education in Latin America. 'Me case of Venezuela'. Paper prpared for the IBRD, November 1990 II-26 Table 16 Chile: Old Universities, Input-Output Ratios in 1980 and in 1990 Institution(l) Average Avenge 110 Averge Avenge I/O enrollment Graduates ratio enrollment Graduates rtio (1986-1990) (1988-1990) 1990 (1976-1980) (1978-1980) 1980 U. de Chile 16,888 1,734 9.74 52,280 7,208 7.2 U. Cat6licade Chile 14,817 2,497 5.93 13,845 2,293 6.0 U. de Concepci6n 11,419 1,088 10.53 12,536 2,282 5.5 U. CatdlicaValprulo 6,932 878 7.90 6,832 759 9.0 U. T6cnicaF. SantaMara 5,653 547 10.33 3,537 326 10.8 U. Santiago de Chile 12,261 1,153 10.63 28,840 3,867 7.5 U. Austral de Chile 5,596 657 8.48 4,540 683 6.6 U. Cat6lica del Norte 2,969 149* 19.92 5,882 548 10.1 Total 76,535 17,966 10.05 128,307 17,966 7.1 Source: Consejode Rectores, Anuario 1980 and IM ' This figure needsto be checked. High input-outputratiosnot only reflectlevelsof inefficiency but also translateinto higherpublic expenditureper graduate. High public expenditure per studentand per graduate will also translateinto high unit costsper studentand per graduate.Computation of studentunitcosts andper graduate costmust include, in the Chilean case, both public and private expendituredirectlyrelated to teaching." Table 17 shows averagecostsper studentand per graduatein the publicsectorinstitutions. Leavingasidethe two PIs, studentunitcost arehighestin the UniversidadCat6lica de Chile and the Universidad de Chile and lowest in the Universidad de la Serena. Averageunit cost for the whole public sector is USS 1.700, well below the figures for equivalentinstitutionsin other countriesof the world. 3 Cost of teaching will be estimatedas the sum of expenditures incured by the public trey (port directo, indirecto, pagard universitario and recuperaci6ude crEdito fiscal universitario)and the expenditurm equivalentto the incomeoriginatedby paymentof tuitionfees. Unit costs will thus be computedas the averageof public and private expenditureper stdet anrolled at the undergraduatelevel. Unit costs of graduateswill be computedas sudent unit cost times the input-output ratiofor eachestablishment 1-27 Table 17 Chile: Average Estimated Unit Costs per Student and per Graduate in the Public Sector Institutions in US dollars, 1990. Institution Student Graduate unit cost unit cost U. de Chile 2,435.6 23,722.8 U. Cat6lica de Chile 2,582.4 15,313.6 U. de Concepci6n 1,613.1 16,985.9 U. Cat6lica Valpam{so 1,787.8 14,123.6 U. T. F. Santa Maria 1,400.5 14,467.2 U. de Santiago 1,454.6 15,462.4 U. Austml Chile 1,520.7 12,895.5 U. Cat6lica Norte 1,623.8 32,346.1 U. de Valpamaso 1,828.8 11,704.3 U. de Antofagasta 1,505.4 17,823.9 U. de la Seren 1,027.8 7,677.7 U. del Blo Blo 1,271.1 18,641.7 U. la Frontem 1,375.9 11,241.1 U. de Magallanes 1,994.6 13,423.7 U. de Talca 1,894.8 11,937.2 U. de Atacama 1,631.7 16,512.8 U. de Tampad 1,255.1 11,797.9 U. Aruro Pt 1,158.7 15,144.2 U. Mebopol.Cs.Ed. 1,431.7 15,147.4 U. Playa Ancha 1,192.3 I.P. de Santiago 877.2 16,465.0 I.P. Osorno 513.1 7,947.9 Average 1,699.7 15,637.2 Unit costsfor graduatesdiffermore widelyas they reflectthe computedinput-output ratiosfor the different institutions.The highestper graduateunit costs are found in two of the old universities,the Universidad del Norte and the Universidad de Chile. The lowest unit costs per graduateare to be foundin someof the newregionaluniversities(Universidadde la Serena, Universidadde la Frontera, Universidad de Valparaiso, Universidad de Tarapac and Universidad de Talca). But also one of the old universities(theUniversidadAustral de Chile)showsa graduateunit cost that is wellbelowthe sector's average. Equity The followingsectionexaminesavailabledata on to the socioeconomic status of studentsin higher education. The results of the various studies cannot be comparedto make any judgementsabout the impactof reformsover time becauseof the differingmethodologies and disparatesourcesof information they use. II-28 A study carried out in 1966 examinedthe socioeconomicprofile of students in the 8 old universities.' In general, 40.4% of students came from high and middle high socioeconomic backgrounds. A secondstudy, performedby G. Briones, comparesthe socioeconomic composition of students enteringthe 8 old universitiesin 1976and 1981.3 A reductionin the representation of studentscoming from familieswith low educationalattainment can be discerned(Table 18). Brionescontendsthat ther has been an elitizationof accessto the 11 only-university based coursesof study in the 8 old universities. Table 18 Chile:-Socioeconomic of University Background StudlntsEntering intothe 8 OldUniversities, in Years 1976and 1981Accordingto Father's Education.(percetage) Father'seducaion 1976 1981 Illiterate 0.8 0.5 lncompleteprmay 11.9 12.6 Completeprimary 17.0 13.0 Incomplete eondoay 21.9 17.7 Complete eondary 2S.0 28.7 Incomplete university S.9 5.8 Completeunivercity 14.S 18.1 Militarystdie 2.9 3.S Source: G.Briones,*LA educaci6n superioren el modelode la economfa nooibeosi' in PIlE, Las Inadrmaiones Educaconslesbpi elRimenMilitsr. PUE, Santiao de Chile, Vol.2, chapter 11. A further study by Gonzalez, Latorre & Magendzoshows the socioeconomicbackgroundof students enrolled in first year classified by tier and sector, according to father's educationin year 1984.' Table 19presents a summaryof the findings. A clear differenceof studentsbackgroundexists between public sector institutions,both universitiesand Pls, and new private institutions. While in the fist type of institutionsstudents comingfrom familieswhere father's educationis below the complete secondarylevel representmore than 40% of total first year enrollment,in the private sector institutions the samegroup is less than 15% in the case of universitiesand less than one third in the case of Pls. Moreover,the socioeconomic profileof studentsin the TTCs is on the whole not very differentfrom the profile of studentsin the publicuniversities. Thesefindingsconfirmthe overallstrong selectivecharacter 32 SeeE.Schiefelbeiny otros, Up Intentode Artlisis Globalde la UniversidadChilens.PLANDES,Boletfn Informntivo,NO28-29, 1968, p.40 D See Briones,Guillermo,'La educacionsuperioren el modelo de la economlanooliberal'. En PHE, Iu TransformacionesEducacionalesbaio el RemimenMilitar. PIIE, Santiagode Chile, Vol.2, cap. 11 34 See Gonzalez,Luis Enrique, Latorre, CarmenLuz and Magendzo,Abraham, Anilisis de ls incidenca sobre la educacidnpostsecundariachilenade los cambiosproducidospor Is reestructuracion universitarisdel afo 1981-. PIIE, Santiagode Chile, 1987 (mimeo),Table 3.2 I1-29 of the Chilean HE system. Appendix Table 11 confirms some of these findings for first year students using a different classification of SES. Table 19 Chile: SocioeconomicBackground of StudentsEnrolledin First Year Classified by Tier and Sector,Accordingto Father's Education,1984 Institution Diterate Primary Secdary Unversity Incomp. CompL Incomp. Compl. Incomp. Compi. Universities With public funtds 0.6 9.9 15.9 18.5 30.8 5.4 18.8 No public funds 0.1 0.9 1.9 7.6 31.9 19.5 38.2 Total 0.6 9.2 14.8 17.6 30.9 6.5 20.4 P. Institutes With public funds 3.0 26.4 20.6 12.5 20.8 5.8 11.1 No public funds 0.4 3.9 5.5 13.6 35.8 9.9 31.0 Total 0.9 8.6 8.7 13.3 32.6 9.1 26.7 TT Centers 0.6 11.4 15.4 17.4 35.3 5.1 14.8 Tota Population 40 yr. and more 14.0 42.0 10.7 23.8 5.5 2.2 1.8 Source: L. E. Gonalez, C. L Latorre,A. Magendzo,AnLliside la incidencia sobrela educaci6nposodcundaria chilena de los cambiosproducidospor la reestnrcturaci6n unversitaria del aDo 1981. PIIE, Santiago, Chile, 1987(mimeo),Table3.2 J. Castafeda has analyzed the redistributive impact of public educational expenditure for year 198 5 .M This study demonstrates that only 25% of direct institutional allocations benefit the poorer 60% within society. If one looks more specifically at the beneficiaries of student loan funds, 51,4% of the funding benefits these groups. Two additional studies reviewed by A. Sanfuentes - one based on the same 1985 figures and the second based on a special household survey carried out by J. Rodriguez- corroborate these results. In effect, they show that the top 20% income distribution group obtains around 53% of public expenditure allocated to HE, while the bottom 20% group only captures 6%MI These M Tarsicio,Pac Combatirla Pobrezn.PolfticaSocialv Descentralizaciften Chiledurante See Castafieda, los'80 CEP, Santiagode Chile, 1990, Chapter2, sectionIV. This analysisis basedon data providedby the CASENsurvey(1987). Althoughtheseresultsare very muchunderdispute,theyprovidegross estimationsthat can be used when interpretedcarefully. Y Sanfuentes, sobrela distribuci6n AndrEs, Antecedenteas del ingresoy gastosgubernamentales par atenuar la extrm pobren'. In RevistaEstudiosSociales,NO60, 1989,p.29 II-30 figures compareunfavorablywith the informationavailablefor other Latin Americancountries,with the exceptionof DominicanRepublic.' More recent household-survey data from the PlanningMinistry, can be used to construct ' an inequalityindexof the ratio of HE share to populationshare for each quintile(Table20). The value 1.0 shows equalitybetweenpopulationshare and enrollmentshare for a given group. Values below 1.0 suggest discrimination against the group. Values above 1.0 over-representationof the group in enrollments.Thus, youngadultsin the poorestgrouprepresentone fifthof the total 18-24age population but only less than 5 per cent of HE enrollments.At the other end, young adults in the upper income group represent a mere 13.5 per cent of the age populationand approximatelyone third of total enrollments. These figuresare similarto those of Latin America,but with a more pronouncedinequality indexin the case of the low status group and a mDlder indexwith respectto the high statusgroup. The socioeconomic backgroundof HE studentsenrolled in private institutionscan be estimated, althoughonly in very gross terms. While only 1.3% of the bottomquintileage-groupenrolls in private sector institutions,13.2% of the top group does. Thus an importantresult revealedby these surveysis that lower incomegroups are far less likely to enroll in private institutionsthan in public institutions. That is, poorer studentswho accessthe systemgo to publicinstitutions. Privateuniversitieshave a much more unequalrepresentationof the variousstatus groups withinsociety, as presentedin Tables 19 and 21. Table 20 Chile:Distnbutionof Enrollments by IncomeGroups,1990 Quintile Percentof total Percentof Inequality 18-24popuation enrollmts Index 1 20.9 4.2 0.2 2 23.5 6.6 0.3 3 23.0 9.4 0.4 4 19.1 15.9 0.8 5 13.5 32.0 2.4 '7 See Winkler,Donald, gyci, Table IV.6, p.62 X MIDEPLAN, 'AnOliais preliminarde la informaci6nde la encuestaCASEN1990: mddulo educacidn, (mimeo, 1991). As will be noticed, HE enrollment figure obtained by way of the CASE N survey do not correspondto the officialfigures providedby the Ministryof Education.In fact, theyare producedon the basisof a survey with only a very limited number of HE students acting as respondents. Nonetheless, they can be consideredas a gross approximationto student's socio-economic backgroundclassfied acoordingto per capita incomeof their families. II-31 Table 21 Chile:Percentageof 18-24Years per CapitaIncomeQuintiles Enrolledin HE and Percentage 1990. Enrolledin PrivateInstitutions, Quintile Enrolled Enrolled in HE in privateinsts. 1 4.2 1.3 2 6.6 3.0 3 9.4 4.2 4 15.9 7.0 5 32.0 13.2 TOTAL 12.0 5.1 Access opportunities to HE have been improving during the last decades, in parallel to the enlargement of secondaryeducation.Moreover,as notedearlier,the ratio of new openingsto first year enrollments has steadilyincreased,to a pointwhere in 1990the wholesystemsupplied1.7 openingsfor eachone studentenrolledin first year. Butthis latteroccurrence has moreto do withprivate institution's supplyside strategiesthan with the actuallevelsof equalityof the system.In fact, over-supplygrows in inverse proportionto the reputationand tier affiliationof establishments, thus allowingthe ratios of applicationsto admissionsto remain stable at the universitylevel whilethey decrease at the level of privatePls and TrCs. REMANING ISSUESAND ACIIONS restructured In sum, the reformsof 1980havedramatically highereducationin Chile. Enrollments havedoubledduringthe decadechieflythrougha proliferation institutions.The of private,non-university doublingof enrollmenthas been achievedat the sametime as public fundingfor higher educationhas been reduced. Theintroductionof marketforces-competitionfor students,and significant costrecoverymeasures- -does not seemto haveharmedqualityin the publicuniversities,the elitepart of the system. In fact, the measurestaken may be seen as havingpreventeddeteriorationthat has elsewherebeen associatedwith rapid expansion. On the whole, Chileboastssomeof the highestqualityuniversitiesin Latin America. There is concern,however,aboutthe qualityof the new privateinstitutions. A more puzzling aspectof the system is the apparentfailure to improvethe efficiencyof the system, especiallythe low outputof graduatesrelativeto the numberof students. Despitecompetition, there is no evidencethat staff are beingmore efficientlydeployedin public institutions. The relatively low student teacher and student staff ratios may be explained,in part, by the utilizationof staff for income generatingactivitiesand secondemployment in private institutions. In terms of equity, the reforms have expandedaccessoverall, but highereducationcontinuesto serve mainlymiddleand upper middleincomegroups. The lower incomegroupsdo not appearto have which mostlyserve upper incomegroupsthat benefitedfrom the rapid expansionof private institutions II-32 are unable to find places at the elite institutions. On the other hand, the moves towards cost recovery havelimitedthe extent of public subsidiesbenefitingupper incomegroups. FwrlherRcfonn A centralissue of concernthat remainsafter the reform, and is currentlybeing addressedunder furtherreforms in 1990and 1991, is the overall developmentof new institutionsand the mechanismfor ensuringqualityin new programs. Three