Tanzania - Social sector review

H_- 19809 October 1999 SAr - a _ e:..~~~~~~ ..0. idli- -A. ow . s l~~~~~~~~4 A6W > _!   A WORLD BANK COUNTRY STUDY Tanzania Social Sector Review The World Bank Washington, D.C.  Copyright (D 1999 The International Bank for Reconstruction and Development/THE WORLD BANK 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United States of America First printing October 1999 World Bank Country Studies are among the many reports originally prepared for internal use as part of the continuing analysis by the Bank of the economic and related conditions of its developing member countries and of its dialogues with the governments. Some of the reports are published in this series with the least possible delay for the use of governments and the academic, business and financial, and development communities. The typescript of this paper therefore has not been prepared in accordance with the procedures appropriate to formal printed texts, and the World Bank accepts no responsibility for errors. Some sources cited in this paper may be informal documents that are not readily available. The findings, interpretations, and conclusions expressed in this paper are entirely those of the author(s) and should not be attributed in any manner to the World Bank, to its affiliated organizations, or to members of its Board of Executive Directors or the countries they represent. The World Bank does not guarantee the accuracy of the data included in this publication and accepts no responsibility for any consequence of their use. The boundaries, colors, denominations, and other information shown on any map in this volume do not imply on the part of the World Bank Group any judgment on the legal status of any territory or the endorsement or acceptance of such boundaries. The material in this publication is copyrighted. The World Bank encourages dissemination of its work and will normally grant permission promptly. Permission to photocopy items for internal or personal use, for the internal or personal use of specific clients, or for educational classroom use, is granted by the World Bank provided that the appropriate fee is paid directly to Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, U.S.A., telephone 978 750 8400, fax 978 750 4470. Please contact Copyright Clearance Center prior to photocopying items. For permission to reprint individual articles or chapters, please fax your request with complete information to the Republication Department, Copyright Clearance Center, fax 978 750 4470. All other queries on rights and licenses should be addressed to the World Bank at the address above, or fax no. 202 522 2422. ISBN: 0-8213-4585-0 ISSN: 0253-2123 Cover photo taken in 1996 of girls who received scholarships to attend secondary school in a pre-test of the program, which was a direct result of the analysis of this document. This first cohort of girls will graduate from lower secondary school in December 1999. Library of Congress Cataloging-in-Publication Data Tanzania: social sector review. p. cm. - (World Bank country study) Includes bibliographical references. ISBN 0-8213-4585-0 1. Human services-Tanzania. 2. Tanzania-Social conditions. I. World Bank. II. Series. HV448.5.T36 1999 362.9678-dc2l 99-047476  CONTENTS Acronyms and Abbreviations ..................... xi Abstract ..................... xiv Acknowledgments ........................... xv Preface ..................... xvi Executive Summary ..................... I Background .35 Objectives of the Social Sector Review .35 Key Definitions .37 Data Sources .37 Organization of the Report .38 Importance of Human Capital and Human Resources to Development in Tanzania .38 Growth and Human Welfare .38 Human Resources as Capital .39 Unique Features of Human Capital ...................................... 41 Producing Human Capital .42 Paying for Investments in Human Capital .44 Summary .45 2 Development and Structure of the Social Sectors .46 Three Eras .46 Phase I: Independence to Early-1980s .46 Phase II: Early-1980s to Early-1990s . 48 Phase III: Current Era (early-1990s) .50 Delivery of Social Services Through Local Government .51 Local Government Institutions .51 Local Government Allocation Process and Accountability .53 Public Spending and the Distribution of Benefits .55 Government Spending .55 Donor Spending .56 Incidence of the Benefits of Public Spending .57 Conclusion .59 3 Characteristics of the Tanzanian Household .60 Household Structure .60 Sources of Income and Living Conditions .61 Ownership of Goods .61 Housing Conditions .62 Ownership of Land and Livestock .62 Human Capital .63 iii  Expenditure and Consumption Patterns ............................................ 63 Conclusion .............. ..................................... 64 4 Education ............... ..................................... 65 Outcomes and Enrollment .............................................. ... 65 Primary School . ................................................. 66 Secondary School ................................................. 69 Higher and Tertiary Education and Training .......... ....................... 72 Increasing Post-Primary Education Opportunities ........ .................... 73 Household Expenditures on Education ............................................. 75 Overall Expenditures ................................................. 