Malawi - Financing and efficiency of education

Report No. 4894-MA1 Financing and Efficiency of Education in Mvalawi June 25,1984 Edu ation dfnlJ Manpower rDevelopmtnit Division Easterm Afrin' Regional Office FOR OFFICIAAL USE ONLY Document of the World Bank This document has a restricted distribution and may be used by rec ipients only in the performance of their official duties. Its contents may not otherwis. be disclosed without World Bank authorization.  CURRENCY EQUIVALENTS Currency Unit =Malawi Kwacha (MK) US$1.00 =MK 1.25 MK 1.00 US$0.8 ABBREVIATIONS JCE - Junior Certificate of Education MCC - Malawi Correspondence College MCE - Malawi Certificate of Education MCETB - Malawi Certificate of EducaLlon and Testing Board MOF - Ministry of Education PAYE - Pay As You Earn PSLC - Primary School Leaving Certificate UPE - Universal Primary Education REPUBLIC OF MALAWI FISCAL YEAR April 1 - March 31 2YALEESY2  FOR OFFICIAL USE ONLY RFPUBLIC OF MALAWI FINANCING AND EFFICIENCY OF EDUCATION Table of Contents Pa,e No. BASIC DATA ............................................ (1) SUMMARY AND RECOMMENDATIONS .............. ...........(i) I. GENERAL BACKGROUND Recent Economic Developments .................................B Education Financing and the National Economy .................o I Objectives of the Study ......................... . . 4 II. MANAGEMENT OF EDUCATION Introduction 5 Education Administration 6 Issues in Education Management 7 Financial Management 8 Issues in Education Budgeting .c 10 III. COSTS AND EFFICIENCY IN PRIMARY AND SECONDARY EDUCATION Introduction .. ................. .. 17 Primary Education 18 Secondary Education * 22 IV. PRIVATE RESOURCES AND COSTS OF PRIMARY AND SECONDARY EDUCATION Introduction c cc... cc. , , 26 Patterns of Private Expenditure , 27 Continuation Rates Under Hypothetical Fee Increases.e......... 28 Willingness to Pay: Household Outcomes 29 Further Evidence on Willingness to Pay 30 Recommendations 32 This report was prepared by Mrs. Nwanganga Shields, Economist, EAPED, and Mr. Laurence Wolff, Sr. Education Planner, EAPED (Sections III and V), and is based on a mission to Malawi in April 1983 in which Mo. J. P. Tan, Economist (EDC), participated. The report was discussed with the Government in March 1984. This document has a restricted distribution and may be used by recipients only in the perforinance of their official duties. its contents may not otherwise be disclosed without World Bank authorization,  Page No. V. COST EFFECTIVENESS AT THE UNIVERSITY OF HALAWI Introduction .... . .......................e ...... .33 The Comparative Costs of Higher Education .................... 33 Analysts of the Costs of Higher Education in Malawi .......... 34 The Present Situation and Recommendations for the Future *.... 39 VI, SOCIAL EFFICIENCY AND EQUITY CRITERIA FOR DECISIONS ON INVESTMENT IN EDUCATION .....*.* *. *. ...... ........ .. ...... 40 Rates of Return ; General Comments ........................... 41 Ratee of Return by Level of Education in Malawi ........e... 42 Recommendations .... ... ...... * l , 45 VII. COST IMPLICATIONS OF ANALYSIS Introduction .............. ...................... . 46 Enrollment Targets for the Medium Term ....................... 48 Recurrent Cost Projections .......................... 51 ANNEXES Table T-1 Functional Classification of Total Government Expenditure Table T-2 Distribution of Recurrent Expenditure 1971/72 to 1982/83 Table T-3 Recurrent Education Expenditures as Percentage of Total Education Expenditures and of GNP - East Africa, 1970-82 Table T-4 Distribution of Unit Salaries Cost, Pupil:Teacher Ratio and Qualified Teachers by Districts, 1983 Table T-5 Primary School Fees and Textbooks - Costs per Child Table T-6 Estimated Unit Costs to Governments of Public Education Table T-7a Population Growth, Expansion of Primary Education, Teacher Requirements and Costs, Kalawi 1980-2015, Constant Fertility Rates Table T-7b Population Growth, Expansion of Primary Education, Teacher Requirements and Costs, Malawi 1980-2015 Rapid Fertility Decline Table T-8 Typical Pattern of Teacher Utilization in Secondary Education and more Efficient Alternatives, Malawi Table T-9 Total Governaent Expenditure on Primary, Secondary and Univeralty Education - Malawi, 1980-83 Table T-10 Mean Annual Private Expenditure for Schooling by Level and Selected Sample Characteristics, Malawi, 1983 Table T-lla Percent of Students Expecting to Continue SchoolingAfter Hypothetical Increase in Fee, Various Sample Sub-group, Malawi 1983 Table T-1lb Maximum Likelihood Estimate of the Determinants of Expectations of Staying on in School When Fees are Hypothetically Raised to Various New Levels, Malawi. 1983. Table T-12 OLS Estimates of the Determinants of the Proportion of Eligible Siblings Enrolled, Malawi, 1983 Table T-13 School Fee Structure in Malawi, Before and After 1982 Table T-14 Percent Relying on Various Sources of Educational Finance and Mean Dropout Rate Among Respondents' Siblings Due to Recent Fee Increase, Malawi, 1983.  Table T-15 Expecte. and Actual Mean Annual Earnings and Their Growth Rate By Educational Qualification and Sex, Malawi, 1983 Table T-16 Expected Returns to Secondary Education Under Alternative Assumptions by Sex, Malawi, 1983 Expectations of Staying on in School when Fees are Hypothetically Raised to Various New Levels, Malawi, 1983 Table T-17 Estimated Unit Costs to Governments of Higher Education Table T-18 Student:Teacher Ratios in East African Universities Table T-19 Per Capita Income and Unit Costs of Higher Education by Region, 1979 Table T-20 HIighar Education Enrollment Rates, East Africa, 1980 Table T-21 Employment of Malawian and Non-Malawian Middle and Upper Level Manpower, 1979 Table T-22 Finance Study University of Malawi Expenditure Breakdown, Table T-23 University of Malawi Student:Teacher Ratios and United Kingdom Guidelines Table T-24 University of Malawi First Year Enrollment (1982/83) and Parents Occupation Table T-25 University of Malawi Enrollment, Teaching Staff, and Costs 1980/81 to 1982/83 Table T-26 University of Malawi, Staff and Students, 1982/83 and Number of Students Who Could be Served at Higher Student: Teacher Ratios Table T-27 Primary Education: Enrollment Expansion, Implications for Teachers, Student:Teacher Ratios, Recurrent Expenditure 1982/83 to 1987/88 Table T-28a Secondary Education - Expenditure Analysis in 1982/83 Prices Table T-28b Secondary Education: Expenditure Analysis Assumes no Reduction in Student:Teacher Ratio 1982/83 Prices Table T-29 Malawi Correspondence College Expenditure Analysis in 1982/83 Prices Table T-30 Teacher Training: Recurrent Expenditure Needs 1984/85 to 1987/88 in Constant 1982/83 Prices Table T-31 Technical Education Expenditure Analysis in 1982/83 Prices Chart 1 Percent of Education Expenditure by Level and Type 1971 to 1983 Chart 2 Malawi - Structure of the Education System Chart 3 Organization of the Ministry of Education Annex Note 1 Data and Sample Characteristics of 1983 Survey of Primary and Secondary Students Annex Note 2 Methodology and Assumptions for Calculation of Rates of Return Annex Note 3 References wD  (1) REPUBLIC OF MALAWI BASIC DATA General Area (including lake) 118,000 sq.km Area (land only) 94,000 sq.km Population (1981) 6.1 million GNP Per Capita (1981) US$200 Annual Growth Rate of Population (1966-1977) 2.9% Education School Year October-August Adult Literacy Rate (1977) 25% Gross Enrollment Ratios (1981): Primary Education (6-13 age group) 62.0% Secondary Education (14-17 age group) 4.0% Higher Education 0.4% Central Government Expenditures on Education as Percentage of Total Government Expenditures (1981) Capital 11.9% Recurrent 11.2% Total 11.4% Total Ministry of Education Expenditures as a Percentage of GDP (1981) 3*5%  (ii) Summary and Recommendations 1. Since 1978/79, the Government of Malawi has been facing serious economic difficulties due to a combination of factors, including the world econom:'c situation. Public expenditure has grown more rapidly than increases in revenue, Although education expenditure in nominal terms has been increasing since 1971/72, its share of total Government expenditure has been falling. The education budget is extremely vulnerable during periods of economic difficulties of the type experienced by Malawi, in that projects which have short term returns and make immediate contributions to reduction in deficits and solution of the balance of payment problems are favored. This is particularly unfortunate in the case of Malawi since only 62 percent of the eligible age cohort attend primary school and only 4 percent of the age group 14-17 attend secondary school. In terms of the labor market, the 1979 Manpower Survey revealed that 13 percent of the professional and technical posts and 43 percent of the administrative and managerial posts were held by expatriates. Education needs are far from satisfied. This study focuses on two issues in education financing in Malawi - (a) efficiency and cost effectiveness of the system and (b) prospects of cost recovery at all levels. 