issuesstand out: Nwnber and size of inswtutions. Out of 40 private universities,only 2 enroll 3000 studentsor more; the next3 enroll 1000studentsor more. These5 universitiesaccountfor 68% of the sector's total enrollment. Most of the other private universitieshave less than 500 students. (AppendixTable 1). Inadequateaccrediation of these insittions. Up to 1990, all new private universities,once officiallyauthorizedby the Ministryof Education,were accreditedthrough the examinationprocedure introducedin 1981. Theprocedureinvolves prior approbation of thenew institution'steachingprograms by a freely chosen public sector instittion. In each program, students and academic degrees or professional belongingto an examined certificatescandidates universitymust taketheir final examsunder a commissioncomposed by faculty members of both interveninguniversities. After satisfactory completionof the examination period, the new private institutionautomaticallygains its full autonomy. The examiination proceduresintroducedby the 1980Reformhavebeen criticizedmainlybecause: * supervision, they do not provideinstitutional * the relativeacademicweaknessof some of the examiningpublic institutions, * they do not providepublic assurancethat academicstandardsare beingachievedby the new institution, - they are based on a commercialarrangement betweenthe two interveninginstitutionsthus arrangements, being liableto extra-academic * there is a lack of public controlover the wholeprocess. Standards and professional status of academic stff. A HE Commissionset up by the Governmentin 1990voicedthe more commonapprehensions aboutthe qualityof theprivate universities teaching bodies: 'absence of objective,merit-basedcriteria for the recruitmentof academicstaff; lack of a core group of full-timeacademicpersonnel;high turnoverrates of faculty;multipleemployment by the part of faculty members thus preventing the maturing of a sense of institutionalidentity and belonging;the fact thatnumerousprivateuniversityfacultymemberscomefrom the old public institutions which have to bear the cost of supportinga full time academicbody withoutobtainingvalue for money; the absence of an academic career structuredaccording to the needs of the new institutions;scarce attentiongivento academicpersonneldevelopment; the inadequateworkingconditions prevailingin some II-33 of the new institutions,and the lack of library, laboratoryfacilitiesand equipment supportfor academic work'.' Recentdevelopments in the accreditation ofprivate institutions. In March 1990,four days before the installation of the democratically electedGovernment a ConstitutionalLaw (i.e., a law that requires a special quorum for its reform) for the whole educationalsystem was passed by the Military Government. The larger part of the law refers to the HE system. Under the new law, a public, autonomousbody -the Higher Council of Education (HCE)- was set up, with two major functions: (1) to accreditnew privateuniversitiesand PIs, to approvethe core nationalcurriculumfor Cii) primary and secondaryeducation. The HCEhas ninemembers:the Minister of Education,whochairsthe Council;3 memberselected one by the state universities,one by the old and new fullyautonomous private universities,and one by the fully autonomous Pls; 3 memberselectedby the scientificcommunity;and two memberselected, respectively,one by the SupremeCourt and one by the Chiefs of Staff of the Armed Forces and the DirectorGeneralof the NationalPolice. The Councildesignates an Executive Secretarywhois in charge of day to day operationsand directsthe Council'sstaff. The HCE is financedthrougha specificannual National Budget allocationand is legally entitled to charge fees for its accreditationservices. The Council'sstaff is composed of 8 professionalsand 3 administrative employees. During its first year of operation,approximately 70 consultants were engagedin assessment activities. The operationalcost of the Council'sfirst year was US$ 85,000. The accreditation proceduresset up under the 1990Constitutional Law havethe followingmajor characteristics: * They providea licensingsystem for the establishment of new private universitiesand PIs centered-aroundthe approvalof both an institutionalproposaland the programsthat will be offered. Oncethesehavebeenapproved by the HCE, the institutionis officiallyrecognized by the Stateand can initiateits activities. - They also providea temporary Institutional assessmentsystemwherebythe development of each approvedinstitutionalprogramis periodicallyreviewed(once a year) and closely monitored,during a minimumperiodof six years, after whichthe HCE candeclarethe full autonomy of a private institution. * Under the regulationsof the 1990law, all universitiesand PIs that were createdafter 1981 and have not been grantedfull autonomyare free to choose betweenremainingwithinthe examinationsystemor submitting to the newlyestablishedaccreditation procedures. These proceduresare thereforeoptional for alreadyexistingprivate HE institutions. " Comisi6nde Estudiode la Educaci6nSuperior,Una Poiftica2ar el Desarrollode la Educaci6nSunerior en la D6cadade los Noventa', Santiagode Chile, 1991,p. 27 11-34 * Institutionsthat might be createdbeforeMarch 1992can decide betweenbeing accredited by the HCE or use the 1980examination procedures. After March 1992,new universities and PIs will have to be accreditedby the HCE. Out of the 40 existingprivate universities,21 have chosento be accreditedby the HCE. They comprise71% of all studentsenrolledin theprivate universitysector. Of the 80 existingprivatePls, 15 of which have alreadygainedtheir full autonomyunder the examinationregime, and 18 have chosento be accreditedby the HCE. They comprise33% of all studentsenrolledin private PIs. Z de Edu-cci6n, Superior See, Consejo Santiago Un Aiiode TTabaio, de Chile,1991 Il-35 APPENDIX TABLES Appendix Table 1 Chile: Average HE Enrollment by Establishments Establishments Number Average enrollment Universities W.p.f 20 5,610 New private 40 488 Private Institutes W.p.f. 2 3,236 New private 80 419 T. T. Centers W.p.f. _ New private 168 463 Source: Divisi6n de Educaci6n Superior, MINEDUC, 1991. n-36 AppendixTable 2 Chile: PublicSectorInstitutions,Their Share in First Year Enrollment and the Proportionof Student-Quality SubsidiesThey Obtain*, 1990 Institution % of 1st year % of participation enrollment in student-quality subsidy U. de Chile 10.25 19.26 P. U. Cat6licade Chile 11.67 15.30 U. de Concepci6n 9.66 11.29 U. CatdlicaValparaiso 7.24 8.31 U. TecnicaF. SantaMaria 5.40 5.03 U. Santiagode Chile 8.26 9.28 U. Australde Chile 5.05 4.75 U. Catolicadel Norte 3.53 2.21 U. de Valparalso 3.41 3.48 U. de Antofagasta 2.19 1.49 U. de la Serena -3.11 0.93 U. del BfoBfo 5.06 3.32 U. de la Frontera 3.27 2.78 U. de Magallanes 1.19 0.31 U. deTalca 1.91 1.94 U. de Atacama 1.25 0.40 U. de Tarapca 4.02 1.53 U. Arturo Prat 1.73 0.85 U. Metropolitana Cs.Ed. 4.48 2.46 U. de Playa Ancha Cs.Ed. 2.17 0.84 I.P. de Santiago 3.19 3.66 I.P. Osorno 1.95 0.59 Source: Consejode-Rectores,AnuarioEstadfstico,1990and Division de EducacidnSuperior, MINEDUC,1991. Only considersthe amountobtainedby publicinstitutions. n-37 Appendix Table 3 Chile: Selected Quality-Related Indicators for Public Sector HE Institutions, 1989-1990 Institution A B C D E F G H I J U. de Chile 16,926 14.7 19.3 3229 5.2 0.17 39.7 41.0 46.3 9.74 P. U. Cat6licade Chile 14,399 12.4 15.3 1166 12.3 0.43 22.1 24.5 22.2 5.93 U. de Concepci6n 11,192 9.6 11.3 1104 10.3 0.34 6.7 5.9 7.4 10.53 U. Cat6licaValparamso 6,737 5.8 8.3 479 14.1 0.46 3.8 3.6 3.3 7.90 U.T.F. SantaMa 5,486 4.7 5.0 255 21.5 0.25 2.9 2.9 2.2 10.33 U. Santiagode Chile 11,324 9.7 9.3 739 17.1 0.18 4.0 4.2 5.3 10.63 U. Austral de Chile 6,863 5.9 4.8 601 11.4 0.35 5.9 5.3 7.8 8.48 U. Cat6licadel Norte 3,013 2.6 2.2 218 13.8 0.40 1.0 0.6 0.3 19.92 U. de Valparaiso 3,001 2.6 3.5 324 9.3 0.11 1.0 0.5 0.9 6.40 U. de Antofagasta 2,757 2.4 1.5 202 13.6 0.34 1.1 0.9 0.7 11.84 U. de la Serena 3,017 2.6 0.9 203 14.9 0.25 1.0 0.7 0.9 7.47 U. del Bio Bfo 4,958 4.3 3.3 276 18.0 0.20 0.2 0.2 0.2 14.66 U. de la Frontem 4,360 3.8 2.8 362 12.0 0.18 1.0 0.8 1.1 8.17 U. de MaguLAnes 1,036 0.9 0.3 90 11.5 0.20 0.6 0.5 0.1 6.73 U. de Talca 2,933 2.5 1.9 14I- 19.8 0.39 0.3 0.3 0.5 6.30 U. de Atacama 1,436 1.2 0.4 97 14.8 0.36 0.4 0.3 0.0 10.12 U. de Tarapaci 3,894 3.3 1.5 307 12.7 0.30 0.9 0.6 0.3 9.40 U. Arturo Pnat 988 0.9 0.9 113 8.7 0.11 0.1 0.1 0.0 13.07 U. Metropolitana Cs. Ed. 4,158 3.6 2.5 366 11.4 0.23 0.5 0.6 0.0 10.58 U. de Plays AnchaCs. Ed.* 2,203 1.9 0.8 181 12.2 0.47 0.1 0.2 0.1 - I.P. de Santiago 3,399 2.9 3.7 236 14.4 0.06 0.2 0.1 0.0 18.77 I.P. Osorno 2,074 1.8 0.6 144 14.4 0.15 0.6 0.4 0.2 15.49 A: Enrollment,1989. Consejode Rectores.AnuarioEstadfstico.1989(*) 1988enrollment figures. B: % of total enrollment. C: X of participationin 1990best students-subsidy. SeeTable 18. D: Numberof FTE faculty. 1989.MINEDUC.Divisi6nde Educaci6n Supenor. E: Student/FTE teacherratio. 1989. F: Ph.D & MA/FTE teacher. 1989. 0: % of active reseahers in PONDECYT during period 1982-1989.Definedas those who participatedat leat onceas 'investigador responsable' or two timesas 'co-investigador'in approvedprojects. Percentageover total of active researchers.5.8% of these come from outsidethe consideredHE establishments.CONICYT.RecursosHumanosen Investieaci6n Cientfficav Tecnol6oica.op.cit. H: % of researchprojectsapprovedduringthe period.Percentageover total of approvedprojectsduringthe period.5.6% of approvedprojects como from outsidethe considered HE establiments. Sourcesme as in G. 1 S of publications withinHE institutions over total originiutded (85.0% of nationaltotal)for 1989registeredby the I.S.S. M. Krauskopf.gpRjj. Table V. J: Grmduates/enrollment ratio. 1986-1990. See below, Table9. 11-38 Appendix Table 4 Chile: R & D PublicExpenditureand Percentagein HE System, 1980-1988 Year USS (million) % GNP % to HE 1980 104.4 0.37 47.0 1984 87.0 0.45 44.6 1988 90.6 0.46 32.5 Source:Basedon CONICYT,Op. Cit. Table S Chile: Faculty Staff, NominalTime and FormalTrainingby Tiers and Sectors, 1986 %FT %HT SPT %PHD %MA %PT %NT IIniversities W.p.f. 59.1 7.3 33.6 7.6 12.5 78.9 1.0 private,n.p.f. 5.5 1.5 93.0 8.5 15.1 72.4 4.0 Prfessional institutes W.p.f. 50.3 12.3 37.4 0.5 9.0 84.3 6.1 Private,n.p.f. 9.7 7.0 83.3 2.2 9.3 80.2 8.3 Source: C. Cox and C. Jara, Datos BasicosDarala Discusi6nde Poifticasen Educacion(1970- I.2=, CIDE-FLACSO, Santiagode Chile, 1989. FT: full time HT: half time PT: part time PT: professionaltitle and/or licencatura NT: no professionaltitle 11-39 Appendix Table 6 Chile: Institutional Profile of the New State Universities, 1990 Institution Number Facultystaff Number of approved Major fieldsof stildO Number %of % of research enrollment concentration (1986) FT post- projects grads. (1988- 1st. 9 2nd. % holders 1990) U. de Vparaso 3,391 639 36,3 7,7 5 Medic./health 34,5 Social Scs. 24,2 U. de Antofagasta 2,825 373 61,1 17,2 10 Technologies 43,2 Health 31,7 U. de la Seren 3,066 304 67,8 13,2 9 Technologies 42,1 Education 35,9 U. del Bfo Bfo 5,806 -495* 21,2 25,1 4 Technologies 46,0 SocialScs. 15,7 U. do la Frontera 4,487 573 50,3 10,1 11 Technologies 33,2 Mod./Health22,6 U. de Magallanes 1,100 192 44,8 0,8 9 Technologies 38,5 SocialScs. 28,3 U. de Talca 2,328 326 54,6 17,2 9 SocialScs. 37,6 Agriculture 22,7 U. do Atscams 1,458 144 54,9 13,2 3 Technologies 81,0 Education 16,9 U. de Tanpcd 4,274 468 58,3 22,9 14 Education 31,5 Tochnologs. 29,9 U. Arturo Pmt 1,206 164 62,8 0,9 - SocialScs. 52,0 Education 14,5 U.M. Cs. Ed. 4,462 485 71,8 11,3 10 Education 100,,0 - U.P.A. Cs. Ed. 2,071 357 58,3 11,2 2 Education 81,7 Humaities 15,1 I.P. de Santiago 4,221 414 29,0 13,0 - Technologies 49,4 SocialScs. 26,7 I.P. Osomo 4,217 202 63,4 12,4 10 Education 51,7 Agficulture 26,1 Souce: Based on Consejode RnctoresUniversidades Chiles, Anuanos Estadfsicos, 1986 and 1990 and F. Saavedra,op ct Does not considerfacultystaff of IP de Chillanwhichwas later aborbed by the U.del Bio Bio I1-40 Appendix Table 7 Institution s/r FrE/S S/AP Ratio Ratto Ratio UCH 3.1 5.3 2.7 PUC 6.7 9.8 6.5 UC 9.1 10.3 5.9 UCV 7.7 12.3 8.3 UTFSM 13.7 20.1 7.6 USACH 9.3 17.2 8.4 UACH 7.7 9.0 4.7 UCN 13.6 14.0 4.9 UV 4.3 7.7 5.4 UANTO 10.4 12.0 7.6 ULS 13.2 17.7 10.6 UBB 14.5 25.2 17.7 UFRO 8.2 12.4 9.4 UMA 6.2 10.2 9.4 UTAL 11.9 17.0 11.8 UATA 14.0 19.8 10.6 UTAR 9.6 13.4 7.5 UAP 6.8 11.5 8.1 UMCE 8.9 11.9 8.5 UPACE n.i n.i n.i IPS 9.2 18.2 15.4 IM 12.8 16.7 8.3 IPV 16.0 20.8 9.9 IPO 13.6 18.0 13.5 Average 7.1 10.8 6.1 Source: Consejo de Rectores. Anuario Estad(stico, 1987. 11-41 AppendixTable 8 Chile:PublicSectorHE Institutions EfficiencyLevels in GraduateOutput, 1990 Country Total Graduate Expected Efficiency enrollments output(2) output(3) index Total 121,876 12,624 10,358 100.0 U. de Chile 16,478 1,793 1,707 105.0 P. U. Catolicade Chile 14,566 2,243 1,509 148.6 U. de Concepcidn 11,692 995 1,211 82.2 U. Cat6licaValparalso 6,976 732 723 101.2 U. T6cnicaF. SantaMarfa 6,145 539 636 84.7 U. Santiagode Chile 11,251 1,111 1,165 95.4 U. Australde Chile 6,562 537 680 79.0 U. Catdlicadel Norte 3,291 140 341 41.1 U. de Valparaiso 3,391 275 351 78.3 U. de Antofagasta 2,825 208 293 71.0 U. de la Serena 3,066 347 216 109.8 U. del BfoBfo 5,806 680 601 113.1 U. de la Frontera 4,487 594 465 127.7 U. de Magallanes 1,100 133 114 116.7 U. de Talca 2,328 442 241 183.4 U. de Atacama 1,458 152 151 100.7 U. de Tarapaci 4,274 360 443 81.3 U. Arturo Prat 1,206 122 125 97.6 U. Metropolitana Cs, Ed. 4,462 358 462 77.5 U. de Playa Ancha Cs. Ed. 2,071 340 215 158.1 I.P. de Santiago 4,221 332 437 76.0 I.P. Osorno 4,217 191 437 43.7 Source: Consejo de Rectores. Anuario Estadfstico, 1990 and Divisidn de Educacidn Superior, MINEDUC.1991 n-42 Appendix Table 9 Chile: PublicSectorHE Institutions,Input-Output Ratios, 1990 Institution(1) Average Average 1/0 enrollment Graduates ratio (1986-1990) (1988-1990) U. de Chile 16,888 1,734 9.74 U. Cat6licade Chile 14,817 2,497 5.93 U. de Concepcion 11,419 1,088 10.53 U. Cat6licaValparafso 6,932 878 7.90 U. TecnicaF. SantaMarfa 5,653 547 10.33 U. Santiagode Chile 12,261 1,153 10.63 U. Austral de Chile 5,596 657 8.48 U. Cat6licadel Norte 2,969 149 19.92 U. de Valpara(so 2,986 471 6.40 U. de Antofagasta 2,813 239 11.84 U. de la Serena 3,510 470 7.47 U. del Blo Bfo(2) 3,576 244 14.66 U. de la Frontera 4,415 540 8.17 U. de Magallanes 1,137 169 6.73 U. de Talca 3,252 516 6.30 U. de Atacama 1,639 162 10.12 U. de Tarapaca 3,929 418 9.40 U. Arturo Prat(3) 1,281 98 13.07 U. Metropolitana Cs. Ed. 4,423 418 10.58 U. de Playa AnchaCs. Ed. n.i n.i I.P. de Santiago 3,698 197 18.77 I.P. Osorno 2,820 182 15.49 Source: Consejode Rectores. AnuaroEstadfstco, 1990. (1) I.P. de Chilln and I.P. de Valdiviaare not considered. (2) Enrollment,years 1983-1987; graduates,years 1985-1987. (3) Output figures presentsteep fluctuations. 