75 Primary School Expenditures . ........................................... 76 Secondary School Expenditures . .......................................... 76 Opportunity Cost of Education . .......................................... 78 Demand for Education . ................................................. 79 Parents' Perceptions of the Quality of Education ........ ..................... 79 Willingness to Pay for Improved Education Services ....... ................... 80 The Supply of Education Services ............................................... 81 Policy Framework ................................................. 81 Organization and Structure of the Education System ........ .................. 84 Teachers, Textbooks, and Management in Primary Education ...... ............. 90 Government and Donor Spending on Education ............ ........................ 95 Recurrent Expenditures ................................................. 95 Expenditure by Program ................................................ 95 Expenditure by Input ................................................. 96 Donor Spending . ................................................. 98 Unit Costs . ................................................. 99 Returns to Education . ................................................. 101 Private Returns .0................................................. 10 Social Returns . ................................................. 102 The Distribution of Benefits from Education Spending ......... ..................... 103 Total Spending on Education ............................................ 103 Distribution of Public Subsidies . ........................................ 103 Conclusion ............... .................................... 104 5 Health ................................................... 108 Outcomes and Utilization ................................................. 109 Mortality Levels .................................................. 109 Causes of Illness and Death . .......................................... 110 Utilization of the Health Care System ............. ........................ III Household Expenditures on Health Care . ......................................... 113 Demand for Health Care ............... ...................................... 115 Perceptions of the Quality of Health Services .115 Willingness to Pay for Improved Health Services. 116 Supply of Health Services .116 Policy Framework . 117 Organization and Structure of the Health Care System. 118 Problems of Personnel, Drugs, Management, and Constraints on the Private Sector . 121 iv  Government and Donor Spending on Health Services ........ ....................... 124 Recurrent Expenditures . ............................................... 124 Unit Costs ........................................................ 127 Summary of Spending Patterns ............. ............................ 127 Total Health Spending and Distribution of Benefits ........ ........................ 128 Estimate of Total Spending on Health .......... ........................... 128 Distribution of Public Subsidies ............ ............................. 128 Conclusion . ........................................................ 130 6 Family Planning ........................................................ 133 Outcomes and Utilization ..................................................... 133 Fertility ........................................................ 134 Health and Fertility .................................................... 134 Utilization of Family Planning, Antenatal, and Obstetric Services ..... .......... 135 Household Expenditures on Family Planning, Antenatal and Obstetric Services .... ...... 138 Demand for Family Planning Services .............. ............................. 139 Knowledge and Attitudes . ............................................. 139 Determinants of Contraceptive Use ........... ............................ 140 Demand for Children . ................................................. 141 Supply of Family Planning Services . ............................................. 141 Policy Framework .................................................... 141 Organization and Structure of Family Planning Services ...... ................ 142 Problems of Quality and Implementation ........ .......................... 144 Public Expenditures on and Unit costs of Family Planning ....... .................... 145 Distribution of Public Spending . ............................................... 147 Conclusion . ........................................................ 148 7 Nutrition . ........................................................ 149 Outcomes . ........................................................ 149 Child Nutritional Status . ............................................... 149 Maternal Nutritional Status . ............................................. 152 Household Nutrient Inputs .................................................... 152 Food Consumption and Energy Intakes ......... ........................... 152 Breastfeeding ........................................................ 153 Supplementary Feeding of Infants ........... ........................... 154 Household Expenditures on Food . ............................................... 