2. The development of the education system in the 1970s and 1980s has led to sizeable growth in the budget and establishment of (MOE) as well as its overall responsibility. This requires careful planning and efficient administration. MOE's administration can promote efficiency by the method It employs to transmit decisions on important issues. Faulty and delayed information is costly in time and efficiency. Currently, many directives such as those on classroom size and teachlng loads, are not implemented because those implementing them either do not understand the directives or are not committed to their goals and general objectives. Furthermore, although effective planning requires some measure of centralization in order to avoid duplication and waste of resources, the Planning Department of the MOE is understaffed and weak and unable to provide effective leadership. It is recommended that (a) the Planning Department be strengthened by additional qualified staff, and (b) the MOE should ensure that principals at the school level understand the objectives Xand goals of any directives issued from headquarters which theY are required to implement. Action should be taken to reduce the unit cost of Malawi Certificate of Education and Testing Board (MCETB). A study of its cost structure and organization should be undertaken. In addition, MOE should consider merging the Examination Unit within the Ministry with MCETB. Prior to implementing this proposal, a study of the most efficient method of administering examinations should be undertaken. In order to ensure that the University implement agreed upon Policy, the MOE should be strengthened through establishment of a unit charged with oversight of higher education.  (iii) 3. The process by which budget decisions are handled affects the efficiency and equity of any education system. The current method of allocation of resources to primary schools is neither efficient nor equitable. Recurrent costs of primary schools are jointly funded by Central Government and the parents. School fees are used for the purchase of books and school materials, while the Government is responsible for teachers' salaries. At present, Government expenditure per child per district varies widely, mainly because of the qualification and length of service of teachers. Rural schools are at a great disadvantage in that they invariably have higher pupil:teacher ratios and their teachers are less qualified than urban schools. Improvements In these areas would increase unit costs, which are very low by African standards. To begin with, the Government should adhere to strict administrative procedure to ensure that qualified teachers go to rural areas. However. as an interim measure, posting of qualified teachers to urban areas should be frozen until all rural schools have at least 80 percent of their teachers' certificated. In a period of serious financial constraints, the Government should not assume major responsibility for primary school construction but should direct its limited funds towards the most needy communities. It could also provide technical assistance in construction and management techniques to such communities. 4. Although school fees are to be used for purchase of textbooks and school materials, evidence presented indicates that the Government is currently collecting on the average more than the annual cost of these books. To enable the Government continue to cover the cost of instructional and school materials, it is recommended that primary school fees be ralsed annually by the expected rate of inflation and 80 percent of the fees should be set aside for textbook replacement. For budgetary purposes, secondary schools are classified as Government boarding and day secondary schools and grant aided secondary schools. Grant aided schools receive grants from the Government to cover some of their expenses. Grant aided schools are not required to account for the use of the grants. This policy should be changed. The method of financing Government and grant aided schools discriminates against students who attend the latter. It is recommended that -he Government should move towards full cost recovery of tuition and boarding costs of secondary schools. This would require substantial Increase in boarding fees and annual adjustment of all fees in line with inflation rate. Secondly, the Goveirment r.hould ensure that grant aided schools, submit as required each year. their audited accounts for the year. 5. At present, the efficiency of primary education in Malawi is low. Only about 23 percent of children who enter the first grade complete the full cycle in eight years. The cost per gradutte is twice the theoretical cost if there were no drop-outs or repetition. The task of improving and expanding primary education requires significant additional human and financial resources. The most urgent lon term task as already stated is to reduce the student:teacher ratio to 50:i and at the same time increase the percentage of trained teachers. In addition as an Interim measure, the Government should consider:  (iv) (a) in-service training of school headmasters and supervisors on proper control and accounting of school fees; (b) establishment of minimum acceptable standards for physical quality of textbooks, teaching equipment. and furniture; (c) establishment of bulk procurement procedures and competitive bidding for bulk purchase of textbooks and other school materials; and (d) encouragement of double shifts to ensure maximum utilization of school buildings. In addition to all these, there are other relatively inexpensive means of improving qualit.y for instance, improving the PSLC examination. Suggested improvements include: (a) preparation of examination questions by qualified experts to ensure that the examinations measure achievements of curriculum objectives; (b) partial automatio'nof examination questions through use of multiple choice format; (c) handing over test administration to MCETB as means of professionalizing the testing process. There is also a need to reduce drastically the repetition rate at the eighth grade level. Recommended steps include regulating the number of times a student can take the PLSC examination and charging multiple repeaters higher school fees or requiring them to register with the MCC. Implementation of these recommendations in 1982/83 would have resulted in a 20.6 percent increase in budget allocation to primary education 6. Evidence presented in Section 3 shows that the financial burden of primary education on the Government can be lowered significantly by adoption of policies leading to decrease in current levels of fertility. If recommended quality improvements in this study are initiated and unit cost increases substantially, by year 2000 the proportion of GNP devoted to primary education would be 25 percent lower than it would be if an attempt was made to reduce fertility rate. It is, therefore, recommended that the Government should consider seriously adoption of policies which would lead to significant reduction in fertility rate. 7. The quality of secondary education is relatively high and the proportions of students who pass the JCE and the MCE have remained constant over the past ten years. However, only 4 percent of Malawi's population 14-17 years old attend secondary schools. This ratio is among the lowest in Africa. Malawi needs to expand secondary enrollment. significantly. Contrary to primary education, there are opportunities for savings in unit costs of secondary education which, if realized can be used to expand enrollment. The following areas of improvement are recommended: (a) increasing student:teacher ratios from 21:1 to 25:1 by increasing teaching load to 3U periods, by reducing the number of class periods to 32 and increasing the length of teaching period to 50 minutes or by increasing average student class size to 40; (b) reducing in boarding places; (c) expanding the production of diploma level teachers and limiting the production of degree level teachers; and (d) expansion and improvement of the MCC.  (v) 8, At present, 36 percent of the recurrent cost of secondary school and 2U-25 percent of recurrent cost of primary schools are recovered through fees. The central issue is whether or not additional funds can be tapped from private sources. This to some extent depends on the willingness of parents to sacrifice for their children's education as well as other competing options for the families' resources. The examination of patterns of private expenditure on education shows that school fees are only a small poitLon of total private expenditure on education. Given the current level of feeb, families are generally willing to pay for education by foregoing other expenditure and by marshalling resources within the extended family network, and even resources from outslde sources. ThiB finding reinforces the suggestion that primary school fees be raised annually by a rate equivalent to the rate of inflation. It is also suggested that the Government should consider other options for supglementing the funds available for primary education. For example, there should be an age limit for primary school attendance. Student who have repeated the final grade could be required to continue their education at the MCC where the fees are six times that of grade 8. There should be substantial fee increases for grades 6-8 in urban primary schools. To minimize inequalities, local authorities should be required to offer assistance to indigent children. Section four reinforces the recommendations in paragraph 4 above, regarding tuition and boarding fees at secondary level. 9. Higher education in Malawi accounts for almost 25 percent of total cost of education. No fees are charged and all students are given allowances for transport, books and clothing. As currently administered. There is considerable room for improvement in the organization of the University. The following suggestions have been put forward: (a) increasing the student:teacher ratio from 7.9:1 to 10.1; (b) reducing administrative costs from 10.8 percent of the budget to 10 percent or even further; (c) reducing salaries paid to teachers overseas who are also receiving other moneys; (d) reduction of the travel and transport costs from 10 percent to 8 percent of budget; and (e) elimination of student allowances. Implementation of the above would have brought a saving of about 20 percent of the l980/81 higher education budget. 10. Given the poor educational base of Malawi at the period of independence, decisions on the distribution of investment by level of education were not based on consistent economic criteria but rather on what type of external aid was available and what levels of education, external donors were prepared to support. This type of investment decision was responsible for the relative shares in the education budget of each level of education. One way to assess the distribution effect of resources to education is to compare the profitability of investment as it accrues to individuals and to sociRty using the best available data. This is done by use of cost benefit analysis. Private and social rates of return were calculated for Malawi using the data from the Urban Household Sample Survey of 1981. The results show that the rates of returns for salawi are, different from the average rates for Africa. At the prlmary level, the low rate for Malawi is a reflection of the high private coVJt of primary  (vi) education. Furthermore, the estimates indicate that private returns on University education are high as a result of the very high subsidy. The social rate of return of 20 percent for University education shows that it is still a good investment. However, the divergence of social and private rates of return of about 50 percentage points reflects the overall level of subsidization of university educaclon, suggesting that cost sharing at the university be brought in line with other levels of education. The social rate of return on investment in secondary education is almost one and a half times that of university and is much more than that of primary indicating that Investment in secondary education at this point in time is a better investment than in any other level. The analysis suggests that the current method of financing primary, secondary and university education discriminates against those students who attend only primary and secondary schools. If fiscal constraints on education expansion are to be eased by shifting part of the cost of education to private households, the most promising level to start charging fees is the university level. It Is therefore recommended that since it would be politically difficult for the Government to discontinue payment of allowances to university students immediatelX, fees should be set at a level higher than the maximum allowances received by Btudents. Furthermore, consideration should be given to charging fees for recovery of boarding costs, with students required to pay an amount equal to the cost of their feeding similar to the requirement for secondary students. Within the next ten years, there should be a move towards imposing tuition fees on university students. Implementation of this scheme should begin with incoming students. To minimize the hardship on truly needy, a loan scheme should be established. The scheme would enable poor students to borrow against their future earnings. 11. Overall the recommendations made in this study would result in (a) an increase of 20.6 percent in the unit costs of primary education; (b) a decrease of 6.2 percent in the unit costs of salaries in secondary education; and (c) a decrease of 25 percent In the unit costs of higher education. If these measures had been implemented in 1982/83, the education budget would have increased marginally by less than 1% because of the higher proportion of the budget devoted to primary education. 12. A switch in resources in education can take place in two ways: either existing proportions between levels can be changed or the rates of future growth can be redistributed. The Government's enrollment targets for the medium term, by restricting the rates of growth of the university and primary teacher training and expanding eurollment at primary level and at MCC, tremendously would increase the share of the education budget devoted to the primary level and MCC. However, a combination of this policy with policy suggestions in this paper would bring about a more equitable and efficient redistribution of resources than would be achieved by manipulating the future rates of growth of enrollment alone. However, if a significant expansion of primary education is to be realized over the next five years, a sector strategy wlll have to be adopted which combines measures aimed at improving the quality of education, reducing the number of repeaters, especially at standard eight, and increasing the total amnount  (vii) of funds available to primary education. To implement a program of this scale within the Government's means would require substantive revisions of sectoral strategy. Three suggestions have been put forward: (a) a new category of teachers should be recruited with lower Initial standard of education. For example, junior secondary school graduates could be given two or three months crash teacher training and hired at a lower initial salary with opportunity in future for upgrading when capacity is created at teacher training colleges; (b) measures-to-improve quality of primary education would be initiated to more than counteract any possible negative impact of reduction in the level of teachers' sualifications. These measures include provision ofadequate textbooks and instructional materials and develoPing a more focused teacher training program than at present. The expanded enrollment therefore would entail additional financial measures such as-(I) an Increase In the share of total recurrent expenditure of MOE devoted to primary education: and (II) educational tax on some luxury items designed to raise money for primary education. 