1-43 Appendix Table 10 Chile: Income Composition of Public Sector HE Institutions, 1990 (in million pesos) Institution Total Income PublicTuition Fees Treasury Paymentsby Other Sources(3) Disbursements(l) Students(2) SS $ 5 $ S 9 U. de Chile 27,363.1 100 8,381.1 30.6 3,856.2 14.1 15,125.8 55.3 U. Cat6l. de Chile 28,175.9 100 5,676.0 20.1 5,792.9 20.6 16,707.0 59.3 U. de Concepci6n 8,252.0 100 3,954.0 47.9 1,861.0 22.6 2,437.0 29.5 U. Cat6l. Valpar. 5,378.6 100 2,164.2 40.2 1,638.4 30.5 1,576.0 29.3 U.T.F. SantaMarfa 2,812.6 100 1,918.2 68.2 705.8 25.1 188.6 6.7 U. de Santiago 6,732.9 100 3,047.5 45.3 1,924.5 28.6 1,760.9 26.1 U. Austral Chile 3,647.9 100 1,917.6 52.6 1,125.0 30.8 605.3 16.6 U. CatolicaNorte 3,616.2 100 1,172.4 32.4 457.0 12.6 1,986.8 54.9 U. de Valpamfso 2,351.1 100 1,079.8 45.9 811.1 34.5 460.2 19.6 U. de Antofagast 3,372.3 100 758.7 22.5 538.0 16.0 2,075.6 61.5 U. de la Serena 1,298.6 100 797.7- 61.4 163.2 12.6 337.7 26.0 U. del BfoBfo 2,737.4 100 1,128.9 41.2 1,121.4 41.0 487.1 17.8 U. a Frontera 2,755.4 100 1,014.6 36.8 867.8 31.5 873.0 31.7 U. de Magallanes 943.4 100 333.8 35.4 335.2 35.5 274.4 29.1 U. de Talca 1,620.3 100 728.1 44.9 616.9 38.1 275.3 17.0 U. de Atacama 1,136.6 100 450.3 39.6 275.1 24.2 411.2 36.2 U. de Tarapcd 2,167.7 100 1,084.0 50.0 551.6 25.4 532.1 24.5 U. AituroPrat(4) 1,114.7 100 211.4 19.0 214.7 19.2 688.6 61.8 U. Metropol.Cs. Ed. 2,712.7 100 1,127.5 41.6 820.3 30.2 764.9 28.2 U. Playa Ancha 1,016.0 100 396.7 39.0 356.2 35.1 262.2 25.8 I.P. de Santiago 1,653.8 100 563.1 34.0 565.9 34.2 542.8 31.7 I.P. Osomo 902.2 100 377.9 41.9 281.9 31.2 242.4 26.9 Total 111,761.4 100 38,283.5 34.3 24,880.1 22.2 48,614.9 43.5 USS 366.5 125.6 81.6 159.4 Direccionde Presupuesto, Source: Basedon Ministeriode Hacienda, 1990. BalancesPresupuestarios, (1) Considers aporte fiscaldirectow, *aportefiscal indirecto', *Pagareuniversitario and *recuperaci6n de credito fiscaluniversitario-. (2) Includesdirect paymentof tuitionfeesboth at the undergraduate and postgraduatelevel. (3) Includes service contractingwith public and private bodies, research funds, venta de activos, renta de inversiones,recuperacion de prdstamos,endeudamiento, donaciones, otros ingresosy disponibilidades an caja. (4) Incomefor tuitionfeeshas been estimated. 11-44 Appendix Table 11 Chile:Socioeconomic Background of StudentsEntering1st Year in Public and Private Universities,1990 Father's Occupaton PublicUniversities Private Universities Mgers, professionals entrepneurs, administrators, 16.6 33.4 Specized employees,tchnicia, niddle-size entrepreneurs 24.2 23.5 Clericaland saleswoikers 25.0 18.3 atisans Smallproprietorsand self-employed 15.0 14.6 Skilledvanual workers 9.8 5.1 manualworkes Semi and umskilled 5.8 2.8 Personal(domestic) srvices and housewifes 3.6 2.3 Total 100.0 100.0 Source: Uznivesdads Chilenas,Procesode Admisi6n.1989-1990. HIGHEREDUCATION IN BRAZIL: ISSUES AND EFFORTS AT REFORM Laurence Wolff, Douglas Albrecht and Alcyone Saliba Table of Contents OVERVIEWOF THE HIGHEREDUCATION SYSTEMIN BRAZIL .... m1-1 1..... Structureand History ............. ...................... 1-l Equityand ExternalEfficiency......... ..................... m-5 GraduateEducationand Research ............................ 11-7 PrivateHigher Education ................................. E-8 ISSUESIN HIGHEREDUCATON .............................. M-l1 Costs of Federaly FundedHigher Education ..... M-l I................ Govermnceand Management of Federl Institutions ................ M-14 Financingof Higher Education.............................. M-17 Relationshipsbetweenthe Federal Government and Private HigherEducation............................. m-21 HIGHEREDUCATION REFORM ............................... 1-24 The Need For Higher EducationReformin Bra2 .................. M-24 The Government's Higher Education Reform Program, September1991- August 1992 .................................... m-25 REFERENCES ............ ............................... IH-28 ANNEX 1. PROPOSEDFUNDINGFORMULAFOR BRAZILIAN FEDERAL HIGHEREDUCATION INSTlTUTIONS ... m-33 ANNEX2. LIST OF CONSULTANT REPORTS ...................... m-38 I OVERVIEW OF THE HIGHER EDUCATION SYSTEM IN BRAZIL Stnrcuure and History Higher education in Brazil currently enrolls about 1.6 million students, 97% of which attend undergraduate programs. As Chart 1 shows,enrollments have grownduring the last thirty years from less than 100,000 in 1960 to 1.6 million by 1989, which is equivalentto 12% of the school age populationand about 58% of secondaryschool enrollment. First-year undergraduateenrollmentis equivalent to 74% of secondaryschoolgraduatesin the previousyear. About60% of total enrollments are in private institutions,of whichone third are in private universities. Federal institutions,most of which are universities,accountfor 22% of enrollments. State institutionsenroll 13%, with the vast majorityin Sao Pauloand ParanA.Municipalinstitutions, also mainlyin the Southeast and South, enroll 5% of students. Enrollrent 1960-89 BRAZIL:Undergraduate univBities, of which 35 are hooTou"M federal, enrolling over 300,000 -6- _ students, 39 are private, enrolling 1400 ._ over 300,00 students, and 19 are state or municipal, enrolling about 150,000. In addition to 10s. universities, the Brazilian higher /oo education system includes multiple-faculty institutions that offer program in more than one knowledge area (such as social 400 --- -- 7 --------------- ___ sciences and technology) and 200 institudonsthat offer single-faculty programs in only one knowledge o . . . . . . . . . . . . . . . . . . . . . . . . .. . area (such as social sciences). SO A70 11-4 o Over 80% of the higher educadon ________________________________ rinstitutions' are of the single- Chart 1 faculty type. Within these two egories, most of the single- and nearlyall of facultyinstitutions institutionsare private. Chart 2 summarizesthe current enrollmentbreakdownby the multiple-faculty governingauthorityand institutional type. 'TheMinistryof Education(MEC) atisticscountseach isolated-facultyas an institution, although,in the private sector,sevenl *isolated'faculties maybe under the controlof a singlegovening boardand, thus, actly institutionswould constitutea single institution. It is esfimatedthat the total numberof private non-university decreaseby as muchas 50% if the governingboardwere the inclusioncriterion. m-2 Programs of study range from Total Higher Education Enrollrnent by GovernmentAuthority 4 semesters, such as in pedagogical and Institutional Type, 1989 programs that prepare elementary T1,0uandM schoolteachers,to 12 semesters,such as in professional programs that ao. preparemedicalphysicians. Degrees ____00__________ are offered in teaching, vocational areas, by discipline, by profession, _o- for example), and for military or religiouscareers. 100< - . Almost two thirds of students Fdisrl |tain Munlidpl P to are emolled in the broad field of arts, Nen-Unhnraltlz IF5"I t.Is- 53M1 5 79 social sciences and humanities. Exact Univoaltl_ 323.76 u4me 3SD 36l5 sciences (11%), engineering and technology (10%), and health -u~uZ Unlwrnlilm msciences (12%) enroll about one third, u-a T.bb Urn with agriculture holding only 2-3% of the enrollment. In 1989 the system Chart 2 produced over 220,000 graduates, about half from universities. Assuminga five-yearaveragefor completion of a programby students,and giventhat the 1985freshman enrollmentwas around 400,000, 50% is a rough estimateof the graduationefficiencyrate in BrazDlian higher education. This rate is relatively low consideringthat Brazil requires competitiveentrance examinations(vestibular) for all highereducationinstitutions. In the mid-1960sthe BrazilianGovernment decidedto meet the socialdemandfor highereducation by encouragingthe establishmentof more private institutions,mostly of the non-university type, and openingnew federal universities,mostlyin the capitalsof the poorer states, where none existedbefore. The implicit policy goal was to consolidatea relatively elite public university system, offering undergraduateand graduate instructionfree-of-charge to the country's best students and emphaszing research, whilecreatinga private sectorprovidingmass education. However, whilefederal 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 Americancountries, such as Argentinaand Mexico,were openingthe doorsof publichighereducationto nearly all secondaryschool graduatesand creatingmass public educationsystemspaired with small private systems. Betweenthe mid-1960sandmid-1970s,Brazilianhighereducationexhibited continuedfast growth, averagingaround20% per year. Althoughall sub-systems grew in studentenrollment,privateenrollment increasedmost rapidly. In 1965, 38% of total enrollmentwas in federal institutions,17% in state, 1% in municipal,and 44% in private; by 1983, theseproportionshad changedto, respectively,24%, 10%, 6%, and 60%. By 1989,federal and municipalenrollmentshareshad decreasedto 21% and 5%, while the state and private shares had slightlyincreasedto 13% and 62%. Since 1983enrollmentin federal institutionshas declinedslightly, whilestate and private institutionshavegrown steadily. Brazil's educationsystem is highlydifferentiatedby region. In the poorest region, the Northeast, roughly 4% of the age cohortattend highereducation,comparedwith 10% do in the richest region, the Southeast. The public sectorpredominatesin the Northand Northeast,whilethe private sectorholdsthe mn-3 majorityof the enrollmentin the other regions,especiallyin the Southeast,where it has more than two thirds of thestudents. There is littlefederalandvery strongprivate(90% of enrollment)presencein Sao Paulo, where nearly all of publichighereducationenrollment is in state schools. Althoughfederal and state institutionscharge small fees for studentservices,and also contract specializedservices (research,training, technical assistance,and health care) to public and private agencies,the constitution prohibitsthe chargingof tuition. Thereforepublicuniversitiesare specifically mainly financeddirectly with governmentfunds. In contrast,private institutionsare financed mainly through tuition and fees, although some private universitiesreceive government funds (mostly for graduateprogramsand research)and contractservicesto publicand privateagencies. Studentsattending private institutions can also benefit from public funds by applying for educationalloans and/or scholarships. Federal expenditure 2 in 1988 averagedUS$7,930per student, about fifty on federal universities timesmore than expenditureat the primaryand secondarylevels (roughlyUS$150per student). This level of spendingis higherthan that of otherLatin Americancountriesand it is similar to that of many developedcountries. The average for Chilean public instititionsis US$1,700 and for Venezuela is US$1,625. Unit expendituresin private institutionsrange from about US$4,500 in the best Catholic universitiesto a low of US$500in some non-university private institutions.Table 1, whichsu enrollmentsand estimatescosts by sector, shows that total public spending in higher educationis equivalentto 1.3% of GDP and accountsfor approximately 80% of total spendingin higher education in Brazil. These figures do not includespendingon research in universitiesby agenciesoutsidethe Ministryof Education,which is equivalent to another10-15%. In Brazil a large share of total public spendingon educationgoes to highereducation. Only2% of total publiceducation enrollmentis in highereducation,but an estimated23% of totalpublic (federal, state and municipal) educationspendingis directedto this level. The Ministryof Educationalone spends about60% of its budget in its federalinstitutions,which serve only 22% of higher education students. Assumingthat enrollmentin graduate educationas well as percentageof faculty with advanced degrees are proxies for output quality, it is possibleto differentiateseveral tiers of higher education institutionsin Brazil. On this basis, as wellas on the basisof informed judgmentsby relevantobservers, the two state universities of Sao Paulo (USP) and Campinas (UNICAMP)are in a category by themselves. These are by far the highestquality institutionsin Brazil and boast significantgraduate programs (26% of the enrollmentin graduateeducationin Brazil is in these two institutions). Under- graduateeducationis also considered by far thebest in Brazil. Scienceand engineering major constitute portions of enrollment. Total enrollmentin thesetwo institutionsis around50,000. 2 Figuresare basodon officialexchangerates. Excludinguniversity hospitaland pensioncosts,the unit costs would be about US$6,000. Seepage M-11 for a moredetaileddiscussionof unit costs. m-4 Enrolient and Sector.1989 EstimatedCostb Hrither Education Ated : JE.-:-ed :::0-UntCostan: Total Cost in Toual Cou ..- i fSecor :;E;; :Enrollmemnt. - USS USSbionian E %ofGDP;; s . f;St- i- i,- (13 f0 4,0 ,ii 3AT- 03- t. U...a, S6 (,) . .. K. 1 ....... ('' .X '"".....: ... . .. ...- . .- . . .. ;..E .. .. . ..-. ::-~~~~~~~~. ... ...- ..E. E. E-. ..-.- f - -- .-. ....E. D!i- i.E Total ~~~~~~~.i: A;; i.. C . 1,570,.E. ;.i. .-- . E 3 2. --.:- E......- 4.9 1. :::- -:-: : i: :.E f . -. -. ... . . . .. ::-:. .- :-:::-..-.......