155 Nutrition-Related Government Actions ............. ............................. 156 Food Production and Availability ........... ............................. 156 Nutrition Policy ...................................................... 157 Current Nutrition Interventions ............. ............................. 157 Conclusion . ......................................................... 159 8 Water and Sanitation ....................................................... 160 Utilization of Improved Water Sources and Sanitation Facilities ...... ................ 160 Household Water Supply . ............................................... 160 Sanitation ........................................................ 162 Household Expenditures on Water . ............................................. 164 Demand for Improved Water Sources and Sanitation Facilities ...... .................. 166 v  Willingness to Pay for Improved Water Supplies .............. ............ 166 Household Demand for Improved Sanitation Services .......... .............. 168 Policy Framework for the Supply of Water and Sanitation Services ........ ............ 168 Expenditures in the Water and Sanitation Sectors ............... ................... 172 Government Expenditures .............................................. 172 Donor Expenditures ................................................... 172 Unit Costs .................................................... 172 Cost Recovery and Water and Sanitation Tariffs ....... ..................... 173 Distribution of Benefits .................................................... 175 Implementing a Demand-Side Planning Strategy: Options and First Steps ..... ......... 176 9 The Social Sectors in Zanzibar ................................................ 180 Education ............................................................... 180 Outcomes and Enrollment . ............................................. 180 Household Expenditures on Education ........... ........................ 183 Structure of the Education System ............. .......................... 184 Cost-Sharing in the Education System ........... ......................... 186 Health .................................................................... 186 Outcomes . .................................................... 186 Utilization . .......................................................... 187 Household Expenditures on Health Care .......... ......................... 188 Structure of the Health Care System ............. ......................... 189 Family Planning . .................................................... 190 Nutrition .................................................................. 193 Water and Sanitation ..................................................... 194 Utilization of Improved Water Sources and Sanitation Facilities ..... ........... 194 Water Tariffs .................................................... 195 Conclusion . ...................................................... 195 Bibliography .197 vi  TABLES S.1: Interactions among Social Sector Investments and Human Capital Outcomes ..... ............ 6 S.2: Annual Household Expenditures in 1994/95 (Cash and Consumed Production) in Tanzania . ................................................................. 7 S.3: Annual Household Expenditures (Cash and Consumed Production) in Tanzania, 1993 . .............................................................. 7 S.4: Total Government Budget Estimates, 1994/95 ......................................... 8 S.5: Total Expenditures on the Social Sectors from All Sources, 1994/95 ...................... 11 S.6: Distribution of Benefits of Social Sector Expenditures, by Sector, Recurrent and Development Budgets by Expenditure Quintile, FY93/94 ........................................ 12 S.7: Estimated Household Expenditures per Student Enrolled in Primary School, 1995 .... ....... 14 S.8: Households' Priorities and Perceptions of Government Primary Schools ..... .............. 15 S.9: Average Household Expenditures per Student Enrolled in Secondary School per Year, by Expenditure Quintile, 1994/95 .................................................. 17 S.10: Eastern Africa Regional Comparison of Burden of Disease, 1993 ........ .. .............. 19 S.11: Purchasing Power Parity Adjusted Per Capita Health Care Expenditures, by Type of Intervention .............. ......................................... 19 S.12: Government Health Care Expenditures, by Type of Intervention ................ .1 l9 S.13: Simulated Impact of Reallocation and Spending Increase on Under-Five Mortality Rate . .............................................................. 20 S.14: First Source of Curative Outpatient Care, by Expenditure Quintile ...................... . 21 S.15: Use of Inpatient Services, by Welfare Level ........ ............. . 21 S.16: Expenditures per Illness Episode Among Those Seeking Care, by Residence in 1994/95 . ................................................................. 21 S.17: Expenditures per Visit, by Type of Provider and Expenditure Quintile, 1994/95 .... ........ 22 S.18: Households' Priorities and Perceptions of the Local Government Health Center or Dispensary . ........................................................ 23 S.19: Household Water Sources, by Residence ................... ........................ 