Who benefits from public education spending in Malawi? results from the recent education reform

99 PSP Discussion Paper Series 19712 September 1996 Who Benefits from Public Education Spending in Malawi? Results from the Recent Education Reform Florencia Castro-Leal September 1996 Poverty and Social Policy Deparment Human Capital Development The World Bank  PSP Discussion Papers reflect work in progress. They are intended to make lessons emerging frore the current work program available to operational staff quickly and easily, as well as to stimulate disciussion and comment. They also serve as the building blocks for subsequent policy and best practice papers. The views expressed here are those of the authors and should not be attributed to the WVorld Bank or its Board of Executive Directors or the countries they represent.  CONTENTS FOREWORD .......................................................... iv ABSTRACT ...........................................................v ACKNOWLEDGMENTS .......................................................... vi EXECUTIVE SUMMARY .......................................................... vii I. INTRODUCTION ...........................................................1 II. SCHOOL ENROLLMENT .................... .......................................2 Important gains in school attendance ..........................................................2 Enrollment rates are low and the poorest children are least likely to be in school ... 3 The poorest girls are least likely to go to school and most likely to drop out ........ .... 4 Enrollment rates are higher in the Northern region . ................................................. 5 There is considerable age/grade mismatch ............................................................ 6 Most primary school graduates never attend higher education .............. .................. 7 m. RECENT CHANGES IN ENROLLMENT ....................................................9 Important inequities remain in primary education ...................................................9 IV. PUBLIC SPENDING ON EDUCATION ...................................................... 12 Public spending on education more than tripled ...................................................... 12 Public spending on primary education more than five-fold ................. .................... 13 Services and quality have deteriorated as enrollments have surged ........... ............. 15 Public spending on primary education increased, but bias against primary education persists ........................................................... 17 V. WHO BENEFITS FROM PUBLIC EDUCATION SPENDING? ................ 20 Public spending in education has become more pro-poor ............... ......................... 22 ii  More primary school children are among the poor .................................................. 28 Girls receive about half the public education spending received by boys ......... ....... 29 Public primary education resources are smaller in the rural Center and rural South because enrollments still lag behind ................... ..................................... 29 VI. POLICY IMPLICATIONS .................. ...................................... 33 Expanding access and reducing remaining inequities in education ........... .............. 33 Increase effectiveness and quality of primary education ....................... ................... 33 REFERENCES ........................................................ 34 APPENDIXES A: School enrollment and public education spending by quintiles, regions and gender ........................................................ 37 B: Methodology used to compute the distribution of education spending by socioeconomic groups ........................................................ 45 . .  FOREWORD The new Government of Malawi has made education, particularly primary education, its top priority. After public recurrent spending on education surged and primary school fees were abolished in 1994, Malawian households responded and primary-level enrollments jumped by more than 60 percent. Drawing upon our operational support to the Region, this study uses the Benefit Incidence Analysis to examine the distribution of public spending in education across different socioeconomic groups. The Malawi example shows that an effective way to increase the equity of public spending in education is to increase the total allocation of public resources to primary education while easing the constraints on the demand for education faced by poor households. Recent public budgetary changes favoring primary-level spending, the removal of all primary school fees and the surge in enrollments at this level had a significant impact on the distribution of public education spending in Malawi. This is a very significant development. The challenge in the near future is to maintain current enrollment levels in primary schools and simultaneously to improve the quality of education. A decline in school enrollments among the poor would mean that the Malawian education reform could run the risk that recent gains in allocating resources to them would be reversed. The Poverty and Social Policy Department hopes that policymakers, researchers, and practitioners engaged in the design of education reform programs will find this study useful and timely. Ishrat Husain Director Poverty and Social