-:.:.....::: :-:--::: f 7E::.:--:- !A. -- :Sa :i:gt::-.:Paul: and:Wolynsoado:cs ;o Roge Vai.IJniip tdoniesu idabke by icostae estimate :~ ~~b : -: ::::gEdon Nim e. ca.Euse of voaie bosi; zes. asuSdl:ipie~ ela th ** sh ; 0ould beconiderd indicatv onl.Th o fincud exedlwe o stwkrsip mearcb i:- :-:.---:n:: :- :a:: f: t:s-. fogM-e::.hers, .fli of wicl s ue:tmtd a USSSSlmi:mllon,: an -hh go: i!~t invusi | , , X , , ,, in ei- *E-iba, Table 1 Aother tier is formedby a rdeltvey large numberof public and private universitiesthat provide :undergraduate educationof varying quality. Only a few of these institutionshave significantgraduate programs: the federal universitiesof Rio, MiDas Gerais, SantaCatarina,Sao Carlos,Rio Grandedo Sul, and Pernambuco, as well as the Catholic Universitiesof Rio and Sao Paulo (PUC-Rioand PUC-SP) enroll 7% or more of their studentsmn graduateprograms. Total enrollmentin these somewhatresearch oriented institutions is around 92,000 in the federal univerities and around 45,000 in the private institutions. Enrollmentin the remainingfederal univerities, several of which are reported to be of reasonablequality (that is, those withoutsignificantgraduateeducation)is around215,000. Other state institutionsof similar qualityenroll 140,000. The remainingprivateinstitutionscanbe dividedbetween universities(enrollment310,000),which generallyoffer undergraduateeducationof reasonablequality, and facultieswhich offer low quality,mostlyevening,coursesin the socialsciencesand law, mainlyfor a lowermiddle class clientele(enrollment 580,000). Table2 is a breakzdown of enrollmentsby thesetiers of higher educationin Brazil. m-5 Equity and External Efficiency by Tvpe of Istitution romUment 1989i' Several studies have documentedthat .... the vast majority of students in all higher Two High QuitY st univerhie -50,000 education both publicandprivate, institutions, - - and UNICAMP) - . (USP . . -- are from the middleor upper classes. Public SaxPedael Univ;es 9wilb }te#uch - - -- - institutions have students of the highest -- d Gdsuaozdmtatios -- :-socioeconomic A- status. A recent survey of income of parents of federal university T.o High Quai Calho Us_-ver:ties 45,0w0 students shows the following breakdowns (1989) in terms of minimum salaries: 0-3 minimum salaries, 18% of students; 4-10, - Other Swecand MwdCIDh Z18,000 37%; and 11 and above44%. This compares with an estimated 10% of the general OdherPrivate Univerities- populationearning more than ten minimum Conf,ssiorul and Communiy 191,000- salaries.' Socio-economic class within higher . . - 155,000- -:-Cunl education varies by subject studied. A study Private-Non-Umveauy52a±matzoea(Paul, 1991) shows that over 30% of enrollees in day engineering, biomedical Total: -- - , studies,and law in a private institutionin Sao -___-_- Paulo had fathers with higher education, compared to 13%in teachertrainingand 16% !t -Dd53O5flUflfl5 OfDS?Of Perf',,~~ in night courses in law and social sciences. j''.OII~ . __' Fully 61% of students in the federal univerity in Cearastudyingcivil engineering had fathers with at least an undergraduate Table 2 degree. This compareswith 20% or less for areas such as literature, nursing,geography, and pedagogy. For Brazil as a whole about8% of the adult populationhas attendedhigher education institutions. The studentprofileof the moreprestigiousprivateinstitutions(someCatholicuniversities) is similar to that of the public institutions. Table 3 providesa summaryof recent studieson the socio- economic status of highereducationstudents. As can be expectedthe earnings of those withpost secondaryeducationof any sort are high. Data from the 1980 census shows that the incomeof those with 13 or more years of schools is more than fifteen times those with five years of schoolingand more than 180 times the incomeof those with less than three years of schooling. Meyer(1982),Farret (1985),Salih (1990),Gomes (1990),andPaul (1991)havedocumented these income inequalities. Estimates of income distribution in the generl population are from Camargo, et al. (1990). he lop docileof the labor force earned an average of 17.1 mininmu wages in 1989. The second docile earned an averge of 5.6 minimum wages. m-6 Backgroundof Entcrds~gSucZ -forSeven]Bnzian Institutions : :: : - : :: :;i-E 7 - : . i E ......... 70 i!~~~~~~~~~~~~~~. :. . .... . . E-,#-g Attemded Father wit -Pit..lic. ...... -~~~~~~~HghrSecondary ... .. ' " ..'-.- .... i,.-.,..,..,.',-',..,,, .. . i;0: . ... .. . .; -.... .-. . . . Ei.:i:-.:. :; :::;: . -! :E: -:: -:t. . :: - ;. i-EA- .: . :.; . .: ::: ::: : : : :-:E ::: .... -f :.....:. : ............................ ...... -,. . ~ ~~~~ ..... .. ~ ..~ ~ ~~~ .. .:.:.:.. ... .- t:: 2:: . -:: :-: .i 7ERE! - :: - ::W!------!----: - :{- . - - ! : :: X :~~~~~~~~~~~~~~~~~~~~~~~~~:-:- ............ :Day- yc lai in Teaher:raining. . 13% : 62%: :: i-- AU nightdaijb ses -. ;~~~~~~~~~~~~~~~~~~~~~~~-:-:-: ":% : ': -':F -' -:0 : ::...... : -:-:::- ""- ':.' .. .. . .. 17% .... . :62% ::- .. :0 .. .. ..: ..: :. .: ... .. .. . ... .. .. .. . . .. ..... ... .... .. ... .. . .. ... .. .... . .. . . . S.t : S :''., ','''' '0. '!' -'- . ., ' : :- S'-.-: S ''iE,' i , i- .: . .. . ..... ..... ... .. . .. -:toiUnieriy Feea of Ce-r n E--f - 331%: 20% - i showdiffernes Stuginesn ind DaeaPrning betee grdae1rmlw n ibpetgecusso%td onteeaionishaips, betweeny therlibg, man5tractu and dctonotu.Oerl hr 31e s ninfcn Staemplnimenst ao eng al ihreuato rdae. oee,ol netidt n-fut fbge edcatioUniveradutesoCareyuhn h aenvrwrcd hs enrlyatn h etisiuin n thihowsdifferenession. Studienes earing boetweena graduates fromaly n h coursesty srvtud lowadhgrsier onsthetetionsh ipes betwee Sothealaborsmarket and higherelducatione foutut athere isauainsgnifican Oeverallyar secondaryschool, and then decidedto go to school in thieevenings. Most of these graduatescontinue 4 Reporte in Claudia de MournCato (1990). m-7 in the samejob theyhad whilestudying,possiblywith some salaryincrementor some modestincreased aspirationsbecauseof their degree. Thesepeople are payingfor low-costcourses and getting modest perceivedbenefitsfrom them. There is little informationon privateand socialrates of returnto educationin Brazil. One study (Psacharopoulos,1985)reports private rates of return of 25% for secondaryeducationand 14% for highereducationand socialrates 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 secondaryeducation,12% for full primaryeducation,and 16%for four years of primaryeducationduringthe periodof 1982-86. GraduateEducationand Research Graduate educationis a small but importantpart of the higher educationsystem. Current enrollmentis estimatedatkbout 50,000, the bulk of whichis in master's programs. Roughlyone third of the graduatestudentsare pursuinga degreein the 'hard' sciences(includingtechnology). Abouthalf of the graduateenrollment and one third is in state institutions,nearlyall (97%) is in federalinstitutions in the state of Sao Paulo. There are more than 1,300 master's and doctoralcourses, or an averageof 40 studentsin eachdegreeprogram. Graduateeducation has recentlygrownrapidly, and is of relatively high quality comparedto undergraduate education,to a great extent becauseof a relativelyeffective competitive fundingsystemoperatingoutsideMEC's regularfundingof higher educationinstitutions. The fundingmechanism for graduateeducationand researchin boththe publicand private sectors is basedto a great extenton open competition and peerreview. Publicand privateuniversitiesand non- profit researchinstitutesget funds for graduateeducationand research. Thesefundsmay not appearin their budget if they are received through foundations'. The most importantGovernmentagency financing(and undertaking) scienceresearchis the Councilfor ScientificandTechnological Development (CNPq). The CNPq providesfundsfor researchersin the form of salaryincrementsand also provides supportin the form of researchgrants,usuallyto individual researchersand for relativelysmallamounts. Along with its grant program, CNPq administerseleven separate research institutes and a large scholarshipprogram(at present,there are 30,000scholarships insideBrazil and 3,500 overseas). The Agency for FinancingStudiesand Projects(FINEP) is a public corporationwhich acts both as a conventionalresearch foundationand as a developmentbank lending funds for technological innovationin Brazilianindustry. FINEP was set up in 1967to developBrazil's engineeringindustry. FINEP has 700 staff, more than 100 with doctoralor masters' degrees, and in 1988supported 1,840 development. FINEP also projects split almost equallybetweenscientificresearch and technological functionsas the ExecutiveSecretariatfor the NationalScientificand Technological Development Fund (FNDCT), a major governmentsource of grant funding for infrastructure,new buildings and new equipment in universitybasedscienceand technology research. Fundingfor FNDCT in 1992as well as for CNPq was severelyreducedbecauseof Brazil's ongoingeconomicdifficulties. The Agency for Training of High Level Personnel (CAPES)of the Ministryof Educationand Culture (MEC)coordinatesgraduateeducation in Brazil. CAPES'sprincipalinstruments of supporthave been grants and fellowshipsfor study in Brazil and abroad for individualspreparing for careers in university teaching and research (in 1989, CAPES supported around 14,000 students at Brazilian universitiesand 2,000 studentsin NorthAmericaand Europe). CAPEScertifiesand rates all graduate programs in terms of overall qualityand providessupportto 700 graduatecourses(two-thirdsof all the m-8 courses in the country). It also runs a programto help build up institutionsso that they can offer new courses. In 1989CAPES' total budgetwas equivalentto US$200million. Private Higher Education Private higher educationhas existed for over a hundredyears in Brazil. Its participationin the overall system increasedduring the late 1960's and 1970's, when the militaryregime, confrontedwith strong pressuresfor increasedenrollmentin public institutions,as well as political oppositioncentered in the public insttutions, resolvedto permit the openingof large numbers of new private institutions. Duringthat time higher education,both public and private, increasedrapidly, but the greatest increases were in private non-university institutions. Since the 1970's, a majorityof studentshave enrolledin private institutions. In the 1960's and 1970'sprivate higher education grew at 15%per year, muchfasterthan the secondarysystem, with the resultthat a more heterogenousclientele,includingstudentswith deficientpreparation,nowattendprivate institutions. Growthover the past six years has only been 2% per annumas a result of the economic downturnas well as continuingrestrictionsby CFE on openingof new programs. The private sector in 1987 comprised72% 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 sectoremphasizesundergraduate educationand humanities,social services,and teaching. It has 75% of the studentenrollmentin humanities,66% of languages,57% of exact sciences and technology, 14% of agriculturalsciences,43% of health sciences, and 51% of arts. For teacher training the private sector has 67% of teachingprograms, with 69% of studentsand 75% of graduates. Althoughthe privatesectorconcenrateson undergraduate education,it is also an importantpartner in graduateeducation: 11% of programs, and almost20% of students. However60% of private sector graduate enrollmentis concentratedin two universities,the CatholicUniversitiesof Rio and Sao Paulo (PUC-RJ and PUC-SP). In comparison,about one third of the federal universitieshave no graduate education,and 50% of graduate enrollmentin federal institutionsis concentratedin three universities, in Rio, Minas Gerais, and Rio Grande do Sul. CAPES' program of peer evaluation and rating of graduate programs shows that graduate educationin the private institutionsis of quality comparableto that of many federal universities. Private graduate educationreceives more than 10% of the funds allocated by the government agencies CAPES, FINEP, and CNPq and supplies 15% of theses and dissertations. Private Univenties. PUC-Rioand PUC-Sao Paulo enroll 25,000 students, offer high quality undergraduateand graduate education,and produceresearch. PUC-Rio has a studentteacher ratio of 10:1, which is similar to that of federal universities;17% of its enrollmentis in graduate education. PUC-Sao Paulo has 15% in graduateeducationand a 16:1 ratio. Thesetwo institutionswere founded many decadesago, have large numbersof graduatestudents, student-teacher ratios lower than in many other private but higher than in public, some full-timefaculty, and costs higherthan manyother private but lower than public ones. ' Mhissectiondraws (thoughnotexclusively)on comprtive casestudies,undertkn by VabI(1990), of four universities,including2 Catholicuniversities(PUCs). commumity m-9 In privateuniversitiesotherthan PUC-Rioand PUC-SP,80-90%of teachersdo not havegraduate degrees;only about 15-20%are fulltime;63% are hired on hourlybases for teachingonly (no research expectation) and receivesalariesof aboutUSS200 per month. In contrast,in public universities55% of teachers did not have a graduatedegree in 1988; and 77% were full time. With regard to student flow, the ratioof 1985first yearopeningsover 1989graduates was 789%, compared to a publicuniversity ratio of 63%. Privateuniversities and use themmorethan twice use facilities,libraries,and laboratories, as intensivelyas public institutions. FiE student/teacher ratios are 24:1 and unit costs average about US$1,000comparedto 8.8:1 and aroundUS$8000,respectively,in federalinstitutions. PrivateNon-University Instiutions.' A large numberof the non-university institutionsoffer all frompre-school levelsof schooling, thusmaximizing to tertiaryeducation, the utilizationof their facilities and direct administrationsystems. They usually do not have campuses,simply occupy buildings, sometimes in more than one location,often offering night courses only in buildings operating as secondaryschoolsduring the day. Somehave good facilitiesbut the majoritydo not. Many single- faculty and multiple-facultyinstitutionsaspire to university status, because of higher prestige and relativelymore