Philippines - Education financing and social equity : a reform agenda

Report No. 15898-PH Philippines Education Financing and Social Equity: A Reform Agenda June11, 1996 Human Resources Operations Division Country Department I East Asia ancl Pacific Region Document of the World Bank  CURRENCY EQUIVALENTS (As of April 1, 1996) Currency Unit = Peso US$1.00 = Pesos 25 US$0.04 = Peso 1.00 WEIGHTS AND MEASURES 1 Meter (m) = 3.28 Feet (ft) I Kilometer (km) = 0.62 miles FISCAL YEAR January I - December 31 ABBREVIATIONS AND ACRONYMS COA Central Office for Auditing DECS Department of Education, Culture and Sports ESC Educational Service Contracting GASTPE Government Assistance to Students and Teachers in Private Education IRA Internal Revenue Allotment LGC Local Government Code LGU Local Government Unit MOOE Maintenance and Other Operating Expenditures NCR National Capital Region NEDA National Economic Development Agency NGO Non Governmental Organization NMYC National Manpower and Youth Council (renamed TESDA) PTA Parent-Teacher Association PY Project Year SUC State University and College TEI Teacher Education Institution TFS Tuition Fee Supplements  Table of Contents I. Summary and Recommendations ...........................................1 1.1 Summary .........................................1 1.2 Recommendations .........................................5 II. Education Finance ...........................................9 2.1 Central Government Financing .........................................9 2.2 Local Government Financing ........................................ 21 2.3 Private Financing ........................................ 24 2.4 Synthesis ........................................ 26 III. Major Issues...................................................................................................................29 3.1 Education and Poverty ........................................ 29 3.2 Fiscal Decentralization and Equity ........................................ 33 3.3 Social Targeting ........................................ 36 3.4 Public Versus Private Education ........................................ 45 3.5 Cost-Effectiveness In Public Education ........................................ 53 Annex 1: Poverty and Regional Equity in Basic Education .......................................... 60 Annex 2: Selective Evidence on Local Education Financing .......................................... 70 This report is based on sector work conducted in 1994-95, as part of the identification of a Third Elementary Education Project. The report has been prepared by Antoine Schwartz (PSP, Senior Economist), with the assistance of Arben Bakllamaja (consultant), and Francoise Delannoy (EA1IR, Task Manager). William McCleary (EAIDR) guided the completion of the report; Birger Fredriksen (AF5CO) and Emmanuel Jimenez (PRDPH) were the peer reviewers. Special thanks go to Vajeera Dorabawila (consultant) for statistical support, and to Gay Santos (PSP) for word processing. i  List of Tables Table 2.1 Central Government Education Expenditure Recent Trends ......................................9 Table 2.2 Evolution Of Government Education Financing 1985-95, By Category Of Exp enditure ................................................11 Table 2.3 Teacher Remuneration Intra-Sectoral Distribution Of Expenditure .......................... 12 Table 2.4 Government Financing, By Level And Type Of Education Administrative Versus Operational Expenditures ......................................................... 14 Table 2.5 Functional Distribution Of The 1994 DECS Budget ................................................. 15 Table 2.6 Actual Versus Budgeted DECS Expenditure Recurrent Expenditure Per Student ... 17 Table 2.7 Recurrent Expenditure Per Student ......................................................... 19 Table 2.8 Evolution Of Per Student MOOE In Basic And Tertiary Education ........................ 20 Table 2.9 LGU Education Financing By Expenditure Category (% Distribution) ................... 22 Table 2.10 Total Expenditure By Level Of Education And Source Of Financing (1994) .......... 26 Table 2.11 Changing Pattern Of Education Financing 1986-94 .................................................. 27 Table 3.1 Family Income, School Participation And Completion ............................................. 29 Table 3.2 Reasons For Not Attending School (Lowest 30% P.C. Income Group) ................... 31 Table 3.3: Comparative Nutrition & Health Indicators .......................................................... 32 Table 3.4: School Performance: Eastern Visayas Region Vs. NCR ........................................... 33 Table 3.5: DECS Resource Allocation: E. Visayas Region Vs. NCR ......................................... 37 Table 3.6 School Staffing Characteristics: E. Visayas Vs. NCR ............................................... 42 Table 3.7 Unit Recurrent Cost Of Public And Private Schools By Level Of Education (1994) .46 Table 3.8 Unit Operating Costs As Percentage Of GNP Per Capita: Philippines Versus Comparators In The Region ..................... 47 Table 3.9 Private/Public Unit Cost Ratios ..................... 48 Table 3.10 Comparative School Performance Indicators In Basic Education (Private/ Public Performance Ratios) .49 Table 3.11 Cost Sharing In Public Education, 1986-94 (Government And Private Shares In % Of Total Unit Cost) .51 Table 3.12 Teaching Versus Non-Teaching Decs Staff In 1992, By Level Of Education (Percent Of Distribution) .................................................. 56 Box 3.1: Relative Incidence Of Private Educational Costs .................................................. 52 List of Figures Figure 2.1 Evolution of Per Student Recurrent Expenditure by Level of Education ............... 18 Figure 3.1 Poverty and Elementary School Completion (1991) ................................................ 