24 S.20: Type of Sanitation Facility, by Residence .................... ....................... 25 S.21: Characteristics of Women between 15 and 45 Responding to Contingent Valuation Question of Child Spacing ............................................. 28 S.22: Constraints on the Supply of Social Services .................. ...................... 30 1.1: Compound Rates of Growth and per Capita Income in 2050 ............................. 39 1.2: Returns to Investment in Education, by Level and Region, Latest Year ..................... 40 1.3: Interactions among Social Sector Investments and Human Capital Outcomes ..... ........... 43 1.4: Fertility and Education in Tanzania ................................................. 44 2.1: Expansion of the Health Sector: 1969, 1978, and 1992 ................................. 47 2.2: Expansion of the Education Sector: 1971, 1981, and 1991 .............................. 47 2.3: Expansion of the Water Sector: 1971, 1980, and 1991 ................................. 48 2.4: Source of Funds for Rural and Urban Councils ....................................... 53 2.5: Total Government Expenditures, 1994/95 ............................................ 55 2.6: Trends in Social Sector Spending: 1990-93 .......................................... 57 2.7: Aggregation of Public Social Expenditure Incidence, 1993 .............................. 58 3.1: Demographic Characteristics of the Household, by Expenditure Level ..................... 60 3.2: Ownership of Some Assets by Gender Among Population 15 Years and Older .... ........... 61 3.3: Annual Household Expenditures (Cash and Consumed Production), 1994/95 ..... ........... 63 4.1: Gross Enrollment Rates, 1993 ..... .............................................. . 66 vii  4.2: Primary School Enrollment Rates, by Expenditure Quintile, Gender, and Location, 1993 ...... 66 4.3: Incidence of Late Enrollment Among Current Primary School Enrollees ..... .............. 67 4.4: Gross Enrollment Rates, by Grade ................................................. 67 4.5: Secondary-school Enrollment Rates, by Expenditure Quintile, Gender, and Location, 1993 .... 69 4.6: Increasing Enrollment Rates Without Altering Unit Costs ............................... 73 4.7: Increasing Enrollments Through Reallocation ........................................ 74 4.8: Share of Total Expenditures on Education, by Expenditure Quintile ....................... 75 4.9: Total Household Expenditures on Education per Year, by Households with at Least One Child in School ............................................... 75 4.10: Total Annual Household Expenditures on Education by Households with at Least One Child in School, by Gender ...................................... 76 4.11: Government and Household Expenditures per Student Enrolled in Primary School .... ... 76 4.12: Government and Household Expenditures per Student Enrolled in Secondary School per Year, by Expenditure Level, 1994195 .................................... 77 4.13: Average Household Expenditure per Student Enrolled in Government and Private Secondary Schools, 1994/95 . ............................................................. 77 4.14: Ratio of Expenditures on Secondary to Primary School, by Expenditure Level ..... ......... 78 4.15: Household Expenditures on Education per Student in Households with at Least One Child in School, by Gender, 1994/95 .................................................... 78 4.16: Children's Weekly Time Use in Tanzania, by Gender and Age Cohort, 1993 ..... .......... 79 4.17: Ratings of Government Primary School Quality ...................................... 80 4.18: Expansion of Private Schools .................................................... 84 4.19: Profile of the Average Primary School ............................................. 85 4.20: Qualifications of Teachers in Government and Private Secondary Schools ..... ............ 89 4.21: Distribution of Recurrent and Development Expenditures Among Education Sector Programs, 1994/95 .................................................................... 96 4.22: Input Use in Education Sector Programs, 1994/95 .................................... 97 4.23: Expenditures for Sample Government and Private Secondary Schools ..... ............... 98 4.24: Private and Social Annual Rates of Return to Education and Training (1990/91) .... ....... 102 4.25: Distribution of the Benefits of 1993/94 Public Spending on Education, by Expenditure Quintile .............. ........................................ 104 5.1: Infant Mortality per 1,000 Live Births, 1992 ........................................ 109 5.2: Total Deaths and Life Years Lost, by Disease ....................................... 110 5.3: Use of Selected Health Services, by Expenditure Level ................................ 112 5.4: Population Sick and Seeking Care, by Sex .......................................... 112 5.5: First Source of Curative Outpatient Care, by Expenditure Level ......................... 113 5.6: Use of Inpatient Services, by Expenditure Level ..................................... 113 5.7: Share of Total Household Expenditures on Health, by Residence and Expenditure Quintile .................................................................... 