autonomy. The vast majorityof these institutions are locatedin Sao Paulo and Rio. Incomefrom tuition accountsfor 50% to 96% of an institution'sbudget, with an average of approximately 80%. Thefederally-funded studentloan-program brings in 5-30%of the budget,with an averageof about 13%. Other sources of incomeare provisionof services,such as health servicesin hospital-schools and at-largeeducationprogramscontrcted by local corporations for thetrainingof their employees,as well as the investment of tuitionmoneyin the capitalmarket. Sincethese institutionsdo not keepdetailedaccountsit is difficultto reconcileincomeand expenditure. Many institutions award partialor full tuition waiversto studentsin financialdifficulty. Waiver levelsvary from 5% to 45%, averaging10%. Abouta 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. One groupof privateinstitutions is that of wcommunity'institutions,whichsee themselvesas non- govermmental public entitiesprovidingsocial services. This group actuallyconsistsof two types of institutions,both sharingthe samemission,but with distinctaffiliations.The most traditionalsub-group is that of the confessionalinstitutions,founded and administeredby consortia of religious orders, generallyin existencemanyyears prior to the 1968reforms and consequentexpansion;manyof them are university,but not all. A smaller and newer sub-group of community institutionsis that of secular institutions,all universitiesfounded and administeredby community-based groups with no religious affiliation. Total enrollmentin confessional wasaround225,000in 1989(25% of totalprivate institutions enrollment). A second group, the majority of private institutions, can be defined as 'entrepreneurial" institutions,which considerthemselvesas educationalenterprisesseeking a profit to all intents and purposes. Manygeneraldirectorsof theseschoolsare their ownfounders. Theirbackgroundis diverse, from teachers to retired army generalsto lawyers. Some were establishedby elite families(such as UNIFORin Ceara, Mackenzie in Sao Paulo, and SBIin Rio de Janeiro),others are ownedby business partnershipscapableof cost-effectively deliveringlow-costprograms (such as UPIS in Brasilia and Objetivoin Sao Paulo). Total enrollmentin 1989in entrepreneurial institutionswas around 700,000 ' This section draws on case studies of 16 higher educationinstitutionsof the state of Rio de Janeiro undertakenby a team of researchersled by Laura Dantasof SBI. m-10 (about75% of total private enrollment). However, unlike in the primary and secondary education sectors, the law governingprivate higher educationdoes not allow institutionsto be profit-making. Nevertheless,ownersfigure out creativewaysto captureprofit. Demandfor PrivateHigher Education. Considering that publicinstitutionshave higherprestige and charge no tuition, they are almostinvariablythe first choiceof students. Private highereducation becomesthe choice when: (i) the student is unable to gain admissionto a public institution;(ii) the studentis a full participantin the labor force and, thus, cannotafford to attend daytimeclasses;(iii) the local publicinstitutiondoes not offer the courseof study desiredby the student;and, in some cases, (iv) the studentsees the privateinstitutionas better qualityin selectedcoursesof study. Once the first-order decision is made to seek admissionin the private sector, a second-orderdecision is often required, becausemostlocationshave morethanone privateinstitution.Theprofileof studentsin thebest Catholic universitiesis similar to that of public institutions. However, most of the low quality, non-university institutionsare attendedby older, less preparedstudents,most of them workingfull time, and staffedby less prepared teachers, sharing low intellectualdemandson each other the ("pacto currupto'). The widespreadperception(largely unmeasured)is that the majorityof these schools are 'diploma mlsW offeringformal credentialsbut very little content. Hard data on this overallimpression,throughstudies of learningor income after attendingthese institutions,is not available. Severa factors account for the phenomenonof people paying for low-qualityprivate higher education. First, there is demandoverflow;demandfor higher educationoutweighssupply by at least 2:1, but the ratio of applicantsby 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 educationin genera) has a high scarcityvalue, acquiringevenlow-qualityhighereducationis betterthan acquiringno highereducation,especiallyin a labor marketwhere the majorityhave a less than primary level educadonalattainment. Third, the publicsector of the labor marketis dominatedby crentiaism, in which career ladders for civil servantsare definedby credentialsand seniority(as well as patronage). Hence,the Governmentactuallyencourages low qualityprivatehighereducation by assigningan intrinsic value to the diploma. Althoughthe private sector of the labor market is generallymore interestedin skills than in credentials,it still has to complywith regulationswhich are createdby trade associations and mandatedby the publicsector. For example:the law regulatingthe professionof journalism states that corporationsproducingany form of publication(such as client newsletters)must employcertified Brazilianjournalists, whetherthey needthem or not, an artificeof thejourmalists union to guaranteejobs for its members; companiesoften work around the legislationby hiring a card-caringjournalist who receivestokencompensation for agreeingto havehis/her nameon the company'smasthead,but who does not do the work. Similar legislationexists for manyother professions. Such labor market regulation guaranteesthat investmentin any privatehigher education pays off, giventhat studentsattendingprivate institutionsreceivediplomaswith the samenominalcertifying value as one obtainedfrom a high quality (publicor private) institution. ISSUESIN HGHER EDUCATION This sectionseeksto document themainissuesandproblemsof the currenthighereducation system in Brazil. It coversthe key areasof the costs,governance,and management of federalhighereducation, the relationshipsbetweenthe federalgovermment and privatehighereducation,and overallfinancingof the higher education system. Costsof FederallyFunded HigherEducation There is much controversywithin Brazil on whether or not the costs of higher educationare excessive. Based on comparisons with state and private institutionsin Brazil as well as through some internationalcomparisons,it can be concludedthat, while federal institutionsemphasizehigher cost programs in science and technology,and some of these institutionsprovide significantprograms of graduate education,overall the unit costs of these institutionsare significantlyhigher than private institutionsin Brazil providingsimilarprograms. The main factorsaccounting for such high unit costs are the low student/teacher and student/non-teaching staff ratios. There are only minor differencesin expendituresbetween research and graduate oriented federal universitiesand purely undergraduate institutions;nor arethere any economies of scaleamongfederalinstitutions.In additionphysicalfacilities in the federal institutionssurveyed are grossly under-utilized. International Comparisons. Becauseof Brazil's rapid inflationand extremely variableexchange rates, it is difficultto arrive at reliable dollar estimatesof unit costs. In the years 1988and 1989, in particular, the differencebetweenthe officialexchangerate and the parallelexchangerate (morethan 40% in 1988)makesreliableinternational comparisons difficult. Estimatedaverageunitcosts in federal institutionsin 1988are US$7,930per student,basedon workundertakenby Jean JacquesPaul and Elisa Wolyneck (1990), which simplydivides MEC's budgetper institutionby the number of students and which uses official exchangerates prorated on a monthlybasis during 1988.7 FranciscoGaetaniand JacquesSchwartman (1991)arguethat theseunit costs wouldbe about40% lower if the parallelrather than the officialexchangerate were used, which wouldput the federalinstitutionsin a more favorable light in international comparisons.Gaetaniand Schwartman also arguethatunit costswouldbe reduced by an additional25%, to about US$6,000 if the costs of universityhospitals (about 10% in many institutions)as well as the costs of payments to retired teachers (15% for some institutions)were discountedfrom the overall budget. However, retirement costs may be considereda part of wages. However, even makingthe exchangerate adjustment suggestedby Gaetaniand Schwartzman,Brazil's unit costs are equal to abouthalf of those in the USA, Japan, and Australia,and are significantly higher than those of its Latin Americanneighbors. Becausethe main determinantof costsin highereducation world-wideis teaching,whichnormally accountsfor 70% of higher educationcosts, the use of student-teacherratios, which are not subjectto problemsof exchangerates correctionsand are relativelystableover time, provideanotherway of doing cross-country comparisons. Full-time equivalent (FIE) student-teacherratios in Brazil's federal 7 otherwise Unless noted,thisreportrelieson theofficiaexchn ate. M-12 institutionsaverage8.8:1, comparedwith 12.1 studentsper teacherin US private institutions,and 17.4 studentsper teacher in US public 4 year institutions'. Another element of importance is the ratio of students to non-teachingpersonnel. While no internationalcomparisonsare available,the current studentsto non-teaching staff ratio in Brazilianfederal institutionsis 3.9:1 (it was 5.1:1 in 1980). In short, despitethe usual caution attachedto international comparisons, Brazil's federalsystemis undoubtedly one of thehigh cost systemsin the world, onceone takes into consideration Brazil's salary levels. The main determinantof these high costs is the large numbersof teachingand non-teaching staff comparedto students. Compaisons within Broi. The overall FTE studentteacher ratio in federal institutionsis 8.8:1, comparedto 10:1 in state universitiesand 29:1 in private universities. A review of the data showsthe following: * vary greatly in studentteacherratios, from about4:1 to about 12:1. A few Federal institutions small specializedinstitutionsin agricultureand medicinehave around 4:1 ratios. * In terms of student-teacherratio, there are no economiesof scale amongthe 23 universities. In fact several of the largest institutions,such as the Federal Universitiesof Pernambucoand Paraiba, have studentteacher ratios ef less than 7:1, and the largest institution,the Federal Universityof Rio, has a studentteacherratio of 8:1. * the higheststudentteacherratiosin federalinstitutionsappearamongthe youngestinstitutions. While this may in part be a resultof a lack of graduate education,it may also be a result of an inaequate abilityto lobby MEC to authorizenew positions. * The state universityof Sao Paulo (USP),the largest and best institutionof higher educationin Brazil, has a student teacher ratio of 10:1. UNICAMP, with the highest percentage of enrollmentat graduate level (31%) has a studentteacher ratio of 7:1, and UNESP also 7:1. Most of the other state institutionshave studentteacherratios of 10:1 or above. - Graduateeducation,which requireslower studentteacherratios than undergraduateeducation, affectsthe studentteacherratio in UNICAMP,with 31% of enrollment at the graduatelevel, and USP, with 259% of enrollment at the graduatelevel. Amongfederal institutionsonly the federal universitiesof Rio and Minashavemorethan 9% of their enrollmentin graduateeducation (18% and 9%). * The highestunit costs are in the FederalUniversity of Vivosa,whichis a specializedagricultural institution. The three universitiesin Rio also have amongthe highestunit costs. The lowest unit costs are in institutionsin the North and Northeastwith little graduate education. Paul and Wolyneck(1990)estimatedthat unit costsin federalinstitutionsare 60% higherthan the unit costs of the Universityof Sao Paulo (USP), in spiteof the fact that 25% of USP's enrollmentis in graduate ' Te argument made by Francio Gaetai and Jacques Schwartzmantha comparsons with Americn insttutions are inapproprite, becu much teacig in the US is done by teaching assstants, results in only marginaladjustmentswhen student-teache rtios are baed an FTE ither ta headoounL M-13 education, and USP is considered thehighestqualityinstitution in Brazil. But Gaetaniand Schwartzman have arguedthat more than 9,000 studentsin USP reportedas full time graduatestudentsare in fact not engaged in fulltime studyor research. After eliminating thesestudentsand takingout funds for retirees, he argues that the differenceis only about 13%. However,in a separatepaper, Paul (1989)has notedthat there are also significantnumbersof *ghost"studentsin federal institutions. The conclusion,even after accepting Gaetaniand Schwartznan's revisions,is that the highestqualitypublic institutionin Brazil, USP, does not appearto be particularlycost efficientbut nevertheless has lowerunit coststhan the majorfederalinstitutions with which it is compared. Institutionalcasestudies undertaken by RogerioVahlbreak down expenditures in 1989by budgeting categories,by faculties,and by departments for two federaluniversities,Minas Gerais (UFMG)and Santa Catarina(UFMG). The averageunit cost of the two federaluniversitiesstudiedis US$10,000(1989data). The nominalstudentteacherratio is 6:1. Full-timeteachersare requiredto teach 8-14 hours (comparedto 24 hours in the privateinstitutions)per week and are expectedto prepare classes and undertakeresearch during the rest of the time. The top salaryfor a fullprofessoris aroundUS$2,300per monthin the federal institutions(comparedto US$1,600per month in the private universitiessurveyed). The numberof non- teachingstaff in the two federal institutions is equivalentto about25% of studentenrollment(comparedto 5% of enrollmentin the private institutions). The average salary of the non-teachingstaff is roughly US$1,000per monthin the two federalinstitutions(compared to S300in the private institutions). Vahl sought to break down costs by program of study. Sinceuniversitiesdo not keep