30 Figure 3.2 Regional Income and Per Student SEF Expenditure (1994) .................................... 35 Figure 3.3 Government Basic Education Expenditure Per Student (1994) and Poverty Incidence by Region .38 ii  I. Summary and Recommendations 1.1 Summary 1. This report is primarily, but not exclusively, concerned with the financing of public elementary and secondary education. Both levels, referred to as basic education, are. the responsibility of the Department of Education, Culture and Sports (DECS). In addition to the DECS budget, central government resource allocations for the education sector as a whole include budget appropriations for State Universities and Colleges (SUCs), and for the National Manpower and Youth Council (NMYC) responsible for non-formal manpower training and skills upgrading programs. Basic education absorbs over 90 percent of DECS' operational budget. 2. Constitutionally, education in the Philippines commands the highest priority in the inter- sectoral allocation of Central Government resources. In reality, several factors have contributed to limit the volume of effectively available budget resources: an extended period of slow economic growth; insufficient public resource mobilization; the heavy burden of the national debt; and the increasing transfer of central government resources to Local Government Units (LGUs) since the enactment of the new Local Government Code of 1991. Net of debt service and Internal Revenue Allotment (IRA) to LGUs, budget allocations for education absorb close to 30 percent of total available resources. This is a high proportion by international standards, even though it represents less than 3 percent of GNP. Overall, the prospective growth of government funding for the sector thus depends on future overall improvements in economic and fiscal management, rather than on increased allocations from the national budget. 3. At the same time, expansion of the public education system and the need for essential quality improvements, many of which mandated by Congress, generate substantial incremental financing requirements. The growing tension between available resources and financing needs unveils itself in two ways: (i) a gradual erosion of recurrent non-personnel expenditure per student, particularly in basic education (Table 2.8); and (ii) increasing cost-sharing for public education services by households and LGUs. The first trend primarily affects the quality of education delivery in public schools. The second has potentially adverse equity implications with regard to the distribution of educational outcomes among regions and population groups. 4. Financing constraints in public elementary and secondary schools are reflected in steadily rising student/teacher ratios and inadequate funding for essential non-salary operating expenses. Relative to student enrollment, budget allocations for Maintenance and Other Operating Expenditure (MOOE) are incompatible with delivery of quality education. Moreover, a substantial portion of DECS' budget appropriations for MOOE is channeled to the administration rather than to operations. At the school level, LGUs and parental contributions thus appear to be the principal sources of MOOE financing.  5. So far, education spending from private and LGU sources has amply compensated the shortfall in central government financing. Since the mid-1980s, more than half of the increase in national spending on education originated from private sources (Table 2.11). Education expenditure from all sources combined thus represented 6 percent of GNP in 1994, altogether a substantial commitment of national resources. At issue, therefore, are the distribution, composition and quality of this national investment, rather than its magnitude. 6. Private financing flows both to private and public schools. Private schools, accounting for a substantial share of post-elementary student enrollment (33 and 79 percent at the secondary and tertiary level respectively), are a vital element of private sector involvement in education. Over the past decade, however, the role of the private sector as co-financier of the public education system has expanded more rapidly than private education itself. It is this phenomenon that deserves particular attention in view of its underlying efficiency and equity implications (Section 3.4). 7. Parents in the Philippines, eager to capitalize on relatively high private returns, show great willingness to invest in their childrens' education. In such an enabling environment, Government's role is to promote social equity in access to quality education, particularly at the elementary level, without discouraging private sector involvement. The evidence, however, shows that Government has failed to sustain adequate funding for public elementary education, while simultaneously increasing subsidization of higher education institutions. As a result, the bulk of the increase in cost-recovery for public education has been concentrated at the elementary level, where households now shoulder close to one-third of the total cost, compared to only about 10 percent in the mid-1980s (Table 3.11). Equity 8. There is a fundamental equity problem in basic education, particularly at the elementary level, with potentially serious long-term consequences for the poor and the country's development prospects. At the elementary level, where the school infrastructure is extensively developed and opportunity costs are still relatively low, poverty affects school completion rather than initial access. Overall, one out of three children entering first grade does not complete the elementary cycle. Underlying this national average are such extremes as near-universal completion in the most affluent areas, and over 70 percent non-completion in the poorest provinces in Mindanao and the Visayas. Surveys among out-of-school children confirm the strong incidence of economic factors on school participation, particularly among post-elementary age groups (13-24). Malnutrition and poor health care are also important factors preventing active school participation among the poor. 9. There is evidence of significant regional differentials in the quality of educational services provided in public schools. The poor receive a lower quality education, not only because they attend public rather than private schools, but also because the quality of public education is comparatively lower in poor areas. Quality differentials are strongly related to the regional distribution of educational inputs, particularly as regards teaching and school management. For instance, the poorest regions have the highest proportion of inexperienced teachers, as well as the highest proportion of schools without principal, and school districts without supervisor. 