Policy Department iv   ABSTRACT Malawi's education system is hampered by problems of poor access, high repetition and drop-out rates, poor infrastructure and inequality. To address this situation the new Government of Malawi has made education, particularly primary education, its top priority. After public recurrent spending on education surged and primary school fees were abolished in 1994, Malawian households responded and primary-level enrollments Jumped by more than 60 percent. This study uses the Benefit Incidence Analysis to examine the distribution of public spending in education across different socioeconomic groups before and after these changes. Recent public budgetary changes favoring primary-level spending, the removal of all primary school fees and the surge in enrollments at this level had a significant impact on the distribution of public education spending in Malawi. In 1994-5, the poorest one-fifth of the population received 16 percent of all public education spending compared to only 10 percent in 1990-1, while the share going to the richest one-fifth of the population declined from 38 to 25 percent between the same years. The Malawi example shows that an effective way to increase the equity of public spending in education is to increase the total allocation of public resources to primary education while easing the constraints on the demand for education faced by poor households. The challenge in the near future is to maintain current enrollment levels in primary schools and simultaneously to improve the quality of education. A decline in school enrollments among the poor would mean that the Malawian education reform could run the risk that recent gains in allocating resources to them would be reversed. v  ACKNOWLEDGMENTS This study was a background paper to the report: Malawi Human Resources and Poverty - Profile and Priorities for Action; prepared at the request of the Government of Malawi as the World Bank's input for their Poverty Alleviation Program. The author expresses her gratitude to the Government of Malawi for its suppcrt and cooperation in providing in:formation and data reviewed in this study The author gratefully acknowledges the support received from Helena Ribe in producing this paper. David Bruns provided generous and timely information on macroeconomic indicators, public education spending and enrollments in Malawi. The author also wishes to thank Trina Haque, Ingeborg Kleppe, Peter Ngomba, Carolyn Winter and Mulu,eta Wodajo for very helpful comments on an earlier draft. Meera Venkataraman provided excellent computational analysis of household survey data. vi  EXECUTIVE SUMMARY' Malawi's education system is hampered by problems of poor access. high repetition and drop-out rates, poor infrastructure and inequality. To redress this situation the new Government of Malawi (GOM) made education, particularly primary education, its top pnority. In 1994, public recurrent spending on education more than tripled in real terms compared with its level in 1990; it went from 268 million Malawi Kwacha (MK) to MLK900 million, expressed in constant 1995 MNK Primary-level spending more than five-fold in real terms during the same period, and its share has increased from less than half to almost three- quarters of total public education spending. This has been a very positive and needed development. As a result of the recent education reform, the share of public education resources received by the poor increased substantially between 1990/1 and 1994/5 (see Table A). In 1990/1, the poorest quintile received only 10 percent of all public educationi spending, while the share going to the richest quintile was 38 percent. In 1994/5, there was a substantial reduction in the inequality of public expenditures for education; ihe share received by the poorest quintile increased to 16 percent and the share going to the richest quintile decreased to 25 percent. Table A Incidence of public education spending on the poorest and richest quintiles, 1990/1 and 1994/5, percentage share Education spending benefiting: Educational Year the poorest 20% of the the richest 20% of th Level population (%) population (%) All Education 1990/1 10 38 Primarv 15 24 Secondary 7 41 Tertiary 3 61 AU Education 1994/5 16 25 Primary 20 16 Secondary 9 39 Tertiary 1 58 Sources: HESSE4 (1990/1); MOE (1993, 1995); GOM National Budget (1995) Note: Ifpublic education spending were equally distributed across population quitnles, then the poorest and rchesr quintiles (every number in this table) would receive a 20 percentage share of spending. vii  In 1994, the government abolished primary school fees as part of its commitment to raising the education levels of the population. Malawian households responded to this new priority and the number of children enrolled at the primary level jumped by 62 percent from 1.8 million to near three million students for the 1994/5 school year. Growth rates in primary school enrollments have been highest in the rural Central and the rural Southern regions - the poorest regions and those regions that had lagged behind in net enrollrment rates in 1990/1. By contrast, enrollment rates in the rest of the countryv increased by less than 30 percent. Despite the recent improvements, there are still serious problems and severe inequalities in public primary education. The quality of education., already severely compromised by