systematic recordsof this sort, only rough estimates,coveringteachingcostsonly (but not maintenanceand materials), are possible. lhe followingare the average cost breakdownsin the two federal insdtutions:science and technology, $9,447; biomedicalsciences, USS11,985(e.g., medicineUS$11,844, dentistry US$13,059, nursing US$11,755);and humanitiesand social sciences, US$6,854 (e.g., law US$6,681, mathematics US$7,290,pedagogyUS$9,863). The costdifferentialbetweenthe hard sciencesand the socialsciencesand law, is about 1.5:1, comparedwith a USA differentialrangingfrom 2 to as much as 5:1. It suggeststhat the incentivesystem is such that there are no rewardsfor makingsavingsin potentiallylow cost programs. of very large lecturesaccompanied In fact federaluniversitiesrarely use the combination by smallerclasses commonto the USA. Anotherelementof interestis that of spaceutilization. The two federaluniversitiessurveyedby Vahl provide28 squaremetersper studenh comparedto 6 squaremeters in the four private universitiessurveyed. Moreimportantly,the formerhas an averageof 1.6 seats for everystudentenrolledin the systemcompared to 0.8 in the private institutions. Assumingthat classes can take place eight hours in a day (a reasonable expectation,with classesrunning8-12a.m. and 2-6 p.m.), and studentsattendaboutfour classesof one hour per day, on averageseats in the two federalinstitutionsare utilizedabout30% of the time, comparedto 85% utilizationin the comparableprivate institutions.Based on this analysis,the physicalfacilitiesin the two federaluniversitiessurveyedare adequate to serve abouttwicethe numberof studentscurrentlybeingserved withoutinitiatingany night classes. The argumenthas been madethat the higherunit costsof the federal institutionsare a resultof higher quality staff, manyof whom undertakeresearch, of programsin science and engineering,and of graduate surveyeddo undertakeresearchand extension programs. It is certainlytrue that the two federal institutions and do havemanyprogramsin highercost areassuchas medicineand engineering.On theotherhand, PUC- Rio and PUC-SPwhich are the two private insdtutionswhich are acknowledged to be amongthe highest quality institutionsin Brazil, with high percentages of graduatestudents,and with input and output quality M-14 measuressimilarto the two federal institutionssurveyedhave muchlower unit costs. Availabledata (from Tramontinand Braga, Vahland Paul) are shown in Table 49: .f: ..... . f-: C . ;:-::: ... ;.-E . . . ;:W ... .. f ,: . 0 2 D t., i 7:7 T: ft t:'.;f;Sf:F-;i .. .:ogts md O i Two Fdedi and Two Pa jml6v; Univen :-: - UEMQ :UFSC PU-ioB PUC-E Perc'i Graduate Studet 11 10 717 1 S Peret Staff wifi Doctorate- 20 i 18 .. 30 S: .1.5....... . .. . ... FM Se Ratos:: -Tecer 1 -: ::-10:1 .... .:-6 . . . .:1 .. S- ace-teachingSi-:-ai 3:1-: - - 11:1-.-;:g: 14..:-1 .ntCasin -S-- P 5 . .53....... -! .'L. ' M. ,,,,,,, E,, , S. A1 4 ,,. . .. .. ....... .o.rce: ,V.a. i, !al T , (1990)- '... -' a Table 4 Unit costsin the two fedeal universitiesare twiceas high as PUC-Rioand aboutfour times as high as PUC-SP,which haveequaly high qualitystaff and majorgraduateprograms. It shouldbe notedthat PUC-SP focusseson the social sciences and has very little enrollmentin medical or hard sciences. Thereforeits unit costsshouldbe comparedwith estimatedunitcostsof ova US$6000for the two federal universitiesin humanitiesand the socialsciences(Vahl). PUC-Rio's enrollment profUeis similarto the two federal instiutions surveyed, since it has a very strong science and engineeringprogram (but no medicalschool), muchof it supportedby grants from FINEP. In short the two best private institutions in Brazil offer programs similarto those of the two federal institutionsat less than half their costs. It shouldbe notedthat the two federalinstitutionssurveyedare amongthe oldestand best federal institutions. In contrast,the FederalUniversitiesof OuroPreto, Mato Grosso,Juiz de Fora and Espirito Santo have negligiblegraduate enrollmentand are reportedto undertakelittle or no research but have FrE studentteacher ratios of 9:1 or less and unit costs rangingfrom US$6500to US$9000. and Management Governance Inslttow of Federni The system of governanceand management of federalhigher educationencourageshigh costs and discouragesefficiency,quality, and diversity. Under the current system federal institutionshave little autonomyof decisionmaking and the federal government has inadequateknowledgeof the performance of these institutions. 9 For purposes of com4ability, data from Paul for 1988 on student teacber rasios and unit cost is used for the four insttutions. Data On the other item ar, for the fedal institions, from Vahl, and for the privae institutions, from Tramontin and Baga M-15 Personnd PoUicy.The federaluniversitiesthemselves have no policy with regard to personnel. All permanentstaff are contractedby the universitiesbut paid by the federal governmentthrough the Civil ServiceSystem accordingto civil service rules. Howeveruniversitiesare permittedto contract personnelon a temporarynon-civilservantbasis. No universityis permittedto establishits own plan of positionsand salaries, nor to establishits own criteria for promotion,nor to establishsalaries for administradve/technical personnelon the basis of marketfactorsor productivity. Withouta doubt this lack of flexibilityleadsto low motivationon the part of faculty. Universitiesare not permittedto hire staff, sincethe federalgovermment setsoverallhiringpolicy,and recentlyhas enforcedfreezeson hiring. Furthermorethe universitiescannotfire staff exceptunder the strict limitationsof federal law for civil servants. The 'Law of Isonomy of 1987regularizes employment of all teachingand non-teaching staff in federaluniversitieson a nationalbasis. The law establishes354 non-teaching (technical/administrative) categoriesas well as the regulationsfor teachingstaff advancement in the career ladder (that is, being promotedfrom auxiliary,assistant,andassociate to fullprofessor). Thelaw includessalaryrelationships for all positions. It requirespaymentof the samesalaryfor eachcategorythroughout Brazil. Underthe law promotionmay be gainedthroughgettinghigherdegrees(mastersor doctorates)or throughtime of service. A teachercan reachthe associaterank solelyon the basisof time of service. The law requires evaluationof professional outputonly for enteringthe professionand for becominga full professor. In no case are advanceddegreesrequired (advanceddegreeswere requiredbefore 1987). Amongthose currentlyat the associatelevel, 23% havedoctoraldegreesand 53% havemaster degrees. Withoutthe requirementfor advanceddegrees,these percentageswill fall and many full professorswill not have advanceddegrees.Legallya teachercannotbe fired becausehis academicwork is deemedmediocreor weak. Furthermore,if a teacher were fired, there wouldbe no assurancethat a replacementcouldbe hired, since such decisionsare subjectto the ongoingfederalhiring regulations. Acadekic Decision Making. While the law states that federal universitieshave *didactic, administrative, autonomy,'in practicethere aremanyrestrictions.The internal financialand disciplinary statutesof each universitymust be approvedby the FederalCouncilof Education(CFE). The area of greatest autonomy is in the area of teaching and research, where, since the end of the military dictatorship,the universitieshave been quite free. The CFE, besides accreditingall universitiesand faculties, establishesthe minimumcurriculumfor each course in all public and private institutionsin accordancewith nationalstandards. Onaveragethe minimumcurriculum covers 94% of the total of the creditsneededto obtain a diploma. WithinBrazil scoresof professionsrequire a valid highereducation diplomato operate eitherpubliclyor privately. Withthe exception of law, there are no objectivenational examinations for entranceto a profession,which is solelydetermined by the holdingof a validdipioma. Sao Paulo is currently experimenting with a professionalexaminationfor medicine. Universitiesare highly constrainedwhenmakinginnovationsin courses,sincethey must obey the federal guidelineson minimumcurriculum. Furthermore,while the universitiesare permittedto increase or decrease the number of students per course, they are presentlyprohibited from contractingnew professors and generallythey are prohibitedfrom firing any teachers,nearlyall of whomare civil servantswith tenure. Finally the Universitylevel Councilof Teaching,Researchand Extension,composedof representatives of all teachinglevels, must approvethe closingdownor extinctionof any course. This practicehelps explain the existenceof facultiesand/or departments with a large numberof teachers and a very small numberof studentsin areaswhere demandis down. Internal Management. The federal universitiesare formallymanagedby a university council, consistingof the rector and his staff, representatives of the main academicbodies, and representatives M-16 of studentsand society as a whole. A council of overseers includesrepresentatives of academia,the overall community(includingthe private productivesector), and MEC. Other councilsare relatedto teachingand research. There are also councilson teachingand for each departmentand faculty. The rector of the university is selectedby the President of the Republic from a list of three presentedby the universityand can be rector for only one term. Since the end of military rule, most universitieshave a popular eection in which all students, teachers, and administrative personnelvote. The candidatewith the highestvotetotal is sent up to the Presidentas the first on the list. Until recently, with few exceptions,the Presidentselectedthe first nameon the list. Studentsand facultyhave gone on strikewhen the second or third on the list has beenselected. The democraticelectionof rectorshas in some campusesresultedin the electionof highly qualifiedand dynamic rectors. However, there is a strong tendencyfor the process to becomehighly politicized,with resultingpolitical battles between various parties. In addition many candidates find it necessary to make commitmentsto their constituencieswhich at a later time make it very difficultto introducetough requirementsor measures to increase cost-effectiveness, such as requiring students to attend classes or eliminatingexcessive numbers of non-teaching staff'. The rector of the universityseects his technicaladministave staff from amongexistingfaculty and staff. These staff do not act as managersbut rather as interlocutorsbetween faculty, students, technical personnel, the rector, and the governient. There is no permanent planning staff in the university. The staff who work with the rector generallyhandle only short term budgetingproblems. There is no institutionalplanningmechanism.Thereforeinstitonal planningoccursby accretionbased on the interests and strengthsof the various departments and faculties. The office of the rector has no authority or capacityto make difficultdecisionsin resourceallocationsbetweenprograms. To a great extent this lack of centralplanningis a result of history. Universitieshave neverhad a strong centralist tradition and they have grown as accretionsof independentfaculties 11 . In additionthe proliferationof large numbersof committees with overlapping jurisdictionshas made it difficultfor universityleadership to take any controversialdecision. Effects on Quality. The lackof evaluation of teachersand subsequentawardsor punishmentbased on such evaluation,as well as the "credendalism* in the system brought about by the requirementfor diplomas,leads in both publicand private institutions to what has been describedas the 'pacto corrupto' betweenstudentsand teachers. For manystudents,obtaininga degree is more importantthan learning. For the teacher, the pacto means he does not have to prepare challenginglectures, carefullycorrect tests and examsand/or spendhis timeworkingwith and advisingstudents. The pacto is most common in programs of lower social prestigeas well as in night classes. In fact, giventhe nature of the system, The most flagat exampleof politicizd electionof a rector was reportedby variousBrzilian newspaper. In this casethe rector of the Federal Univasity of Rio de Janeiro(UFRJ)hired, throughtemporarycontrats, Iarge numbersof administative personnelwho supportedhim and helped to ens his e-election-in spite of the fact that universityby-lawsdid not permita secol tm After a year of turnoil the rector resid and the situation was normalized. " The case of the rector of the Universityof Juiz de Fora, also reportedin the newspapen, illustates the lack of power of the rector. This rector resigned after the UniversityCouncilpae a resolutionto incre the restaurant subsidy from 30% to 70%. With the pasing of this resolution, the subsidy for the resturat wu equivalentto 82% of the university's discretionary fimds for non-salry operting costs. M-17 it is surprisingthat manyteachers(to some extentin all areasbut especiallyin the sciences,engineering, and graduateeducation) are dedicatedand that qualityis adequate. 