2  10. Fiscal decentralization and devolution have generated winners and losers among LGUs, essentially because the incremental cost of devolved services bears little relation with the distribution of extra income. Although public education has not been devolved to LGUs, financing constraints at the central level are de-facto leading to selective decentralization of education financing. Winners among LGUs have resources available for extra support to public schools. They do so voluntarily, each according to means. As a result, there is a danger that the education quality gap between rich and poor areas will increase. 11. Present criteria and procedures underlying DECS' regional budget appropriations do not aim at counterbalancing the regressive impact of fiscal decentralization on local education financing, not to speak of redressing initial inequities. An overriding issue, in this context, is the lack of a coherent DECS strategy for actively targeting resources towards the most pressing needs, the most disadvantaged areas or population groups. The issue is essentially political. There appears to be great reluctance on the part of the national and local legislature to face the implications of the country's considerable geographical, cultural and social diversity, and accept discriminatory use of national resources for the sake of addressing priority regional needs. As a result, DECS is largely deprived of the authority and control required for effective social targeting. 12. Ineffective targeting is apparent in all budget categories. The absence of a truly coherent national strategy for the expansion of the public school system, and DECS' lack of effective control of locally financed school construction, often lead to unplanned additional claims on the recurrent budget. This, in turn, further reduce DECS' flexibility for effective expenditure targeting. Several allocation rules, particularly as regards sector staffing, are based on nation-wide student enrollment characteristics. These rules are applied uniformly throughout the country, regardless of specific local circumstances, actual needs and priorities, and tend to favor urban over rural areas. 13. Social targeting has inspired the creation of the GASTPE program (Government Assistance to Students and Teachers in Private Education), funded through the DECS budget, to enable youth from lower income households to attend private high-schools and colleges. The bulk of the program's support, channeled through tuition fee supplements (TFS) for students, and educational service contracting (ESC) with private schools, is aimed at the secondary level. Although innovative in design, and impressive in size, the program has a major flaw. For a given budget allocation, the GASTPE program has essentially sought to maximize the number of beneficiaries -- these currently represent about half of total private secondary enrollment. As a result, the benefits are substantially below the actual cost of private schools. This has two consequences: (i) TFS students are required to pay the difference between subsidy and actual cost, which eliminates a large segment of the target population; and (ii) the better private schools, unable to collect their full operating cost from ESC beneficiaries, are withdrawing from the scheme. 14. The "nationalization" of village highschools in 1988 was another attempt at social targeting. Village highschools, previously funded by LGUs, practiced a significant degree of cost- recovery through tuition and other fees. Nationalization, which entailed elimination of all fees, was the expression of the guarantee of free universal access to basic education adopted in the 1987 3  Constitution. The measure generated a sharp increase in financing requirements. Since then, Government has been unable to adequately sustain this additional commitment. Cost-Effectiveness 15. In view of the relatively low levels of unit recurrent expenditure, there does not seem to be ample room for immediate cost savings in the operation of the public education system. In the longer term, the most promising venue for increasing the cost-effectiveness of the system is to improve its internal efficiency, with particular focus on the country's lowest performing school divisions and districts. However, this requires substantial initial investment in quality improvements. 16. There is scope, however, for reducing the cost to the national budget of the rather inefficient State Universities and Colleges, essentially by promoting much higher cost recovery for educational services, and giving them greater financial autonomy. Management and financial decentralization would also make these institutions more responsive to the specific needs of the various regions, hence foster the external efficiency of their program offerings. 