13. There is no "magic" by which funds devoted to education can be reduced and there is at the same time a clear need to increase enrollment ratios significantly at all levels of education and therefore to increase the amount of funds available to education beyond the present 12 percent of the recurrent budget. The proposals made in this study, therefore, are designed to permit this continued expansion in a manner which ensures the best use of these limited but necessarily Increasing financial resources.  I. GENERAL BACKGROUND Recent Economic Developments 1.01 The last decade in Malawi has been full of economic uncertainty and disequilibria which took the form of continuous fluctuations in the level of economic activity, inflation rate, balance of payments and fiscal deficits. Economic growth, savings and investment have declined sharply in recent years. The Growth of Gross Development Product (GDP) fell from 5.5 percent in 1979 to only 0.4 percent in 1980 and 0.8 percent in 1981 as virtually every sector of the economy contracted. The growth rate in 1982 was 3.0 percent and similar g,rowth rate is expected to be achieved in 1983. While domestic savings declined somewhat, investment dropped sharply from 34.3 percent of GDP in 1979 to 21.2 percent in 1981 and 20.3 percent in 1982. Malawi's fiscal position also seriously deteriorated. 1.02 From independence until 1977/78, government expenditure expanded roughly in line with GDP and deficits rarely exceeded 7-8 percent of GDP. Since 1978/79, however, the Government's fiscal position has deteriorated. Revenue has increased slightly faster than GDP; while recurrent and development account expenditures have grown even mre rapidly owing to rapid expansion of development projects. As a result, rapidly rising expenditures outstripped revenues and pushed government deficits up to an average of 13.6 percent during 1979/1980 to 1981/82 before falling to 9.4 percent in 1982/83. In these four years, government expenditures averaged about 33 percent of GDP compared to 25 percent in the previous six years. To finance the deficits, the government resorted to increased domestic borrowing, partly on commercial terms. Inflationary pressures grew to almost 14 percent per annum between 1979 to 1981. On the other hand, the outlook for the growth of the Malawian economy, given global economic conditions and the political situation in Southern Africa, is not very promising. Recession in industrialized countries has hit Malawi's budget by suppressing growth in demand for its exports and increasing its debt service burden. Frrthermore, due to prevailing high interest rates, the climate for flow of capital to Malawi from industrialized countries is bleak. Education Financing and the National Economy 1.03 In nominal terms, the recurrent and development budgets of the Ministry of Education have grown from 11.4 million Kwacha in 1971/72 to 54.5 million Kwacha in 1982/83. However, as a percentage of total Government expenditure, the allocation to education fell over time, from  - 2 - 16.8 percent in 1971/72 to 8.6 percent in 1979/80 (Annex T-1). In 1982/83, the share of education increased to 14 percent of total Government expenditure. As a prx.portion of the GDP, the share of expenditure on education has risen from 2 percent to 3.5 percent during the period. At the same time, there have been tremendous increases in enrollment. For instance, with the help of external financial assistance, enrollment in primary schools doubled between 1971/72 and 1980/81, and access to secondary level education expanded both through i creased enrollment in the formal education system and through increased enrollment in the informal system, i.e. Malawi Correspondence College (MCC). There has been significant improvement in the quality of education due to improvements in the quality of teachers through upgrading primary teacher training as well as provision of textbooks at the primary level. However only 62 percent of the eligible age group were enrolled in primary richools in 1982/83 and only four percent of the 16-17 year olds were in secon.dary schools. In addition, the transition rate from primary to secondary school in recent years has dropped from 10 percent to 6 percent. The output from tertiary level education institutions is insufficient to meet the needs of the economy for miAdle and high level manpower. For instance, the 1979 manpower survey revealed that expatriates occupied 13 percent of all professional technical positions and 43 percent of administrative and managerial positions. As a result, there is considerable pressure for continued expansion of education services at all levels. 1.04 The education budget is particularly vulnerable during periods of economic difficulty. Projects that have short term returns and make immediate contributions to reduction in deficits and balance of payments problems may be more attractive although their long term development impact is less clear. The unfavorable economic climate for development has further consequences for public expenditure. The policies of adjustment proposed in such circumstances usually favor tight controls on public expenditure and the reduction !n its level. The effects of such macro economic policy are likely to make further funding of education at existing levels impossible. Furthermore, in addition to pressures on the education budget which result from difficulties in maintaining aggregate expenditure levels, there are other pressures in the economy acting to reduce the priority accorded to education, especially in Malawi where investment in education has never been accorded high priority. The emergence of other major sectors (energy, health, road transportation, rural development) in the 1980s has increased competition for public expenditure as the Malawian Government seeks to provide more services in different sectors. 1.05 Although education expenditure has been rising in recent years (Annex T-1 and Chart 1), this situation may not continue. Current figures do not show a substantial drop in proportional allocations, since  -3- educational budgets are usually "sticky" because a large part of recurrent expenditure is salaries, which cannot be reduced rapidly. Consequently, when changes in priorities take place, there is a time lag before the effect is felt on the budget. Secondly, even when the proportional spending on education is maintained, low or negative economic growth rates and inflation may be reducing the resources allo'ated to education in real terms. Thirdly, changes in allocation of resources in Malawi within the education sector have benefited the university (Annex T-2). This has had the effect of reducing, in both real and proportionate terms, resources available to other sub-sectors such as primary and se-,ndary education. If public expenditure in Malawi is to be reduced in redi terms, the pressure will undoubtedly fall heavily on the educational budget with adverse consequences for medium and long tenm development goals. The impact of any reduction in expenditure will be in those areas of budget that are politically expendable - these are basic education and non formal education. Non salary recurrent expenditure is also vulnerable although savirgs in this area may be very small. Parts of the system important to the intarest of the urban elite, such as the high cost University of Malawi are not likely to suffer, although they may represent the least effective use of available resources. 