2 Financing of Higher Educalion' PublicFinancingof FederalInstitutions. MEC isthe mainfundingsourcefor federaluniversities, providingapproximately 85% of the incomeof the federaluniversities. MEC's total budgetfor higher education in 1989was around US$3 billion,of whichUS$2.5millionwent to federaluniversities. The way MEC determineseach institution's budget has an importantimpact on the behavior in each institution.Additionalfunding, especiallyfor graduate educationand research, is provided by other governmentagencies,as wellas by privateindustryand by localgovernment. A few federaluniversities (e.g. Brasilia,MinasGerais)havetheir own endowment (patrimonia), usuallyin the form of real estate which can be sold or developed. The budgetof federalinstitutionscanbe brokenup intotwo basic components,salariesfor staff, who are civil servants,and muchsmallerdiscretionary budgetswhichinstitutionsuse to fund everything else. Personnelpaymentsare made directlythrougb the NationalTreasuryand do not enter the budget of the institution,sinceMEC has complete controlover them. Overall94% of paymentsfrom MECgo to personneland 6% financeother costsand to capital. Theuniversityhas no controlover the 94% since they are simplybased on paymentto staff. The 6% of budgets which institutionscan control covers utilities, travel, materialsand equipment, maintenance,restaurant,and library. Ordinarymaintenance generallyaccountsfor 70% of thesefunds. At the beginningof each fiscal year, each institutionis budgetedonly a portion of its expected requirements. Each institutionreceives these funds on the basis of historical criteria, e.g., the percentageof the total budgetit receivedthe previousyear. Thefirst paymentis madeon March 1, with inflationcorrections. After six months, MEC provides a supplemental budget. Universitiesare not permittedto spend funds which have not been budgeted,so it is very difficultto plan the purchaseof materialson a yearly basis. Furthermore,the universitiesare not allowedto keep funds in interest bearing accounts,so in a high inflationperiodthe real valueof their fundsis rapidlyeroded. Purchases must thereforebe made on an emergency basis. While universitieshave a fair amountof flexibilityin determininghow to applynon-salary finds, they appearto exerciselittle or no controlson use of these funds by staff. For exampleuniversitieshave no internalcontrolon use of energy, telephones,copy machines,etc. The budget negotiationsbetweenthe federal universitiesand MEC generallyresult in budget increases.But the numberof staff and faculty members,the physical capacityof institutionsand the allocationsfor non-salaryexpendituresare made without any reference to what is expectedof the institutions.The five year periodbetween 1983and 1988exemplifies the problem. During this period, budgets rose while enrollmentsfell an average of 7%. Institutionsreceivedextra funds but decidedto maintaintheir elite status. The extra per studentfundingcouldhave been used to improveteachingor researchquality, but there is no indicationthat this has been accomplished.Furthermore,a significant part of the non-salarybudgetis used to provideextra studentsubsidies,such as discountedmeals, health servicesand other welfareservices. 12 This a descriptionof the financingsystema of January 1991. The determinationof budgets without any considerationof cost or of output results in waste. Publicmoneyhas beenused to purchaseteachingand physicalcapacitythat is not beingused. Sincethe budgetingis done withoutany attentionto outputs or costs, the funding increaseshave not resulted in perceivedqualityimprovements. Sources of Fundsfor Fedeial Institutions, Induding Cost Recovery. Publichigher education institutionsare legallyprohibitedfrom chargingtuition. They do, however, chargea variety of small fees for servicesto studentsas wellas to thecommunity and industry. Overallthe two federalinstitutions surveyedin depth get 80-85%of their fundsfrom MEC. The remaining15% comesfrom servicefees chargedto students (graduationfees, restaurant charges, etc.), contractingagencies (e.g., in-service training programs, technical assistance and studies provided to government and industry), health authorities(reimbursements for health servicesprovidedby universityhospitals), and research donors (CNPq, FINEP). Becausethey are subjectto complexcivil service regulationsregardinguse of funds, many universitieshave establishedindependent'foundations' to managemore efficientlymanagenon- MEC funds, especiallythose coming from FINEP, CNPq, CAPES, and PADCT for research and graduate education. The budgets of these foundationshave been growing and they are providingan importantelementof flexibility. Traditionally subsidize publicinstitutions medicaland restaurantservicesprovidedto students. The averagerestaurantsubsidyin UFMG was 48% and in UFSC was 70%. The average yearly subsidyin per studentwas US$31. It has beenreportedthat the subsidyat the Universityof Juiz the two institutions de Fora was 70% in 1989. The legal prohibitionagainstchargingtuitionreducesboth the efficiencyand equity of the higher educationsystem. Since studentsdo not pay for their education,they have little incentiveto complete their course-work in a timely manner. Because students in Brazil's public institutions are disproportionately from the upper economicclasses, the governmentspendsfar more of its moneyfor the benefit of the rich rather than the poor. Sincethe provisionof basicsocial servicessuch as primary educationand health care are not effectivelyreachingthe poor, fundinghigher educationonly from the broad tax base results in a regressivetransferof fundsfrom poor to the rich. Public Financial Suppot of the Private Sector. Until the end of the 1960s, the federal government provided a significantVmountof direct public support to private, especially Catholic in some years. ln the period 1970- institutions,coveringas muchas 50% of the costsof these institutions 1980 this amount changed to around 10% of expenditures. In the early 1980s, the Government effectivelyended all direct support of private institutions. However, in the period 1986-1989,the Governmentagain indicatedits willingness to support'community' institutions,increasingto 7% of the total expendituresof about20 community institutionsin 1987. In 1989-90such direct supporthas again fallen to almost nothing. The Governmentcontinuesto support graduate educationin a few private institutionsthrough FINEP, CAPES, and CNPQ. Of particularnote is FINEP's long term support of science and engineeringin the PUC-Rio,whichhas helpedmake it the best private institutionin Brazil. The Student Loan Program. Recentlythe main source of public financial support to private institutionshas been the provision of subsidizedstudent loans. In 1976, the Brazilian government implementedthe first student loan program to provide financialsupport for students to pay fees and by the Ministryof Education, executed maintenancecosts in privateand publicinstitutions. Established by the Caixa EconomicaFederal (CFE, or FederalSavingsBank),the Programa de Credito Educativo (PCE, or EducationCredit Program)has sinceprovidedloans to 712,688students. Althoughpresently M-19 PCE fundsare availablefor the paymentof tuition by (undergraduate only) studentsattendingprivate institutions,duringthe first phase (1976-83)of PCE funds were also directedto stipends, so students attendingprivate as well as public institutionsbenefittedfrom the credit line. The program has experienced severefinancialproblemssinceearly 1980sand it has been reformulated twice. Eligibilitywas prioritizedaccordingto a householdincomeformula adjustedfor the numberof 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. Becauseresources were plentifulin 1976, loans were granted to almost everyonewho applied. Once studentswere awardeda loan in a givenyear, they were automatically eligiblefor loan renewal for a period equivalent to the average length of course-work, with a one year automatic extension,if necessary. The repayment terms includeda fixednominalinterest chargeof 15% (while the annualinflationrate averagedalmost100%duringthe period 1976-83). Studentswere granted a 1 year grace period and repaymentwasto be no longerthan the periodof the loan itself. The studentloanprogramwasadministered by the CaixaEconomica Federal, and losseswere to be paid to the Caixa by the Ministry of Education.The Caixa, a banking structure with branches throughout the country,wasregardedas the mostappropriatebodyto administerthe programsinceit had extensiveexperiencewith private lending. Furthermore,it was believedthat the Caixa's controlover credit ratingswouldbe a significantdeterrentagainstdefault. By 1983,the programwas essentially bankrupt. As inflationexceeded100%,the nominalinterest rate of 15% on the loans impliedthat the loans had a subsidyof around 90%. This was so high that studentsfound themselvesobligedto pay only nominalamountsper month. The amountswere so low that many did not bother paying back their loans, nor did the Caixa pursue them since the costs of recoveryoutweighedthe value of outstandingdebt. When administrative costs and default lossesare includedin the financialassessmentof the program,it is clear that the loans were more expensivethan outrightgrants. Despite the collapse in 1983, a strong demand remained for educationalcredit. The PCE's effectiveness as a meansto bring relativelypoorer studentsinto the educationsystem,particularlythe privatesector, led the governmentto reformulateit. The first task was to securea financingsource for the program. The governmentdedicateda fixedpercentage of annualincomefrom the federallottery to the new loanprogram. In an attemptto makethe programself financing,both the eligibility criteriaand the terms of repaymentwere altered. Supportwas restrictedto needystudentsin the private sector, for the purposeof payingfees only, and credit was limitedto the maximum officiallengthof course-work Between 1987 and 1989, PCE granted approximately40,000 loans to new students each semester. Table 5 summarizesthe experienceof those three years. During the secondphase of PCE, the termsof repaymentwere tightenedsomewhat,but a large subsidyremained. Considering terms, the high inflationrate of the periodand the (modified)repayment the interest subsidy on the program amountedto over 55%. This subsidy level was a substantial reductionover the previousone, but not enoughto makethe programfinanciallyviable. A few other reforms were introducedin the secondphase of PCE, includingthe requirementfor a loanguarantorand the paymentof an up-frontinsurancefee. Both these changeswere believedto be importantsteps to minimizelossesthrough default and evasion. Nevertheless,the program continued beingfinanciallyunsustainable exacerbated becauseof the high interestsubsidy. Hyperinflation program m-20 losses and the insufficiencyof dedicatedlottery income again brought the program to the verge of bankruptcyin 1989, when the programwas again reformed. Two importantchangeshave been madeto reducethe interestsubsidy. First, the interest charge willbe 6% abovethe BTN (a new indicatorof inflation). Second,the periodof repaymenthas been cut in half, to equalthe periodof borrowing. on the PCE LonProirmm by hs IneetSubsidieg , i-~~~~~~~~~~~~ . AX- Phase .~ i - S\ = 9I_- % - :. :7% i 5---- - -C5%-; '' - '"'' fX000--. "'... ..... . ... ... ... ..... ''' '''' :0S! ; ; ,;iS -::, ..i:. S-. . ,i .... ':; . ti-'- i, , ,, f: ,,., : : , ,...,. . .... ; E i SS .i. i: i i i;S;j: . ....... ~~~~~............................ :; . . 9, 9.. ............. . -S.-g- S.-999.- "-0o") i;:02-e::i:_ S9g5 9~~. i9....... .......... . i S 9 9 S : aN9 Szg9' lo4ns ar asuoed A:l to beS1,000peryear, u5 forinhdon. Subsid.es a.c alad by g areadiount Ale:;: of 3% ad ahd.ayEs ypesdnfli y. Por IbseUadI ons, e E:>YN andT>BTNacontribuio for aptalSconDSt ructiO from:n governme .and ndstry... ...currt ,. .e ,;--onstructionprogram c was80%finane bygovernment and20% byprivate inidustry.The physica.l$S --aci.itiesare'a,dequatein size:and wellmanaie and 'manicured.'. UABsalaries"-are equivalent'-.o publ*iesecor.'crie b iteh' ot.hedlarge.:n'7::.',e amountsoffrbingS bn't "1em' on- "e .af memerfo e 0 students ecaeo ac of f sB oes researc only. :in> .:the B'bil -- nsprovEides anamnder,graduate:-.education:-wi - lo is' equtialen-t fthe beNs 0.:-- . pitbltc -. .- 0 0 .1987 -figures,S% :ofenering ;.:, nsttutonsA'ccording to -. '-'' . .',.-'y-:fLX '.... .''0'. stdents 't?--:0,-''''' . on the- }; scored above 60j entrance :aminat-:ns. 81 of enteringstudentsere wTlhe in :the: meddleorupper 3 classes. institutionhas :10,000 applicants for 2,00 spaces the firstyear of education. .Students at.UAB seelcto complete in a reasonable thieir--studies periodof timeandentertheclaborforce. .Based On: -enrollments :: .- live years.earlier,.approxcimatelyG33%eof enteringstudents graduate.However, therebpo :::islittlerepetitioandstudents-who failusuallydropout. T eccostper graduate is aboutUS$4,800.iD0 le totalbudgetofUS$6.6millionresultsiabout MO 1100 graduates. IV-IS TheUniversityof Carabobo (UC) is a publicuniversity with problemswhich,albeitsimilar to be more severe. UC currently enrolls 52,000. to those of.other public institutions,.appear. students',whichmakes it one of the thireelargest publicinstitutionsin Venezuela. However,it is estimatedthatit least'10% of these students do notzatend classes. Ihe physical facilities at .. Caraboboare extremely"outdated and.inadequate..t has.0.8 square meters for each enrolled,' stdent,: a figurewhich'is far'lower thanthe.interiona y a epted re of five'sare meters..; -perstudent andis.lower than any other institution in Venezuela. .Pysicaly most buldingsaren', :.:deplorable:shape and'teachingequipment.anda lc'lkofalibrar. withitotallynindequiatescientific been unableto resist pressur.sto accept'largenumers of unquaified sudent. '.boos.-.UC,.has' Onlyabout"I o enteringstudentsscred Abve 60'in the entra'n exaia . About47*%are' . in the -middte:or uper classes.. * e' situation -isuch hat' on 17% of ene ng'meering "students, of which there are 1000,passed the end.ofthe first yearexamination.:. Similarly:'UCwas., accepting800per year into medicine,whilethe"insitution has agreements with the local hospitals -to train a maximum of 100. There'are8500studentsenrolledin'the facultyof law. Utl' lastyear.. studentswere permittedto continuetheir studiesno matterhowtheyiscoredonthe,finalexamination.. -andno checkswere madeon howmanyclasses theyattended.,.. ;...... -About.19%ofienteringstudentseventuallygraduate. Becauseof high repetition rates, the -system 'requires i 222shoolyes to rod onegrde.., Crty; g adu om o medicine take :avcrageeof9aOf half years colete the 6 year cour.. h education and 'engineering those who`docompd the.:courso ' aveeage .sixeao..ac''coralte. ' e.r lei erall: "the fitcos U$100rnslteintoaompc raduate-couo. about S o' 0. -- aks tam6t .... ~tih ourti esusex .... eniv.a ~~~~~~~~~~. . . ;-;. . -. i. ..