large class sizes even before the recent surge in primary enrollment, has deteriorated further. The bulk of the recent increases in primary education spending has been spent on teachers' salaries. In 1994/5, recurrent spending on salaries consumed 97 percent of all primary recurrent expenditures. Although public education recurrent spending has increased to 7.5 percent of the Gross Domestic Product (GDP) in 1994/5, there continues to be a dearth of even the minimal number of essential teaching materials and human resources necessary to educate children. There have been substantial gains in attendance at all levels over time, and the number of Malawians who have attended school has increased significantly. However, the poorest children are least likely to be in school and are most likely to drop out or to leave and return to school repeatedly, thus decreasing their chances of completing the primary cycle. Among the poorest children, girls are least likely to go to primary school and more likely to drop out than boys. Only 18 percent of all children enrolled in primary school complete the full eight years of primary education. Thus, most children never attend higher education, owing also in part to the lirnited number of places available at secondary schools. Recent public budgetary changes favoring primary-level spending, the removal of all primary school fees and the surge in enrollments at this level had a significant impact on the distribution of public education spending in Malawi. The Malawi example shows that an effective way to increase the equity of public spending in education is to increase the total allocation of public resources to primary education while easing the constraints on the demand for education faced by poor households. The challenge in the near future is to maintain current enrollment levels in primary schools and simultaneously to improve the quality of education. Inequities in the distribution of education resources still remain across regions, within educational levels and by gender. A decline in school enrollments among the poor would mean that the Malawian education reform could run the risk that recent gains in allocating resources to them would be reversed. viii  I. INTRODUCTION The recent policy reform in the education sector as well as the rapid increase in primary enrollments in 1994 are very significant developments in Malawi. This paper explores both issues separately and then brings them together to see if there are any changes in the distribution of social benefits. There are two main factors that influence prevailing inequities in the amount of public education resources received by the population. First, there is the allocation of public resources to individuals in different socioeconomic groups, regions and by gender as a result of their consumption of public services. Second, there is the allocation of government spending within the education sector. The distribution of public education spending can be examined across different socioeconomic groups, regions and by gender through the allocation of per unit public spending according to individual utilization rates of public services. The methodology is known as Benefit Incidence Analysis and it measures how well public services are targeted to certain groups of the population, for example the poor and girls. The 1990/1 Household Expenditure and Small-Scale Economic Activities (HESSEA) survey, conducted by the National Statistical Office of Malawi, was used to obtain school enrollment information. HESSEA is the only data set available in Malawi that can be used to analyze education indicators linked to household income and expenditure characteristics. Enrollments by educational levels for 1992/3 and 1994/5, as well as sectoral public spending information, were provided by the Ministry of Education (MOE). HESSEA and MOE data were used to compute per capita public education spending by income quintiles in 1990/1 and 1994/5. This paper looks at educational indicators in Section B. Public education spending and the allocation of public resources within the education sector are explored in Section C. Section D analyzes the distribution patterns of public education spending across socioeconomic groups and gender. The last Section discusses the most relevant policy implications coming from this study. l  H. SCHOOL ENROLLMENT * Important gains in school attendance There has been a substantial increase in the share of young adults receiving some primary education, completing primary school, and even attaining higher levels of education compared with the share of older adults who reached those levels in the past (see Figure 1). The number of Malawians who have never attended school has been significantly reduced. About one in three Malawians currently aged between 16 and 3 5 years old have never attended school, compared to one in two for those aged 35 years or older. The share of the population aged between 16 and 35 years old that has attained an educational level of at least Standard V to VIII has doubled compared with the populaticn over 35 years of age. Figure 1 Highest education level attained by adults in two age groups, 1990/1 70 T MALES I 70 FEMALES 60 60' 40 400 3020 I 20 10 10 0 0 No Std I-IV Std V-V7II Secondary No Std. I-IV Std. V-V/7I1 Secondary Education & Tertiary Education & Tertnary i 35+ years old M '-35 years oi1 35 + years old * 