17. Thirdly, special efforts need to deployed to improve the effectiveness of budgetary processes in DECS. At present, the expenditure monitoring system is weak. The recording of actual expenditure commitments does not match the detailed format of initial budget appropriations. It is therefore difficult to monitor the actual implementation of specific budget programs and objectives. A particular concern, in this respect is whether and to what extent MOOE appropriations, a large proportion of which is not directly allocated to operations, are eventually used for the intended purpose. Government Strategy 18. Government's education strategy, the result of an extensive national consultation effort, is essentially directed at improving the quality of basic education, with major emphasis on the elementary cycle where achieving universal completion is a must. A major, and potentially costly issue to be resolved, in this context, is the need to increase the length of the secondary cycle from four to five years. A beginning has been made with the rationalization of budget support for the State Universities and Colleges, but without appreciable savings. Subsidization remains input- rather than performance-based. 19. Government is conscious of the equity issues facing the public education system. So far, these are essentially being addressed through the Presidential Task Force for the Fight against Poverty, with the participation of DECS. The task force has designed a development strategy for 20 Priority Provinces, with an integrated basic education component (See Annex 1, Figure 2). This is certainly a commendable first step. It is to be hoped, however, that this strategy will not lead to neglect of poverty-related education priorities in non-selected provinces, or foster uniform 4  approaches to widely different regional needs and circumstances.1 Moreover, the sustainability of the adopted strategy will only be assured if criteria and procedures for recurrent budget appropriations are adjusted accordingly. 1.2 Conclusions and Recommendations 20. Inadequate government funding has led to a substantial increase in cost-recovery in public schools, particularly at the elementary level. Public schools today are relatively more expensive for lower income households than private schools are for upper income households. In view of the rather inequitable distribution of income in the Philipines, / the emerging imbalance between public and private education financing affects a relatively large segment of the population. Continued reliance on private sources to meet the shortfall in central government financing therefore increasingly requires compensatory measures to protect the poor. 21. Overall, student enrollment at the elementary level is lagging behind the rate of population growth. At the same time, enrollment in public elementary schools is growing at a much lower rate than in private schools, suggesting that the private cost burden may have reached a critical level for lower income households. There are three venues to reduce this burden: (i) improving the efficiency of education delivery (see Section 3.5), so as to reduce the need for extra financial contributions from parents; (ii) increasing cost-sharing with LGUs where further possible, with selective targeting of central budget support towards disadvantaged areas; and (iii) direct subsidization of private costs for children from the poorest households. 22. Providing free access to secondary schooling of adequate quality would require a substantial increase in government funding. Cost-recovery at that level need therefore not be reduced. However, greater and, above all, more effective support for students from lower income households is a must. In this respect, public school fees should be regularized, notwithstanding the constitutional guarantee of free access. At the tertiary level, both the extent of cost-recovery and the cost-effectiveness of programs and operations ought to be targeted for drastic improvement. 23. Government's sector objectives call for considerable additional resources to be channeled to public elementary and secondary schools, particularly in view of the need for lightening the cost burden of school participation to the poor. Since education already absorbs a substantial part of the national budget, the required additional resources will have to be mobilized primarily through: * Reallocating resources within the overall education sector budget towards basic education; These issues are being are being addressed in the proposed Third Elementary Education Project, identified jointly by DECS and the World Bank, which would promote bottom-up participatory planning processes geared to meet specific needs in targeted areas. The latter would not be limited to the 19 priority provinces. 2 According to the 1991 Family Income and Expenditure Survey, 39 percent of households have a per capita income below the poverty threshold (47 percent of rural households). See The Philippines: A Strategy to Fight Povert. World Bank Report No. 14933-PH, November 1995. 5  * Reassessing expenditure priorities for central government funding in the sector; * Raising the cost-effectiveness of DECS' operations and programs. In the implementation of the national education agenda, moreover, DECS should aim at maximizing the social returns of its activities and programs by seeking more effective ways of targeting resources towards priority needs. Increasing available resources 24. Increasing resource availability for DECS can be achieved in two ways; i.e., by: (i) increasing DECS' share in the total education sector budget; and (ii) rationalizing expenditure within DECS' budget. (a) A first step for DECS to gain more financial room for addressing priority needs in the sector would be to devolve some of its present responsibilities (e.g. school construction and maintenance) to LGUs. DECS would provide selective compensatory financing to disadvantaged LGUs. This support would have to be granted in ways that do not discourage local taxation efforts, nor the extent of local resource commitment to education -- an Education Resource Equalization Fund, providing matching grants, might be considered in this context (b) The burden of State Universities and Colleges on the central government budget ought to be