1.06 The proportion of public expenditure spent on education in Malawi is less than that in many other African countries (Annex T-3). Education needs are a long way from being satisfied. If changes are needed, these are most likely to be in the distribution of educational expenditures between different categories and in effective use of available resources rather than in a reduction in the overall level of expenditure. Acting against the pressures for a reduction in the ovc .311 level, of education budget and priorities accorded to education are forces that may produce opposite direction. Because the formal education system in Malawi contains incremental salary scales for teachers and administrators (salaries represent more than 95 percent and 50 percent of the recurrent expenditure at primary and secondary schools respectively) there is a built in tendency for costs to increase annually for a long time, with the g2owth in average age and length of service of teachers and administrators. Another important factor which must not be underestimated is the effect of increased demand for education. The perceived private rate of return to education in Malawi Is high (section IV). Incomes of school graduates are many times higher than those of illiterates or drop outs from the system. Because of their scarcity, acquisition of school certificate provides access to secure jobs in the public and private sectors. Where opportunities to earn income are severely limited and the impact of poverty is not softened tnrough social welfare schemes, educational success is critically important in determining the futures of individuals. Furthermcre education in Malawi has many vested interests associated with  - 4 - it. For example, the Ministry of Education (MOE) is the largest single employer and has the most articulate employees and this creates pressure to increase expenditure. Employees and school materials suppliers all have vested interests in maintaining the level of expenditure. The administrators of Grant-Aided schools have a strong interebt in maintaining or Increasing their subsidies and the University authorlties with direct links to the senior government official have great influence in determining levels of expenditure. Also the high proportion of externally funded education investment, has the effect of increasing future recurrent expenditure at the period of its maturity. Consequently, the MOE's budget should grow at least as fast as the combined effects of salary adjustments (and other recurrent costs) and increases resulting from the provision of more student places due to population growth in order to ensure at least the maintenance of existing enrollment ratlos. This may prove difficult In Malawi when the overall economic growth rate is very low. Objectives of the Study 1.07 In view of the foregoing economic difficulties of Malawi and the pressure on the education budget, this special study focuses on two key elements in the education sector in Malawi: (a) efficiency and cost effectiveness of the education system; and (b) prospects for cost recovery at all levels of the education system. A number of background documents on Malawi's education system are available, some of which have been financed under Bank projects or authored directly by Bank staff 1/. A substantial body of sector knowledge has been acquired through appraisal and supervision of five Bank Group assisted education Projects 2y. The Third and Fourth Education projects provided financial resources for research and for the preparation of the National Education Plan. The available Information on appraisal reports and the most recent Education Sector Memorandum provide good analyses of the education system and highlight the key problems. However, none of these studies specifically addresses the topics of efficiency and financial management or the issues of resource allocation between the different levels of education or cost recovery at the University level. This study focuses on these key issues in the hope that it will assist the development of appropriate sector policies wlth respect to them. 1/ For example, Evaluation of Human Capital in Malawi, by Stephen L. Heyneman (World Bank Staff Working Paper No. 420), and Charging User Fees for Social Services: The Case of Education in Malawi, by Mateen Thobani (World Bank Staff Working Paper No. 572), as well as an unpublished Ministry of Education Paper on education financing prepared by Rachel Jones. 2/ First Education Project (C102-MAI) Signed 5/4/67; Second Education Project (C.590-MAI) Signed 11/24/75; Third Education Project (C.910/MA1 EEC-12-MAI) Signed 6/19/79; Fourth Education Project (C.1123-MAI) Signed 5/29/81; Fifth Education Project (C.1330-MAI) Signed 5/5/83.  1.08 The Government is currently discussing a draft National Education Plan for the period 1984-1994. The draft plan makes a number of recommendations for greater cost-efficiency. This study is designed to supplement these recommendations with more detailed analysis and justification, as well as additional recommendations, which, it is hoped, will result in adoption of cost-efficiency measures as a national policy. 1.09 The rest of the study is organized as follows: Section II discusses the ways in which the administrative and financial management of education system in Malawi can promote efficiency and equity in the management of resources. Section III focuses on efficiency and cost effectiveness at the primary and secondary level. Given the recent increases in fees at the primary and secondary levels, section IV discusses the orospects for further cost recovery at these levels. In section V the issues of efficiency and cost effectiveness at the University are addressed. Section VI uses cost benefit analysis to address the overall issues of efficiency and equity in investment decisions on education. Finally Section VII discusses the recurrent cost implications of the proposals made in this paper based on enrollment targets set by the Government for the medium term 1984/85 - 1987/88. II. MANAGEMENT OF EDUCATT)N Introduction 2.01 Education in Malawi has followed a system of eight years of primary education, two years of junior secondary education, two years of upper cycle secondary education which divides into general secondary education, technical education and primary teacher training. On completion of the upper cycle of secondary education, it is possible to enter the University and pursue courses in the social sciences, agriculture, education, engineering, medicine including nursing and pure sciences (see Chart 2). 2.02 The development of the Education system throughout the 1970B has led to a sizeable growth in the MOE establishment. Total MOE established staff excluding the university numbered 1,125 in 1970/71. By 1982/83, the MOE has an establishment of about 15,000 persons and a total budget of about K 54.5 million. 2.03 The size of the MOE's establishment and the diversified duties required of it especially at a time when additional resources are not readily available will increasingly require careful management of the  -6- Administration as well as of the Budget of the MOE. In this section, the administration of education in general, the budgetary process and the decision making process within the MOE, regarding education finance, are discussed. Education Administration 2.04 As in mar'r African countries, education administration in Malawi is centralized in t!e Ministry of Education. Centralization in Malawi implies more than one layer of authority. The organizational chart of the MOE is presented in Chart 3. Excluding the autonomous units, there is within the MOE, one Minister, a Principal Secretary and five general directorates including the personnel office, accounts section, school administration, planning, inspectorate and examination. These major levels are sometimes subdivided with differing degrees of autonomy and authority. There are also regional and district offices with administrative oversight responsibility over primary ;ducation either at regional or district levels. Most countries have relatively highly centralized education systems with national ministries establishing education policies and procedures for all public schools as well as regulating standards for private schools. An important reason for this is that with a low average level of income and with relatively weak tax structures, individual communities cannot pay for education with self generated resources as in developed countries. Central funding, often based on export and import tax revenues, sales taxes, taxes on major crops, corporate profit and pay as you earn (PAYE) income tax systems is conducive to centralized administration. 