114 5.8: Total Expenditures per Illness Episode Among Those Seeking Care, by Residence in 1994/95 .............. ........................................ 114 5.9: Distribution of Health Expenditures by Those Who Sought Care, by Expenditure Quintile .............. ........................................ 114 5.10: Expenditures per Visit, by Type of Provider and Expenditure Quintile 1994/95 .... ....... 115 5.11: Ratings of Government Health Facility Quality ..................................... 115 5.12: Health Facilities, by Ownership, 1992 ............................................ 120 5.13: Distribution of Recurrent and Development Expenditures Among Health Sector Programs, 1994/95 ........................................ 125 viii  5.14: Input Use in Health Sector Programs, 1994/95 ...................................... 126 5.15: Estimated Cost Per Outpatient Visit, 1993 ................... I ..................... 127 5.16: Distribution of the Benefits of Public Spending on Health, by Expenditure Quintile .... .... 129 6.1: Total Fertility Rate in Sub-Saharan Africa, 1992 ..................................... 134 6.2: Fertility by Zone of Residence, 1991 .............................................. 134 6.3: Current Use of Contraception by Age, 1994 ......................................... 136 6.4: Current Use of Contraceptives by Region, 1991 ...................................... 137 6.5: Share of Health Facilities Providing Mother-and-Child Health and/or Family Planning Services, 1993 ............. ......................................... 143 6.6: Availability of Family Planning Services in Government and Private Hospitals and Dispensaries .............................................. 144 6.7: Illustrative Tally of Expenditures for the National Family Planning Program, by Source, 1992 . ........................................................... 146 6.8: Share of Those Obtaining Family Planning Services and/or Supplies from Hospitals, by Expenditure Quintile ............. ......................................... 147 7.1: Nutritional Status by Demographic Characteristics ................................... 149 7.2: Indicators of Undemutrition in Sub-Saharan Africa, 1980-1992 ......................... 150 7.3: Percent Distribution of Living Children by Breastfeeding Status, According to Child's Age in Months, 1991/92 . ........................................................... 154 7.4: Duration of Breastfeeding by Residence and Mothers' Education ........................ 154 7.5: Share of Individual Food Groups in Total Food Expenditure, by Expenditure Quintile .... ... 155 8.1: Primary Household Water Sources, by Residence .................................... 160 8.2: Household Water Sources, by Expenditure Quintile ................................... 161 8.3: Type of Sanitation Facility, by Residence ........................................... 163 8.4: Annual Household Expenditures for Water, by Expenditure Quintile ..................... 164 8.5: Share of Households Collecting Water Outside the Home and Estimates of the Time Spent Collecting Water . ........................................................... 165 8.6: Foreign Assistance to Water and Sanitation, 1989-1995 ............................... 172 8.7: Indicative Cost Data for Externally Supported Rural Water Supply Projects in Tanzania ...... 173 8.8: Charges for Water in Urban Areas, 1980-1993 ....................................... 174 8.9: Charges for Sewage Services in Urban Areas, 1994 ................................... 174 8.10: Distribution of Subsidies in the Water Sector, by Expenditure Quintile ..... ............. 175 8.11: Distribution of Public Subsidies by Expenditure Quintile, 1993 ........................ 175 9.1: Infant and Child Mortality Rates, 1994 ............................................. 187 9.2: Expenditures per Illness Episode, 1993 ............................................. 189 9.3: Indicators of Undernutrition, 1991/92 .............................................. 193 9.4: Household Water Source by Residence ............................................. 194 9.5: Household Water Source, by Expenditure Quintile ................................... 194 9.6: Type of Sanitation Facility, by Residence ........................................... 195 FIGURES S.1: Tanzania's Record of Growth Since 1980 ............................................. 2 S.2: Estimated Consumption per Capita Compared to GDP Estimate for 1993 .................... 3 S.3: Reversal in Education Enrollment and Stagnation in Health Indicators ...................... 5 S.4: Growth in Civil Service Personnel in Health and Education .............................. 9 S.5: Components of Per Student Primary Expenditures ..................................... 14 ix  S.6: Willingness to Pay All Costs, One Year, for Primary School ............................. 16 S.7: Components of Expenditures per Visit ............... ............................... 22 S.8: Willingness to Pay for Health Center/Dispensary Visit Including All Costs ..... ......... 23 S.9: Willingness to Pay for One Visit, Health Center or Dispensary, by Decile and Residence ....................................................... 24 S.10: Willingness to Pay, or Accept a Subsidy, for Spacing the Next Child an Additional Year ........................................................... 28 1.1: Tanzania's Record of Growth Since 1980 ............. ............................... 