substantially reduced, and the savings reallocated to basic education (DECS). This could be achieved by promoting fuller cost recovery, decentralized financing, and greater financial autonomy for these institutions. Continued subsidization from the national budget should be selective, reflecting national priorities, rather than across the board, and based on output performance rather than input criteria. Equity considerations would of course require higher cost recovery to be accompanied by compensatory adjustments in existing scholarship and student loan programs. Since many SUCs are the product of local initiative, serving regional needs, such adjustments, to a large extent, could be co- financed by LGUs. (c) DEC S' elementary and secondary school construction program, to the extent that it is not devolved to LGUs, needs to be reassessed, considering the fact that not every barangay needs or can accommodate an elementary school, nor every municipality a highschool. At the elementary level, expansion of the school network is clearly becoming a lower priority than proper maintenance of the existing stock, a relatively large part of which is in dilapidated condition. Moreover, construction activities likely to be financed from other sources need to be taken into account and coordinated, if only to minimize the likelihood of unexpected subsequent claims on DECS' recurrent budget. Evidence further indicates that substantial savings, as well as quality gains, could be realized, by 6  entrusting the implementation and supervision of construction activities to local authorities and PTAs. (d) Rationalizing budget management processes in DECS is a necessary condition for achieving greater cost-effectiveness in the various programs and operations. In particular, better coordination between budget preparation and execution, more transparency in budget commitment and disbursement procedures, and more functional and systematic monitoring of actual expenditure are needed to insure that strategic sector objectives are being met. In a continuous perspective, improving budget management processes would also contribute to reduce the incidence of stop-go financing patterns that have adversely affected various educational programs and operations over the years. Effective targeting of available resources 25. With more resources at its disposal, DECS should strive to: (i) consolidate and sustain present quantitative achievements in basic education; and (ii) move towards nation-wide implementation of its quality improvement strategy, particularly at the elementary level; while (iii) targeting a higher proportion of expenditure towards the most disadvantaged areas and population groups. Social targeting will be most difficult, because it is a politically sensitive topic in any country, and because DECS, at present, does not have sufficient control over the use its own resources. While one should obviously strive for what is politically feasible, there are nonetheless several improvements that appear to be implementable within the prevailing constraints. (a) In general, and wherever possible, DECS should attempt to improve the extent and efficiency of expenditure targeting, away from the prevailing principle of strict proportionality -- i.e., per student, per teacher, or per classroom allocations, or other standard norms applied throughout the country -- towards addressing priority needs identified at the basis of the education system. This implies deconcentration of sector management, program budgeting and execution. (b) Overall, budget appropriations for MOOE ought to be increased substantially, both at elementary and secondary levels. Regional allocations, however, should again be based on actual needs. In operations, expanded provisions for MOOE should flow directly to schools. This will prove to be an essential condition for successful implementation of DECS' stated strategy towards empowering school principals. (c) Prevailing norms underlying sector staffing and allocation of educational inputs tend to have an adverse impact on cost-effective use of available resources. With regard to staffing, for instance, student per class criteria used for the attribution of teacher positions promote padding of student enrollment declared by schools. It is recommended that a thorough assessment of such rules and their effects be conducted, in order to identify possible remedial changes. 7  (d) Improving the quality and retention of public elementary schools is rightly one of the cornerstones of Government's fight against poverty, hence the first priority in terms of intrasectoral allocation of additional resources. Nonetheless, funding for secondary education needs to be increased as well, and quite substantially so if, in addition to quality improvement, the Constitution's commitment to free universal access to basic education is to be sustained. Moreover, raising learning achievements to international college-entrance standards will eventually require the secondary cycle to be lengthened from four to five years. In view of the financing needs of the public secondary education system, the justification for government subsidization of private secondary schools through the GATSPE program would seem questionable. (e) There appears to be a rather large salary differential between teachers in teaching positions and teachers in non-teaching positions, in favor of the latter. Teachers are promoted into non-teaching positions on the basis of seniority and, in principle, of merit. As a result, teaching tends to lose its most qualified practitioners. It may thus be worthwhile to consider extending career opportunities within the teaching profession. (f) The justification of the GATSPE program, in its present form, needs to be reassessed. In any case, if this program is to benefit students from lower income households, and to provide access to private education of acceptable quality, it might be wiser to increase the level of subsidization per student, even if it means reducing the number of beneficiaries. Means testing for the selection of beneficiaries is a must. (g) There is convincing evidence of a strong adverse impact of economic factors on school participation rates. Out-of-school youth represents as much as 18 percent of the 7-12 age group. A case can therefore be made for subsidizing elementary school participation for children from the poorest households, in order to reduce the private cost of education. (h) Surveys also reveal a relatively high incidence of health-related reasons for non- participation among elementary school age children. Systematic institutionalization of school-based feeding and preventive health care programs in the poorest areas therefore ought to be considered. 