2.05 In Malawi, the MOE is responsible for administration of public education institutions at all levels except the University of Malawi (and its component parts) which has its own Senate of which the Principal Secretary of the MOE is a member. It has the responsibility for the structure and content of education, including the examination's system, the curricula, the provision of textbooks and instructional materials, provision and administration of the teaching staff of all schools and colleges, the location of Government funded schools, regulation of the level of school fees as well as school inspectorate. The MOE is also responsible for education planning and most recently cultural activities, which includes the oversight of the Chichewa Language Board. The budget of the University is included in the MOE budget as a line item, but its administration and financial management are the responsibility of the University. The MOE also provides major financial support including payment of teachers' salaries to privately run, mainly church affiliated secondary schools and colleges through a system known as "Grant in aid", whereby block grants are given to the schools and actual administration of the funds is the responsibility of the school board and governors.  -7- Issues in Education Management 2.06 Top level administration can promote efficiency in two ways (a) by directives to lower level functionaires requiring action leading to goal accomplishment; or (b) by use of an incentive system, which rewards efficient performance by lower level functionaires. With regard to the latter, in the Malawian national context, an incentive system is not feasible. Apart from the fact that the educational system in general has unclear goals and outputs that are difficult to measure, the salary and wage structure of workers within MOE's control are decided outside MOE in aecordance with norms applicable throughout the civil service. With regard to the first point, the flow of informations downwards and backwards takes time and, often in the process, the directives get blurred and are misinterpreted and inaccurate responses are transmitted back to the MOE. Faulty and delayed information is costly In time and efficiency in addition to the cost of preparing and transmitting the directives. Even when workshops and in service training are held to explain new policies, such as changes in curricula, many teachers c'r these who are at the implementation level may continue to act as if the cnanges had not been introduced. In Malawi, for instance, directives on classroom size or teaching loads are not enforced. This is particularly true at the Primary level where class size of over 100 is not unknown especially at standard one. This leads to inefficiency. In MOE, directives are used to transmit decisions and there is relatively little delegation of authority. The point here is that for new policies and dirrctives to be enforced and implemented, those who will actually be responsible for the implementation of the directives should be fully involved at the decision stage so that they are committed to its goals and general objective. 2.07 Effective planning requires some measure of centralization so that duplication and waste of scarce resources can be avoided. Unfortunately in Malawi, the Planning Department is weak and understaffed and not able to execute its function properly-3/. Strengthening of the Planning Department should be given the highest priority. This would enable the department to carry out its policy functions as well as monitor the implementation of any long range plans for the sector. 2.08 Another area for significant economies is at the Malawi Certificate of Education and Testing Board (MCETB). This institution is primarily responsible for administering the examinations given at the end of the senior secondary school, while MOE is responsible for primary and 3/ It has the following established positions: a chief and two senior education planning officers and a statistician. In addition there is an expatriate education planner financed under a Bank Group assisted education project and a UNESCO financed studies coordinator on a two year assignment. One of the senior education planning officers is on study leave and the other post is currently vacant. There are two junior officials assigned to the two expatriates as counterparts.  junior secondary exaaninations. MCETB is expected to charge fees for the use of its computer for data processing. The charge to MOE's recurrent budget for the MCETII was K 625 000 (US$540,000) in FY1982/83. Since about 3500 students sat fcr the certificate examination that year, the cost per examinee to MOE is E 190 (US$152). This appears to be excessive in relation to the benefits derived by MOE, since MOE still has to pay in commercial terms for any additional data processing requirement from MCETB. Unit costs can be raduced if all school examination systems are centralized. The potential exist for MCETB to be a net revenue earner to MOE. At present it has considerable underutilized computer capacity. It is recommended that the Government should seek the services of cost accountant to examine ways of making the MCETB more efficient than at present. 2.09 Recommendation: It is recommended that (a) the Planning Department should be strengthened by additional qualified staff with specific duties and terms of reference, (b) MOE should ensure that principals at the school level understand the objectives and goals of any directives issued from headquarters since they are required to implement these directives; and (c) action should be taken to reduce the unit cost at MCETB preferably by employing the services of a short term cost accountant. and by centralizing all examination bodies within one unit. MOE should give serious consideration to merging the examination unit within the Ministry and the MCETB as a means of reducing unit cost of examinations. Consideration should be given to employing a short term consultant to study the most efficient method of merging the two examining bodies. Financial Management 2.10 Budget process. The annual draft budget for education is determined jointly by the MOE and the individual schools and colleges. The initiating memorandum from the Ministry of Finance (MOF) usually sets the growth ceiling for the forthcoming budgetary year. During the budget preparation period, requests are sent out to all schools and colleges for submission of estimates of their future recurrent expenditure to their respective administrative section heads within the MOE. Each submission is carefully scrutinized by each administrative unit to ensure that it meets the guidelines provided by the MOF. Requests for staff increases are given special attention since in the event of revenue shortfalls, salaries and wages are usually not affected and only non-salary components of the budget can be reduced. The MOE usually conducts a series of internal working level meetings of section heads to arrive at a consensus on what the overall budget submission of the MOE should be and on which sections should benefit from any additional revenue. The resulting budget is submitted to and discussed with the MOF before it is submitted to Parliament for the funds to be appropriated.  - 9 - 2.11 In theory, the procedure for decisions on the level of the development budget is more centralized than that for the recurrent budget, although local authorities are invariably consulted regarding the location of new schools and additional classrooms. However, the appropriation process for the development budget differs significantly from that of the recurrent budget since the level of the education's development budget is determined by (a) the level of external financing specifically earmarked for education; (b) the MOE's implementation capacity, and (c) the ability of the MOE to raise the counterpart funds for those projects which require matching local funding as well as the Government's perception of the level of its future indebtedness at the period of maturity of the credit or loan, as the case may be. 2.12 The main focus of the Government's development strategy has been agriculture, industry and the infrastructural investment needed to maximize such investments. Because of the long gestation period of investment in education and its invisible rate of return in nominal terms, education has, in the past, received very low priority in the Government's investment strategy. Consequently, until fairly recently, most of the development budget of education was externally financed or by contributions in kind by parents 4/. The level of the MOE's recurrent budget at any point in time is determined by past investment decisions. Once the investment decisions are made, the MOE is obligated to provide the recurrent resources for running the facilities on completion, at least as far as funding the salaries and wages of key personnel. 