16-35 years old Source: HESSEA (1990/1) Note:Appendix TablesA.] andA.2includepercentage numbers. It is also clear from comparing these two age groups that disparities between men and women in educational attainment have declined, although the percentage of women who have never attended school remains higher than the comparable percentage of men. Among males, the share of those who have never attended school has declined from one in three for the older group to one iin five for the younger group (see Figure 1). Among females, this share has declined from two in three for the older group to less than one in 2  two for the younger group. Thus, gender disparities are still substantial among adults aged 16 to 35, but inequities between males and females are declinin- over time. The largest increase over time in school attendance in Standards V to VIII was among women. One out of every four women in the younger group reached at least Standards \V to VIII, which is three times the number of the older group of females (only 9 percent) who reached this level previously. * Enrollment rates are low and the poorest children are least likel to be in .school Only about half of all children six to thirteen years of age were enrolled in primary schools in 1990/1 and children in the lower expenditure groups are much more likely to be out of school. Net primary enrollment rates are more than 40 percentage points lower in the poorest than in the richest income quintile (see Table 1). 1.2 The gross primary enrollment rate is substantially higher than the net rate and this difference is more marked for the poorest than for the richest income quintile. Gross enrollment rates in primary school for the two lowest income quintiles are about double the net enrollment rates, confirming that late entry and repetition are more common among lower-income groups. Thus, the poor are more likely to drop out or to leave and return repeatedly, decreasing their chances of completing the primary cycle. Table 1 Net and gross enrollment rates in primary education, 1990/1 by quintiles and gender Hh Expenditure Net Enrollment Rates Gross Enrollment Rates quintile Boys Girls National Boys Girls National I - Poorest 34 31 33 65 51 58 n so 45 48 83 69 76 II 52 57 55 88 83 86 IV 66 61 62 104 89 97 V- Richest 76 75 75 113 106 110 All 52 50 51 86 75 81 Primary education in Malawi takes a minimum of 8years and it goesfrom Standard] to Standard 1711. Net enroLlment rate: 6-13 year olds enrolled in primary school as percentage of 6-13 year oldpopulation. Source: HESSEA (1990/91) Gross enrollment rate: All children enrolled in primary school as percentage of 6-13 year oldpopulation, Source: HESSE'- (1990/91) I Income quintiles are created by ranking every individual from the poorest to the richest and then dividing the population into five groups each containing exactly 20 percent of all individuals. Thus, the poorest income quintile contains the poorest one-fifth of the population while the richest income quintile contains the richest one-fifth of the population. 2 We use expenditure quintile and income quintile interchangeably throughout the paper, but bear in mind that the welfare measure for ranking individuals is total household expenditure per adult equivalent. 3  Gross enrollment rates are much lower for Standards V to VIII than for Standarcis I to IV. The decline in gross enrollment rates between the first- and the second-half of primary education is sharpest among the poorest children: two in three drop out compared to less than one in two among the richest children (see Table 2). In addition. the poorest children also have the lowest enrollment rates in both the first- and second-half of primary education. Table 2 Gross enrollment rates in standards I to IV and standards Vto 171H, 19901/ by quintiles and gender Gross Enrollment Rates Gross Enrollment Rates 1 H/h Expenditure Standards I to IV Standards Vto VlI quintile Boys Girls National Bovs Girls National I - Poorest 96 70 82 36 27 32 1 108 98 104 53 38 45 III 116 116 116 55 42 48 IV 132 115 123 76 58 68 V- Richest 133 151 142 88 f68 77 oll 114 102 108 56 43 50 Primary education in Malawi takes a minimum of8 years and it goes from Standard I to Standard Vii7. Gross enrollment rate in standards lto IV: All children enrol,ed in standrds Ito lVas percentage of 6-9 year old population, Gross erolbnentrate in standards Vto V7II: All children enrolled in standards Vto V111 as percentage of 10-13 year old population, Source: HESSE4 (1990/91) The poorest girls are least likely to go to school and most likely to drop out Gender disparities in net enrollment rates at the primary level do not appear to be significant (see Table 1), but the gross rate is systematically lower for girls than for boys across all the income quintiles. The poorest girls have the lowest enrollment rates in the first- as well as the second-half of primary education (see Table 2), and enrollment rates decrease substantially between these two cycles. These findings are an indication that girls aged 13 or older (captured in the gross rate) are more likely to drop out of school than boys aged 13 or older. Thus, it is crucial to put in place incentives for early primary school enrollment and retention of girls, particularly among the poorest and in the Southern region, as girls are more likely than boys to drop out as they get older. Gender disparities in overall gross enrollment rates for primary education are considerably higher in the South than in the North or Central regions (see Appendix Table A.3). Boys in the rural South have an overall gross enrollment rate about 25 percent higher than the rate for girls, while in the rural North and Central regions, the gross rates are about the same for boys and girls. 