-- toal.bud iii. .. ... . :- ut etsao US 5 ilo was ,-nis fraed-ofu.lages in tne sc000 studentsproests. it mmMcr schooli" the ofare'als -'- o' amnthe. and famuntyof000, Fmealschooly f acrcepted200 rater tag nitsi permcitted'amountiof10om0 this wadiinthe face'of large S. e st;ulet protests.A numberof theichool are alssponsoring pre-,, ente oursesto helpraise up thestandard-of c nteringstudents. ofC is also fial biing tomenrce rules f readONITboo 'student attendance, for requeiraiyg stuidents to pass at least50%,,.. of the cou-re work or -to drop out To uvimdroadinistration, ec has establishedw a pivate"C foundationto administratethe library aed to:p p '-va d - - ety of infort sy'stemsfor'students:..' faculty,andadministrators. Fnallythe UChas recevived'aagreement inprincijplefrom government. that additionalfacilitieswfil,bebuilt. The total'costs wouldbe :in. excess,OfUs$50 million.... 'Currentlthe faculty of lawis under:construction., lDespite beingoneof the weakest pulcuieste ntrsof research,UChasatrce some fundsfrom CONICrr for engin'eering research,'especiall in healthiand environmentally,. relatedareas., A totalo6f 3% of UC's budget'isallocated to research. In.accrdancewit gUidelines, I75 ialoated to libraries. IV-16 OPTIONSFOR REFORM To create an environmentthat will provide the necessary incentivesto improve quality and efficiencyin the publicsystem, enhancefacultyperformance,and encourageself-assessment, a new set of transparent funding criteria and mechanismswill need to be introduced, based on how much the Governmentestimatesthe public system should cost rather than on how higher educationinstitutions would like to spend such funds. As noted above the CNU is already discussingmoving away from incrementalfunding through establishinga formula which combinesdifferent criteria for allocating resources to teaching,research and extension.However, the current attemptshouldbe consideredonly a modestattemptat settinga formulawhichwouldprovideappropriate signalsfor institutionsto improve their efficiency;giventhe collegialnature of the CNU eventhese modestattemptsmay fail. While the currentproposedformulawouldhelp to regularizeuse of staff time, would reducethe number of non-teaching staff, and wouldput researchsupporton a systematicbasis, it wouldnot deal with the most fundamental problemof publichighereducation,whichis high dropoutand repetitionrates and the excessivetime i takces for studentsto graduate. It is highly inefficient,both for studentsand for Government, for public institutionsto accept large numbers of students and then to fail them repeatedly. The simplestand most powerfulformulato encourageimprovedefficiencywould be one which financesinstitutions,at leastpartially,on its numbersof graduates. A modelwhich couldbe considered would be that of the Netherlands,which has an utomaticadmissionspolicy. The proposed Brazil formula could also be adaptedto Venezuela. In the Netherlands'fundingformula, universitiesreceive 4.5 years of annmal unit cost fundingper graduate and 1.5 years for dropouts-regardlessof how long studentstake to completetheir studiesor at whatpoint in their studiesthey drop out from the institution. The dual incentivesare both to weed out poorly performingstudents early on, and to get people to graduate as quicklyas possible. Odter criteriashouldbe based on performanceand efficiencyin the use of resources and should help ensure that institutionsdo no arbitrarilyaward degrees to get increased funding. The results of such changeswouldbe a major reductionin the costsper graduate and a much higher social and private rate of return to investmentin higher education. The implementation of such fundingmechanisms wouldrequirea changein the status of the CNU, which shouldbecome independent of the institudonsh oversees. A majorityof the CNU would consist of representativesof Govermment and societyas a whole. 'Buffer' institutionsof this sort have been shown to work well in the UnitedKingdom,Netherlands,and the UnitedStates. There is also a strong case for reducingthe absolute amountsof public expenditureon higher educationand for increasingthe private contributions.The most straightforwardway of achievingthis aim would be to charge tuition fees to all students in public institutions and simultaneouslyoffer subsidizedloans and scholarshipsfor the needieststudents. At a minimumsuch fees couldbe equal to the current charges in private secondaryschools which one third of public university students have attended,and loan and scholarshipscouldbe offeredto the remainingtwo thirds of students. Venezuela currentdyhas some small but relatively well organized student loan schemes (FundacidnAyacucho, Educredito)which couldbe expandedunder this scenario. As in most of Latin America there is very strong oppositionin Venezuelato chargingtuition in public institutionswith their long traditionof free education. It has beenarguedthat cost recovery could IV-17 alsobe achieved through a graduateincometax system,in whichgraduates of publicinstitutions would pay, say, an additionaltax equivalentto 29%of their incomeover a fifteen year period. An alternative currentlyunder discussionin Brazil wouldadd a portionof the costs of public higher education tO the gross incomereportedon tax returnsof parentsof studentsattending publicinstitutions,who wouldthen be taxedat the currentmarginalrate. Howeverboththesealternatives requirea better functioningincome tax systemthan Venezuela'scurrentsystem. In addition,higheradmission standardsin publicinstitutions will resultin increasedprivatesector growth. This is alreadyhappeningunderthe pressureof reducedfundingfor public institutions. Under this scenario public universities would increasinglybecome higher quality, research and graduate educationorientedinstitutions. This policywouldalso require expandedloan and scholarship schemes for needystudentswishingto attendprivateinstitutions. Increasedfundingon the basisof opencompetition for innovativeresearch,teachingand extension programsshouldalso be explored. The recentexpansion of fundingby CONICITon the basis of open competitionfor researchcontractsis a strongstep in the right directionbut more can be done by CNU itself in areas such as extensionand innovativeteachingprograms. Along with fundingmechanisms based on formulae,contractualarrangements,studentfees and other similar provisions,it will be necessaryto strengthencurrent initial efforts towardsutilizationof institutional self-assessment proceduresas well as the use of externalevaluationsand accreditation.Such procedures,as well as the reportingto the publicof results, can serve to guideresourceallocationand to createfurther incentivesfor qualityand performanceimprovements. In additionto these system changes,Venezuelaneeds to examinecloselythe role of the higher educationsystem in trainingand upgrading primaryand secondaryschoolteachers. The current system seems to be succeedingin providingbachelor's degreesto large numbersof currentlypracticingand futureprimaryschoolteachersbut withlittle discernibleimpacton studentlearningor reducedrepetition or dropout. It maywellbe appropriate to reducethe requirementsfor primaryschoolteachercertification from those of the five year bachelor's degree to a three year 'certificate' while at the same time refocussingpre-servicetrainingon increasedpracticeteachingthrough'sandwich' coursesandsupervised internships. Another way that the higher educationsystemcan providea serviceto the lower levels wouldbe to undertakeresearchon the resultsof the universityentranceexamination and then reportthe resultsof this researchto secondaryschool teachersand schooldirectors. This researchcouldidentifythe higher order thinkingskills in whichstudentsare deficient. The resultswouldthenbe incorporated into in- and pre-servicetrainingprograms. CONCLUSIONS The adoptionof fundingmechanismsbased on formulas,contractualarrangements,studentfees and other similarprovisionswould greatlyenhancethe alreadyemergingtrend towardsthe utilizationof self-assessment institutional proceduresand the continuoususe of externalevaluationsand accreditation. Evaluation, at its simplest need not be linked to funding criteria but it can serve to guide resource allocationandlor create incentivesfor quality and performanceimprovements. Anotherpolicy option which would help ensure that higher educationserves societymore effectivelywouldbe reformof the IV-18 systemof trainingteachersto emphasize lengthy increasedpracticeteachingandto shortenthe excessively periodof trainingprimaryschoolteachers. In addition,the CNU couldundertakeresearchon why some students do better on the university entrance examinations, as well as remedial programs for underprivileged studentsdoing poorly on such examinations. IV-19 REFERENCES Albrecht,Douglasand Ziderman,Adrian, FundingMechanisms for Higher Education: Financingfor Stability.Efficiencyand Responsiveness TheWorldBank, DiscussionPap, Washington:1991. Im Plan Nacionalde Cienciay Tecnologfa, CONIC1T, Caracas: 1991. Navarro, Juan Carlos, WVenezuelan Higher Educationin Perspective,' Higher duclati2n 21(2):177-8, the Netherlands:March, 1991. OPSU (Oficina de Planificaci6ndel Sector Universitario), CNU (Consejo Nacional de Educaci6n Superior),BoletfnEstadfsticode Educaci6nSugeriorno. 13, vols. I-IV, Caracas: 1990. Psacharopoulos,George, and Fizbein, Ariel, 'A Cost-Benefit Analysisof EducationalInvestmentin Venezuela,1989,' The WorldBank, A View From LATER, Washington:1992. Reimers,Fernando,'The Feasibilityof IntroducingLoanSchemes to FinanceHigherEducationin Latin America: The caseof Venezuela, The WorldBank, unpblished paper, Washington:1990. UNESCO - CRESALC,Visi6n gualitativade la educacidnsu2erior en America Latina I el Caribe, Caracas: 1991. l~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ - Continued from back Page No. 19 'Inve sme in Science Researchand Training:The Case of Brazil and Implicationsfor Other Countries'by Laurence Wolff, with contributionsfrom George Psacharopoulos,Aron Kuppermann,Charles Blitzer, GeoffreyShepherd, Carlos Primo Braga and Alcyone Saliba, September 1991 No. 20 'Prenatal and Perinatal Health Carc- A DiagnosticInswnment- by Francisco Mardanes, September 1991 No. 21 "MaternalAnthropomeny in Prenatal Carm: A New Maternal Weight Gain Chasrt by Pedro Rosso, September1991 No. 22 "PovMy and Inequawlityin Lain America and the CaribbeanDuring the 70s and 80s: An Overiew of tie Evidence' by Dominiquevan de Wale, September 1991 No. 23 'Social Indicators in Lxin Amenrica and the Caribbean.A Compilationof Statistisfrom 1970 to the Present" by George Psalaropoulos and Bill Wood, October 1991 No. 24 01CElEX - A StudemtLoan Succss Story in Colombia' by Samuel Carlson, October 1991 No. 25 'EducationalDevelopment and Costing in Meico, 1977-1990:A Cross-Statre lne-Series Analysis' by Juan Prawda and George Psacharopoulos,November 1991 No. 26 OACost-Benefi Analsis of EducationalInvestment in Venezuela,1989' by Ariel Fiszbein and George Psacharopoulos,November 1991 No. 27 'Educational Decontr ion in Latin AnriL Lessons Learned" by Juan Prawda, March 1992 No. 28 'Education and the Labor Market in Uruguay' by Geore P and Eduardo Velez, lcharopoulos June 1992 No. 29 'Some OperationalGuidelinasand bsus for the Evduation of Investmentsin Pmwy Health Care in the IAC Aegion' by Francisco Mardonesand Barton Burkhater, June 1992 No. 30 'The Economics of Higher Education in Brazil' by Jean-JacquesPaul and lAurence Wolff, July 1992 No. 31 'Repetition and InadequateAchievanent in Latin Amenrca'sPrunay Sdwools: A Rview of Magnitudes, Causes, Relationshipsand Strategies- by Enesto Schiefelbeinwith Laurence Wolff, August 1992 No. 32 'Early ChildhoodDevelopmentPrograms in Latin America: Toward Definition of an Investment Strategy' by Robert G. Myers, August 1992 No. 33 'PrimarySdiool Quality in Mceico' by Juan Carlos Palafox, Juan Prawda and Eduardo Velez, November 1992 No. 34 'Higher Education Wfonmin Chile, Brazil and Venezuela. Towards a Redefinition of the Rok of the State' edited by Laurene Wolff and Douglas Albrecht, November 1992. Views from LATHR No. 0 'lhc Magnitude of Poverty in Latin America in the 1980s' September, 1990 No. I 'An Ounce of Prevention is Worth How Much Cure? Thinkingabout the Allocation of Health Care Spending' by Philip Musgrove, September 1990. No. 2 'Decentralizationand EducationalBeraucraciaes- by Juan Prawda, November, 1990 No. 3 'What Should Social Funds Finance?.: Portfolio Mix, Targeting,and Efficiency Citeria' by Margaret E. Grosh, December 1990 No. 4 "Financal Balanc in Chil: The ISAPRES (Institudones de Salud Previsional) Health Care System and the Public Seaor" by Philip Musgrove, Janury, 1991 No. 5 *Popukation,Health and Nutrition Issues in the Latin Amerncanand CaribbeanRegion and the Agedirfor the 90's' by Oscar Echeverri, January, 1991 No. 6 'Population and Family Planning in the 1990's: ReconcilingMacro and Micro rsues' by Boce D. Carlson, February, 1991 No. 7 "TheFeasibility of Student Loans in Latin America: A Simulation' by Samuel Carlson and GuozhongXie, March, 1991 No. 8 'Jansforming the Viidous Orcie - The Costs and Savings of School Inefficiency in Mcdco' by Samuel Carlson, April 1991 No. 9 'Colombia's "ErcuelaNueva": An EducationInnovation' by Eduardo Velez, May 1991 No. 10 'Health TechnologyDevelopment and Assessment:Do LAC Countries Have a Choice?' by Oscar Echevei, June 1991 No. 11 'The RecurrentCost Faaor in the PHR Sector' by Jacob van LutsenburgMam, July 1991 No. 12 "TheBurden of Death at Differm Ages: Assumptions, Parameters and Values" by Philip Musgrove, August 1991 No. 13 "GovernmentExpenditureon Social Secors in Latin America and the Caribbean. Statistical Trends' by Hongyu Yang, August 1991 No. 14 'From Manpower Planning to Labor Market Analysis' by George Psacharopoulos,September 1991 No. 15 'An Update on Cholera in the Americas' by Francisco Mardones, August 1991 No. 16 'What do we think about Health Care Finance in Latin America and the Caribbean?' by Philip Musgrove, September 1991 No. 17 "PopulationGrowth, Ezernalities and Poverty" by Nancy Birdsall and Charles Griffin, September 1991 No. 18 'Wage Trendr in Latin America' by Alejandra Cox Edwards, September 1991 - Continuedon inside Page