8  II. Education Finance 2.1 Central Government Financing 26. Constitutionally, education commands the highest priority in the inter-sectoral allocation of government resources.3 At first view, however, the education sector4 appears to receive a rather modest share of the national budget (Table 2.1): 13-14 percent of total expenditure (15-16 percent of recurrent expenditure) in the period 1991-93. Two factors account for this outcome: (i) the heavy burden of the national debt; and (ii) increasing Internal Revenue Appropriations (IRA) to Local Government Units (LGUs) since the enactment of the new Local Government Code of 1991. Debt servicing absorbed over 40 percent of the national budget in that period, and the incidence of IRA transfers increased from 4 percent of the budget in 1991 to about 12.4 percent in 1993. Net of these two obligations, budgetary allocations to education represented 28 percent of effectively available government resources; altogether an appreciable level of commitment. Table 2.1 Central Government Education Expenditure al Million Pesos % of National Budget Year Current 1993 ExI. Debt & % of GNP Prices Prices Total IRA 1985 8,371 16,961 10.5 15.1 1.5 1986 14,046 28,235 12.7 19.5 2.4 1987 15,730 30,484 13.0 33.5 2.3 1988 20,716 36,895 15.2 33.7 2.6 1989 26,741 42,221 15.5 32.0 2.9 1990 32,922 45,772 15.1 31.4 3.0 1991 32,873 38,492 13.3 28.4 2.6 1992 36,633 39,422 13.6 26.2 2.6 1993 38,015 38,015 13.5 28.8 2.5 1994 b/ 46,184 42,377 14.3 31.4 2.6 1995 55,324 47,440 14.0 26.0 2.8 Sources: DECS, DBM, NSCB; 1994 General Appropriation Act, 1995 General Appropriation Bill. a/ DECS, SUCs and NMYC combined (actual expenditure). b/ Estimate. 27. The actual volume of resources available for education financing has also been limited as a result of inadequate resource mobilization, which so far kept government revenues at a relatively modest proportion of national income. All in all, the ratio of government education expenditure to GNP, about 2.5 percent on average through 1991-93, is among the lowest in the East Asia region. 3The 1987 Philippine Constitution indeed prescribes that education should be guaranteed the largest appropriation from the national budget. 41.e., the combined budget of DECS, the State Universities and Colleges (SUCs), and the National Manpower and Youth Council (NMYC). 9  Finally, an extended period of low economic growth (about 1.2 percent p.a. on average between 1989 and 1993) further constrained the level of available government resources. Recent Trends 28. The evolution of government education financing since the early 1980s mirrors the sharp ups and downs of the Philippine economy in that period: (i) severe adjustment and fiscal austerity during the first half of the 1980s; (ii) rapid economic expansion throughout the remainder of the decade following the demise of the Marcos regime; (iii) recession and renewed fiscal austerity in the early 1990s; and (iv) recovery since 1994 onwards. 29. The effects of economic adjustment and fiscal austerity during the first half of the 1980s were severely felt in the education sector. Government spending for education declined by about 30 percent in real terms during that period, and was down to 1.5 percent of GNP in 1985. The financial crisis affected the quality and, to a lesser extent, the expansion of the public education system. Enrollment in public schools continued to increase at all levels, but at substantially lower rates than during the previous decade. The slowdown was particularly significant at the elementary level, where the growth of enrollment fell below the rate of population growth. Real salaries of teaching staff declined dramatically, both in absolute terms and relative to comparable public and private sector employment. 30. From the very beginning, the Aquino Administration initiated a drastic increase in resource allocation to the sector, from less than 10 percent of the national budget in 1985 to 15.5 percent in 1989.5 Total government expenditure for education increased almost fourfold during the remainder of the decade, reaching 3 percent of GNP in 1990. A substantial part of this increase was absorbed by upward salary adjustments. Under the 1989 Salary Standardization Law, teachers' basic salaries were upgraded by 76 percent. This adjustment was subsequently extended to the rest of the civil service, so that the relative position of teachers was only moderately improved. 31. Two important measures further contributed to increase sector financing requirements in that period. First, in 1988 DECS took over the operation of village high-schools, previously funded by LGUs. This "nationalization" was essentially inspired by concern for the relatively low quality of educational services delivered in those schools, and by the commitment towards universal free access to basic education (i.e., including the secondary cycle) expressed in the 1987 Constitution. Moreover, it was feared that the uncontrolled proliferation of these schools would lead to the recruitment of less qualified teachers. Village high-schools practiced a significant degree of cost- recovery through tuition and other fees. The abolition of such fees in the nationalized schools generated a considerable additional increase in public secondary enrollment. The financial consequences of this measure were reflected in a four-fold increase in DECS spending on secondary education between 1987 and 1990.6 5Measured in terms of effective govemment resources (i.e., net of debt service and IRA), the share of education peaked at 33.7 gercent in 1988. From P1.5 billion in 1987 (10 percent of the total education budget) to P6.7 billion (20 percent of the education budget) -- see Table 2.2. 