4  Drop out from primary school amono girls is particularly severe in the rural South. The gross enrollment rate for girls in the rural South drops by 60 percent between the first- and the second-half of primary education (see Appendix Tables A.4 and A. 5) Girls in the rural South have an 89 percent gross enrollment rate in standards I to IV wvhile it is only 29 percent in standards V to VIII. * Enrollment rates are higher in the Northern region Children in the Northern region, from urban as well as rural areas, are the most likely to be enrolled in primary school. By contrast, rural net and gross enrollment rates in the Center and South are almost half of those in urban areas (see Table 3). Less than 50 percent of the children six to thirteen years of age were enrolled in primary schools in these rural regions in 1990/1. Disparities between rural and urban areas are also narrower in the Northern region than in the Center and South. Overall regional disparities are largely explained by lower rural enrollment rates in the rural Center and South. Table 3 Net and gross enrollment rates in primary education, 1990/1 by region and area of residence Net Enrollment Rates Gross Enrollment Rates Region Rural Urban Total Rural Urban Total North 73 84 75 117 129 118 Center 45 76 49 74 112 78 Souths 43 81 47 70 114 75 National 47 79 51 77 115 81 Source: HESSEA (1990/1) Children in the Northern region are the most likely to remain enrolled throughout all years of primary school. By contrast, children in the rural Center are most likely to drop out from primary school and children in the rural South have low primary enrollment rates throughout primary education. Children in the rural Center have higher gross enrollment rates in Standards I to IV than children in the rural South (see Table 4). However, these two regions have the same rates for Standards V to VIII (see Table 4). Educational policies for the rural Center and the rural South ought to approach two different sources of low completion. Primary schools in the rural Center need to increase their ability to retain children in school. In the rural South, enrollments in primary schools need to increase altogether. 5  Table 4 Gross enrollment rates in standards I to IVand standards Vto ilII, 1990/1 by region and area of residence Gross Enrollment Rates Gross Enrollment Rates Region Standards Ito IV Standards V to V1I1 Rural Urban Total Rural Urban Total North 139 150 140 94 106 | 9 I Center 103 138 107 38 85 4- South 98 146 103 38 86 44 National 104 143 108 45 87 70 Source: HESSEA (1990/1) * There is considerable age/grade mismatch Late entry is very common, and older students are more likely to be enrolled at every educational level than are students of the correct age group (see Table 5). The age/grade matching analysis also shows a bottleneck to complete primary school at Standards V to VIII, confirming that repetition is considerably higher at the end of this cycle. T'able 5 Age/Grade Matching, 1990/1 Number Age Group Level of students Percentage of total enrolled in level enrolled 5 or less 6-9 10-13 14-15 16-17 18 or Total ______________ __________ ____-___ more Standard I-IV 993,377 2 34 44 12 5 4 100 Standard V-VIII 407,305 0 2 25 25 24 25 100 FormsI-_Ia 41,876 0 0 4 9 18 69 100 Forms Ill-IlV 17,839 0 0 0 2 8 89 100 Forms V-VI 145 0 0 0 13 11 76 100 Tertiarj' 6,991 0 0 0 0 0 100 100 a Includes M. C-D.E. (Malawi College ofDistance Education) students b Tertiary includes Primary Teacher Training. Technical Training and University Source: HESSEA (1 990/1) About 66 percent of all students enrolled in Standards I-IV are older than nine years of age, and around 75 percent of all students enrolled in Standards V-VIII are older 6  than 14 years of age (see Table 5). In Forms I-IV in secondary school. 75 percent of all students enrolled are older than 18 years of age. A recent study found that the average age in Standard I is 10 years of age and, in Standard VIII it is 22 years of age..' The main disadvantage of attending primary school at an older age is that these children are often needed to help to support the household. This becomes a disincentive for them to complete that level of education and to proceed to higher educational levels. a Most primary school graduates never attend higher education4 The number of students completing each educational cycle and enrolling into the next is extremely low. In 1990/1, only 18 percent of the children enrolled in primary school completed the full eight years of primary education (UNICEF, 1993). In addition, secondary schools offer only a limited number of places with the result that most children never attend secondary school. Only one out of every 100 children entering the primary cycle will be admitted to the secondary cycle. The overall net enrollment rate in secondary schools is 2.2 percent, and the gross rate is about 10 percent. Given the recent surge in enrollments and the deterioration in the quality of primary schools, it is possible that a proportionally smaller number of children