2.13 Once the budget is approved, the supervision and admlnlstration of the recurrent and capital budgets becomes the responsibility of the Principal Secretary of the M4OE. Actual day to day responsibility for supervision of the budget is the responsibility of the designated warrant holders who are given the responsibility for approving expenditures up to a certain level for sections under their control. For example, with regard to Government secondary school expenditure, the principal education officer in charge of secondary schools is the warrant holder for the day schools and he has designated the principals of day secondary schools as the sub-warrant holders. The actual payment of funds so approved is still the responsibility of the Principal Secretary or 'his' designate. For the primary schools, the regional education officers and their district counterparts are the warrant holders. 4/ The first external financing for education was the Bank Group financed First Education Proiect (Credit No. 102-MAI) in 1967 for US$ 6.3 million dollars equivalent and it was not until 1975 that any new major investment decision was made.  - 10 - Issues in Education Budgeting 2.14 The process by which budget decisions are handled affects the efficiency and equity of any educational system. A budget decision made at the Central Government level can be viewed as a source of finance from the perspective of the next lower level of educational organization. The educational services provided also affect the financial resources that can be raised. Each and every educational budget decision at every level invitlves the potential benefit or effect in relation to costs. This is particularly important in Malawi where the Government may be the largest source of financing but the delivery of services is shared between the Government and other entities. In Malawi, education services at the primary and secondary level are provided by the Government and the missions and other private agencies, although the Government has legal control over inspection, licensing, laws governing curricula, fees and salaries. Schools and colleges are either Government, or Government aided or private depending on the degree of Government input on the financing of the school. Government schools receive all their funds from Government sources and fees are paid into MOE account as appropriation in aid. Assisted schools receive part of their recurrent expenditure in the form of flat grants from the Government and the rest of their funds come from parents or the external agencies with which they are affiliated. 2.15 Financing of primary education. The Central Government in Malawi plays a very minor role in the funding of Capital expenditure at the primary level. Prior to the Third Bank Group assisted education project, the construction of school buildings and their equipping were entirely the responsibility of local authorities. This meant that the quality of the buildings depended entirely on the amount of resources at the disposal of each local authority botlh in terms of cash and in the form of free labor. The result is considerable divergence in quality of primary school buildings iT. Malawi. With the Third Education Project, although classroom construction remains essentially the responsibility of local authorities, the Central Government has extended its aid to local authorities by provision of building materials like roofing sheets and cement as well as technical assistance for supervision of construction. 2.16 However, the amount of Central Government funding of construction of classrooms is still to a large extent dependent on the ability of local authorities to demonstrate that they can meet some of the costs of construction such as of bricks and labor. On equity grounds, since the Government's objective is to provide basic education at the primary level, the current system of funding of classroom construction is not helping to achieve that objective. Local authorities endowed both with material resources as well as community spirit are decidedly at an advantage and are  - 11 - able to construct more classrooms and to have higher enrollment- ratio at this level than other local authorities. In a period of serious financial constraints, the Government should not assume a major responsibility for primary school construction. Nonetheless, the Government could direct its limited funds towards the most needy communities where it could also provide technical assistance in construction and management techniques. 2.17 Recurrent costs of primary schools are jointly funded by local authorities, the Central Goverament and the parents. School fees are used for the purchase of books and other school materials and are paid into a special treasury account for that purpose. Teachers became part of the civil service in 1977. Before that, teachers were employed by local education authorities and paid for by the Central Government through block grants which covered the salaries of qualified teachers only. The budget allocation for Districts and therefore schools from the Central Government is calculated on the basis of the number of teachers and their length of service, since teachers' salaries are dependent on their qualification and their position in the salary scale. The result is that Government subsidies per child are not evenly distributed among schools. Available information (Annex T-4 shows that Districts with the lowest unit recurrent costs have the highest pupil:teacher ratios and the lowest proportion of tl, t2 and t3 teachers 5/. At present urban schools with better housing facilities and better amenities have a higher proportion of qualified teachers and a lower pupil:teacher ratio than rural schools. For example, as shown in Table 2.1 all urban towns have lower pupil:teacher ratios, higher proportion of tj and t2 teachers than rural schools. This indicates that the cost per pupil to the Government is higher than in rural schools. Improvement in pupil:teacher ratios and quality of teachers at primary levels would not reduce the unit costs at this level. 5/ Primary school teachers are classified tl, t2, t3 depending on their initial level of education prior to teacher training. Thus a JCE graduate with two years teacher education is classified T3 and an MCE graduate with two years teacher education is classified T2 . T1 is a promotion grade.  - 12 - Table 2.1 Unit Recurrent Cost, Repeater Rate, Qualified Teachers and Pupil Teacher Ratio selected Areas I~~98 Unit Recurrent Repe.taters Qualified Pupil Cost Rate TeacherE Teacher Salaries only as % of Ratio Mk std std total Classes ______ ______ 1 2-7 1___ _ _ 7_ _ Mzuzu 21.5 18.1 9.3 85.9 44.9 Chitipa 16.0 22.4 16.6 77.9 69.8 Lilongwe rural 14.7 18.1 14.4 84.5 65.4 Lilongwe urban Z,..5 17.3 13.0 85.9 38.2 Blantyre city 19.3 23.7 14.5 85.3 49.5 Blantyre rural 12.6 20.7 14.5 80.5 78.6 Zomba urban 25.2 14.7 12.9 82.5 40 2 Zomba rural 12.9 17.9 12.2 77.1 77:6 Source: Bank Staff Calculations from Data provided by MOE The point to be made is that this current syst.7m has efficiency implications. First, schools with high student:t'cher ratios tend on the whole to be less able to prepare their students adequately as measured by the repetition rate, since teachers are not able to give each child the attention required to bring out the best in them. Poor students needing greater attention from teachers often are neglected in such a situation. Secondly, costs per student graduate are higher in such schools since the length of time it takes a student to pass through the system will be longer than normal especially in a school system such as that of Malawi which relies heavily on examination results for promotion from one level to the next. As shown in the table above, except for the Blantyre area, there appears to be a close correlation between pupil:teacher ratios and repetition rates in classes 1 through 7. 