10  32. Secondly, a program for Government Assistance to Students and Teachers in Private Education (GASTPE) was launched in 1989. The GASTPE programn was essentially designed to provide financial assistance to students from lower income families enrolling in private highschools and tertiary institutions. The program started with an appropriation of P590 million (2.5 percent of DECS' 1989 recurrent budget). Funding was gradually increased to P953 million in 1993, but declined afterwards. 33. The nationalization of village highschools in 1988, and the inception of the GASTPE program in 1989, are estimated to have absorbed about 40 percent of the total 1985-89 increment in the DECS budget. The latter was so generous, however, that ample room was left for expansion of other outlays, both recurrent and capital (Table 2.2). Incremental allocations for maintenance and other operating expenses (MOOE), for instance, far exceeded the overall growth in public enrollments -- in real terms, MOOE per student grew by about 17 percent per year in that period. Table 2.2 Evolution of Government Education Financing 1985-95, by Category of Expenditure a 1985-89 1989-93 1993-95 1985-95 I. Average Real Growth p.a.: Total Education 25.6 -2.6 11.7 10.8 DECS Personnel Services 26.9 -1.4 13.6 12.1 MOOE W 21.4 2.0 5.6 10.1 Total Recurrent Exp.cl 27.2 -1.1 11.5 12.0 (Same, per student) (22.8) (-3.1) (9.0) (8.9) Capital Expenditure 47.0 -26.6 28.9 8.4 Total 28.2 -2.4 11.9 11.9 1985 1989 1993 1995 II. Distribution of DECS Exp. (%): Personnel Services 83.2 80.0 82.6 84.7 MOOE 12.6 10.1 12.1 10.7 GASTPE - 2.6 3.0 1.5 Total Recurrent Expenditure 95.8 92.7 97.7 96.9 Capital Expenditure 4.2 7.3 2.3 3.1 Total 100.0 100.0 100.0 100.0 Sources: see Table 2.1 Total education includes the SUCs and NMYC. Actual expenditures for 1985-93 and estimated actual for 1994-95. b/ Excluding GASTPE. et Including GASTPE. 34. Recession and renewed fiscal austerity in the early 1990s brought an end to the rapid expansion of government's education sector spending. Allocations to the education sector fell back to 13-14 percent of total government expenditure throughout 1990-93, equivalent to 2.6 percent of GNP. Total expenditure for education, in real terms, declined on average by 2.6 percent per year between 1989 and 1993, and the overall cost of DECS personnel services by 1.6 percent per year. 11  The volume of MOOE per student is on a downward trend since 1990. Capital outlays underwent drastic cuts, both in 1992 and 1993. 35. The pace of education spending picked up again in 1994, essentially as a result of new government-wide salary adjustments. On balance, a considerable improvement in the sector financing situation has thus been achieved during the last decade. Nevertheless, in view of the bleak initial conditions, it appears unlikely that this improvement was sufficient, overall, to fully meet the growing quantitative and qualitative needs of the public education system. There are four areas of concern, in this respect: (i) the rapid increase in salaries; (ii) the erosion of MOOE; (iii) the chaotic year-by-year evolution of the various categories of expenditure; and (iv) the adverse impact of fiscal decentralization on DECS' effective budgetary resources. 36. Salaries. Overall, the total compensation package of teachers recorded a fourfold increase between 1985 and 1995, equivalent to a 5 percent annual improvement in real income. Although this compares favorably with the 1.4 percent growth of per capita GDP achieved during the same period, the evidence suggests that the remuneration level of public school teachers is still relatively moderate (Table 2.3). Not suprisingly, therefore, the raise in basic teacher salaries and other compensations granted in 1994 and 1995 are part of a longer term package, with additional adjustments in the pipeline. Taking into account normal wage drift and incremental recruitrnent, this could end up boosting DECS' wage bill to an even much larger extent. Crucial non-salary outlays are thus at risk of being crowded out under the mounting pressure of personnel costs. Table 2.3 Teacher Remuneration Ratio of Teacher Salary' to: Poverty Threshold Minimum Wage GNP per Capita 1985 .9 1.1 2.0 1988 1.2 1.6 2.4 1991 1.1 1.3 2.4 1994 1.2 1.3 2.5 Source: DECS; Philippine Statistical Yearbook; World Bank. aStarting salary elementazy teacher (including various allowances) 37. MOOE. On a per student basis, this category of expenditure has been on a downward slide since 1990. The overall recovery in government education spending in 1994 did not interrupt this trend. And although budget appropriations for MOOE suggest a sizable increase for 1995, past discrepancies between budget allocations and actual spending raise some doubt as to whether the budgeted increase has been fully realized. In total, MOOE per student declined by as much as 34 percent in real terms between 1991 and 1994. This erosion of earlier achievements is a source of concern, as it may have significant adverse impacts on school maintenance and the provision of essential educational inputs. In addition, there are serious questions as to the proportion of MOOE actually reaching down to the school level. 12  38. Stop-go financing patterns. Underlying the evolving trends illustrated in Table 2.2, are rather wide year-to-year fluctuations in expenditure patterns, particularly as regards capital operations and MOOE. The resulting unpredictability of funding levels frustrates rational educational planning efforts, and is a major impediment for effective sector management. 