who complete primary schooling will attend secondary school in the near future, particularly in the Center and South. Table 6 Net and gross enrollment rates in secondary education, 1990/1 by quintile and gender Hh Expenditure Net Enrollment Rates Gross Enrollment Rates quintile Males Females National Males Females National 1- Poorest 0.2 0.4 0.3 5.8 1.2 3.6 11 0.4 1.5 1.0 5.7 3.2 4.4 III 3.0 1.4 1.7 11.9 3.4 7.7 IV 2.4 3.0 2.7 19.5 12.7 15.9 V- Richest 10.4 6.5 8.3 40.5 20.0 29.4 Al 2.5 2.0 2.2 13.6 6.8 10.2 Secondary education includes M.C.D.E. (Malawi College ofDistance Education) students Primary education in Malawi takes a minimum of 4 years and it goesfrom Form I to Form IV Net enrollment rate: 14-1 7year olds enrolled zn secondary school as percentage of 14-1 7year old population. Source. HESSEA (1990/91) Gross enrollment rate: All children enrolled In secondary school as percentage of 14-1 7year old population, Source: IIESSEEA (1990/91) 3 Hyde et al. 1994. 4 Higher education refers to all post-primary education and thus, it includes both the secondary and tertiary levels. 7  Practically no children from the two poorest income quintiles attend secondary school (see Table 6). Even in the highest income quintile, only 8 percent of youngsters aged 14 to 17 years old are enrolled at secondary school. Regional disparities in enrollments at the secondary-level indicate that Central and Southern gross enrollment rates are only two-thirds of the Northern rate (see Appendix Table A.6). Gender disparities in enrollments at the secondary-level are substantially wider than at the primary level. Overall, gross enrollment rates for males are twice as large as for females (see Table 4). Females in the Central region are particularly disadvantaged, the gross enrollment rate is two and a half times higher in favor of males (see Appendix Table A. 7). Because transition rates, the proportion of students completing each cycle and enrolling in the next, are extremely low, it is also rare for Malawians to attend tertiary education. There are only about 7,500 tertiary-level students in Malawi. including students attending Primary Teacher Training, Technical Training, and University. In 1994/5, these enrollments represent a mere 0.3 percent of the total number of enrollments at all levels of education. 8  III. RECENT CHANGES IN ENROLLMENTS The response from households and students in Malawi to the elimination of school fees and the priority given by the new government to primary education has been overwhelming. This section analyzes the distribution of the increase in enrollments across income groups, regions and gender using recent data from the Ministrv of Education * Important inequities remain in primary edlucation Estimated gross primary enrollment rates indicate that children from lower income groups who were identified as being disadvantaged in 1990/1 are likelv to remain disadvantaged in 1994/5 (see Table 7). Major increases in gross enrollment rates occurred in the 1994/5 school year at every income level, however, half as many children from the poorest income quintile are enrolled in primary school as from the richest income quintile. The gross enrollment rate for the poorest income quintile is 74 percent, while the richest income quintile has a 134 percent rate. Table 7 Gross enrollment rates in primary education, 1990/91 and 1994/5, * by quintiles and gender Hh expenditure 1990/1 1994/5 quintiles Boys Girls National BOYS Girls National 1- Poorest 65 51 58 100 69 74 II 83 69 76 117 88 102 III 88 83 86 118 98 114 IV 104 89 97 134 104 131 V- Richest 113 106 110 134 120 133 AI 86 75 81 121 96 108 ' 199415 enroliment rates are estimated on the basis of enrollments by regions provided by the MOE (1995) and tak/ng HESSEA data as our baseline (seeAppendixBfor a description of the algorithm). The population by age grops in 1994/5 is estimated on the basis of the 1990/1 population and using a 3.3 percent annual population growth rate. Gross enroUmentrate: All children enrolled in primary school as percentage of 6-13year old population Source: HESSEA (1 990/91); MOE ( 993); MOE (1995) Gender disparities in gross enrollment rates have increased for all income groups, but girls in the poorest income quintile have the lowest gross enrollment rate in the country - a gross enrollment rate of only 69 percent, almost one-third lower than boys. Estimated gross primary enrollment rates by region indicate that, although the regional disparities are narrowing, the same regions that were relatively disadvantaged in 9  1990/1 are still disadvantages in 1994/5. Growth rates in primary school enrollments have been highest in those regions that had lagged behind in net enrollment rates in 1 990/1. Overall net enrollment rates for primary education in 1990/1 were lower in rural than in urban areas, and net enrollment rates in rural areas in the Central and Southern remions were almost half those in urban areas in the same re-ions5. In recent vears. these disparities have been narrowing. Between 1992/3 and 1994/5, primary enrollments in the rural Central region increased by 70 percent, and in the rural Southern region. by 80 percent (see Figure 2). By contrast, enrollments in the rest of the country increased by less than 30 percent. Figure 2 Growth in primary enrollmentsfrom 1992/3 to 1994/5 bY region 90 - URBAN 9g - RURAL 80 80- 70 70 -_ 60T 60- 407 40+ 30t [ 30