6/ 2.18 Recommendations. Although equity consideration might seem to require the Government to spend the same amount per child in all districts, some divergence in costs per district will occur due to differences in length of service of teachers. It is therefore recommended that the Government should adhere to strict administrative procedure to ensure that qualified teachers go to rural areas. However, as an interim measure postlng of qualified teachers to urban areas should be frozen until all 6/ Class 8 was excluded because of the distortions in the data created by students repeating in order to obtain a better score in the Primary School Leaving Certificate (PLSC) so as to get access into seccndary schools.  - 13 - schools have at least 80 percent of all their teachers certificated. The average unit cost MK 16.1 at primary level is extremely low by African standards. Enforcement of a pupil:teacher ratio of 50:1 in all rural districts, gradually bringing this ratio to the level of urban Lilongwe (38:1) as well as ensuring that the qualified teachers are equitably distributed according to a set formula would lead to a more equitable system. 2.19 Primary school fees and their management. Unlike most East African countries, Malawi charges fees for primary education. Fees collected are used for purchase of textbooks, writing and teaching materials. Until FY 1977/78, collection of fees and the usage were the responsibility of local authorities. The result was that basic textbooks in language and mathematics were provided to only one third to one half of all the children at the primary level and very little additional supplies were available 7/. Since 1977, MOE has assumed the responsibility for fee collection, fees are paid into a special account and used for the purchase of exercise books and other school supplies. With Bank Group assistance (Third Education Project), the Government is now providing English, Mathematics, and Chichewa textbooks withi more durable covers and binding. Extra fees are used for school repairs and for payment of water and electricity bills. About 80 percent of the tuition is applied towards textbook purchase and 20 percent is used for the purchase of consumables. However as a result of the Bank Group Textbook Program, the decision was made to set aside 1 kwacha per child per year for the replacement of textbooks. Fees are collected by the Government to cover all costs other than teachers' salaries and general administrative expenses of primary education. Fees were increased in October 1982 by 50 percent for rural children and 43 percent for urban children in Standards 1-5 and by 15 percent for Standards 6-8. The current structure of fees is provided in Table 2.2. Cost of fee collection is high and is absorbed in the administration budget of the Ministry of Education. In 1982/83, the MOE expended about K 0.4 million to collect K 3.24 million in fees - 12.4 percent of the amount collected. Table 2.2 Current Structure of Primary Fees: Annual Fees Tuition Extra Fees Urban Rural Standard I-V K.3.0 K.1.50 K.O.50 Standard VI-VIII K.5.0 K.2.50 K.O.50 Source: MOE 7/ S. Heyneman: Evaluation of Human Capital in Malawi. Staff Working Paper P,r. 420, October 1983  - 14 - 2.20 The textDook program supported by Bank Group assistance presents a unique opportunity for investigating whether or not the portion of fees devoted to textbooks is sufficient to cover this portion of education expenditure. The results of an exercise based on the above program are summarized in Annex T-5. These ca:Lculations were made on the assumption that the average life of the textbook is three years. As could be seen in Annex T-5, 1 kwacha per child will not cover the annual cost of books in 1983 prices. On the other hand, if the initial arrangement of setting aside 80 percent of the fees for textbook replacement was followed the table demonstrates that the collection of fees for textbooks would more than cover their average cost for the entire primary school sector, a surplus of MK 0.79 per child would infact be collected. If fees keep up with inflation, then a surplus in the textbook account should be expected every year provided the cost of the textbooks also goes up at the same rate as inflation. This surplus could be used to finance othe.. quality improvements in the primary schools. 2.21 Recommendations and Conclusion. Primary school fees presently cover almost all non-teacher recurrent costs and it is recommended that this should continue to be the case. However, these fees should be raised annually by the expected rate of inflation, so that they can continue to cover the cost of instructional and school materials. Eighty percent of the fees shotld be set aside for textbook replacement. Teachers' salaries should be borne by the Central Government since significant benefits from primary education '-crue to society and therefore, society should be expected to bear a major responsibility for the cost of education at this level. 2.22 Financing of secondary education. For budgetary purposes, there are three types of secondary schools in Malawi: Government boarding, Government day and grant aided. These categories are differentiated by the source of financing0 Government schools are exclusively constructed with public funds, whereas grant aided schools derive their funds for capital provided by private, mainly religious sources. However, in recent years, this differentiation has become blurred, since renovation and expansion of grant aided schools have been financed through Government external borrowing. 2.23 Irrespective of category, salaries of teachers of all secondary schools are paid by the Government since all secondary school teachers are civil servants. The exception to this rule are missionaries who teach in grant aided schools. Such teachers are paid stipends, the amount of which varies with the marital status of the individual concerned. The stipends are lower than the salary of civil servant teachers with the same or similar qualifications and experience. Furthermore, grant aided schools  - 15 - receive a sum to cover travel and leave passage of missionaries, house maintenance, Board of Governors meetings, clerical staffs and other expenditure. In 1982/83, these grants amounted to K 363,955 or about K 49 per student. In addition, they are reimbursed K 9 per student of the tui-tion fees collected and given boarding grants of K 40 per student, aLAhough they are allowed to keep the boarding and tuition fees collected. 2.24 Government day schools, many of which have boarding facilities, are treated as day schools and boarding fees collected from students are not paid into the treasury but used strictly for feeding the students. Boarders in Government boarding schools pay their fees into the MOE account as appropriation in aid and all the running cosLs of such schools are met by the Government. However, principals of boarding schools are provided a sum equivalent to the boarding fees collected at each school and such fu, 1s are managed by each school's boarding committee and are expected to meet the cost of food. 2.25 The financing of Government schools differs from that of grant aided schools. The former present estimates of recurrent expenditure which are used as basis for the allocation of funds to each school. Mid year allocations are usually made in the light of requirements which have arisen since the original estimates at the beginning of the year. This makes for considerable flexibility, in that, it is possible for the MOE to allocate additional funds or withdraw surplus funds according to circumstances. On the other hand, the grant aided schools, apart from Malawian teachers and cover salaries of missionary teachers and grants for other approved expenses on a prescribed per pupil basis without any possibility of adjustment; any additional funds required are provided by the school or from fees. Although Grant Aided schools are required to submit annual audited accounts to the Government, this procedure is not always followed. Table 2.3 presents an analysis of unit cost to the Government of Government schools and Grant Aided schools.  - 16 - Table 2.3 1982/83 Expenditure per Puil - Government and Grant Aided Ser4ndiiry Schools (in Kwvacha) t _ , _ _ _1 _ _ _ _ t t Gove-auxent t Grant Aidedj t_ ~ J Boarding J t | Day B Boarding iSalaries and Wages : 172.4 t 385.2 $ 155.1 t MI.ssionaries - - 1 30.4 1 - - I Sub-Total | 172.4 1 385.2 1 185.4 1 Boarding - 83.4 40.0 1 1 lTuitlon 1 25.3 46.5 j 9 Other 36.3 150.3 1 49.2 1 Sub-Total I 1.6 280Z2 96.2 | Total t 234.0 665.4 1 283.6