39. Fiscal decentralization. Rising IRA transfers to LGUs have exacerbated DECS' resource constraint. According to the new Local Government Code of 1991, the financial responsibility for school construction and maintenance in basic education was to be devolved to LGUs. So far, however, there has been no formal devolution of educational services. With central government resources diminished by IRA transfers, DECS should therefore have received an increasing share of available resources to sustain adequate delivery of basic education services to a rapidly growing school age population. Moreover, considerable investments are required (many mandated by Parliament) to improve the quality of those services. In reality, the required proportional adjustment in resource allocation to education did not materialize. Instead, the share of the national budget (net of debt service and IRA) accruing to DECS declined from 27 to 23 percent between 1990 and 1992, and remained since then within this margin. As a result, earlier commitments, such as free secondary education, are becoming increasingly difficult to sustain effectively. Intra-sectoral Distribution of Expenditure 40. Regrouping the various education-related components of the government budget into a reliable breakdown of expenditure by level and type of education is a cumbersome operation. Public elementary and secondary education is, in principle, financed through DECS. However, there are also elementary and secondary schools attached to teacher training faculties in State Universities and Colleges (SUCs), which are funded trough direct budget appropriations. SUCs, moreover, also offer vocational and technical courses which do not normally belong to higher education. DECS also transfers resources to private schools at the secondary and tertiary level (GASTPE). 41. With these caveats in mind, the intrasectoral breakdown of education expenditure is shown in Table 2.4. About 85 percent of central government appropriations for education are channeled through DECS. Most of the rest is absorbed by the State Universities and Colleges (SUCs), which are funded through direct appropriations. Since 1991, minor but rapidly growing financing is also being allocated to the National Manpower and Youth Council (NMYC2, which is essentially responsible for non-formal manpower training and skills upgrading programs. 42. DECS' share of total education financing has remained virtually constant throughout 1987-94. Neither repeated hefty salary adjustments,9 nor the nationalization of village highschools, 7Since 1993, the volume of IRA transfers from central government to LGUs is roughly equal to the entire public education budget 8Minor amounts of education sector financing, essentially for on-the-job staff training, can further be identified in the respective budget allocations of several Departments. These are omitted here. 9Normally, under constant relative staffing patterns, a proportional wage increase in two sub-sectors should raise the funding requirements of the most labor-intensive one relative to the other. On the whole, DECS (i.e., basic education) is considerably more labor-intensive than the SUCs -- personnel expenditure were absorbing 75 and 66 percent of the respective budgets in 1994. 13  or the establishment of the GASTPE program, have managed to raise DECS' share in the government's total education budget. On the other hand, the establishment of NMYC did not alter DECS' share either; so far, this new claim on public resources has been at the expense of the SUCs. This uncommon rigidity in the global distribution of education finance suggests tension between prevailing resource appropriation rules and procedures, on the one hand, and rapidly evolving sector circumstances and needs, on the other. It is a first indication that DECS may not have been granted sufficient additional resources to meet the new commitments entrusted to it in that period. Table 2.4 Government Financing, by Level and Type of Education (% Distribution) 1 1987 1988 1989 1990 1991 1992 1993 1994 Elementary 60.0 56.6 42.3 40.4 41.9 50.8 54.1 55.5 Secondary i 9.8 18.5 21.3 20.2 18.3 14.8 15.9 16.7 Tertiarya, b 16.8 16.9 17.7 18.0 16.1 16.4 17.6 16.4 Vocational d 2.1 1.2 .7 .6 3.1 3.0 3.3 3.3 Other d- .0 .0 .0 .4 DECS Central Adm. 11.3 6.8 18.0 20.8 20.6 15.0 9.1 7.7 Overall Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 (o.w. Total DECS) (84.9) (85.0) (84.5) (84.4) (86.5) (85.9) (84.7) (85.1) Source: 1987-94 General Appropriations Acts 1989-90 SUCs Expenditure Program v Including GASTPE from 1989 onwards. b/ SUCs, plus DECS-managed tertialy institutions. C/ DECS-managed programs, plus NMYC from 1991 onwards. d/ Pre-school and non-fornal programs (DECS). 43. Over the medium term, the nationalization of LGU-funded highschools has not dramatically altered the intra-sectoral distribution of resources. The final outcome, however, is the net result of two counterbalancing rounds of adjustment. First, the share of secondary education shot up from an initial 10 percent to 18.5 percent in 1988, and up to 21 percent the year after. This sudden increase was matched by a decline in the share of elementary education, from 60 to 42 percent. In those two years, DECS' regional operations budget increased by P7.8 billion. As much as P4.1 billion of this increment (52 percent) was allocated to public secondary education, which accounted for only 21 percent of DECS-controlled student enrollment.'0 From 1990 onward, the increasingly severe constraints facing DECS' budget affected primarily secondary education, and the respective shares of sector resources moved back towards their original level. These consecutive rounds of adjustment illustrate DECS' difficulty in balancing conflicting budget claims.