38 1.2: Increases in Literacy in Selected Countries, 1850-1980 ................................. 42 BOXES 2.1: Principles of the Social Sector Strategy . ............................................ 51 2.2: Hiring and Firing of School Teachers . .............................................. 52 2.3: Transfer of Funds ........................................................... 54 2.4: How Much is Enough? Underfunding in the Social Sectors . ............................. 56 3.1: Gender Issues in Tanzania .... ................................................... 61 4.1: Factors Influencing Primary School Enrollment .......... ............................. 69 4.2: Selection to Secondary Schools . ................................................... 70 4.3: Girls' Secondary Education Support Program .......... .............................. 71 4.4: Features of the Tanzanian Educational System .......... .............................. 81 4.5: Physical Access to Schools ........................................................ 82 4.6: Examination System ............................................................ 85 4.7: Structure of the Teaching Workforce ............... ................................ 86 4.8: Community Education Fund . ..................................................... 88 4.9: Availability of Textbooks in Primary Schools .......... .............................. 94 4.10: Uses of Funds for Sample of Government and Private Schools .......................... 97 4.11: Recent Donor Initiatives in the Education Sector .................................... 100 5.1: Features of the Tanzanian Health System ............ ............................... 117 5.2: Community Health Fund ........................................................ 119 5.3: Recent Donor Initiatives in the Health Sector .......... .............................. 132 6.1: Changes in Modern Contraceptive Use in the Early 1990s .............................. 136 6.2: Women In Need of Family Planning Services .......... .............................. 137 7.1: Child Support, Protection, and Development: Structure and Implementation ..... .......... 158 8.1: Households' Willingness to Pay for Water ........... ............................... 167 8.2: Agencies Involved in the Water and Sanitation Sector ................................. 171 8.3: The East Kilimanjaro Water Supply Project: Transferring Ownership and Control from Central Government to Community Ownership ......................... 177 9.1: School Drop-Outs Compared to Parents' Education and Occupation ...... ................ 182 9.2: How Do Zanzibari Parents Perceive the Quality of Education? .......................... 183 9.3: How Do Zanzibaris Perceive the Quality of the Health System? ......................... 190 x  Acronyms and Abbreviations ADB African Development Bank AIDS Acquired Immune Deficiency Syndrome BMI Body Mass Index CBD Community-Based Distribution CIDA Canadian Intemational Development Agency CMS Central Medical Stores CPI Consumer Price Index CPR Contraceptive Prevalence Rate CSPD Child Support, Protection and Development DEO District Education Officer DHMT District Health Management Team DM Deutschmark DMO District Medical Officer DSM Dar es Salaam DWD Department of Water Development EDP Essential Drugs Program EKWSP East Kilimanjaro Water Supply Project EPI Expanded Program on Immunization ESR Education for Self-Reliance FAO Food and Agriculture Organization FAST Focused Area Studies Technique FINNIDA Finland International Aid Organization FY Fiscal Year GDP Gross Domestic Product GTZ German Technical Assistance Agency HC Health Center HESAW Health through Sanitation and Water HH Household HIV Human Immunodeficiency Virus HMIS Health Management Information System HRDS Tanzania Human Resources Development Survey ICD Institute for Curriculum Development IDD Iodine Deficiency Disease IEC Information, Education and Communication IFC International Finance Corporation IMR Infant Mortality Rate IUD Intrauterine Device JICA Japan International Cooperation Agency JNSP Joint Nutrition Support Project KAP Knowledge, Attitudes and Practices xi  Acronyms and Abbreviations (continued) MCH/FP Maternal and Child Health/Family planning MOE Ministry of Education (Zanzibar) MOEC Ministry of Education and Culture MSTHE Ministry of Science, Technology and Higher Education NCHS National Center for Health Statistics NGO Nongovernmental Organization NORAD Norwegian Agency for International Development NUWA National Urban Water Authority OC Oral Contraceptive ODA Overseas Development Agency PCM Protein-Calorie Malnutrition PER Public Expenditure Review PHCC Primary Health Care Center PHCU Primary Health Care Unit PMO Prime Minister's Office PSLE Primary School Leaving Examination RDD Regional Development Director REO Regional Education Officer RHMT Regional Health Management Team RMO Regional Medical Officer RWMP Regional Water Master Plan SIDA Swedish International Development Authority SSR Social Sector Review TB Tuberculosis TBA Traditional Birth Attendant TDHS Tanzania Demographic and Health Survey TES Tanzania Elimu Supplies TFNC Tanzania Food and Nutrition Center TFR Total Fertility Rate TIE Tanzania Institute of Education TKAP Tanzania Knowledge, Attitude and Practice Tsh Tanzanian Shilling TTC Teacher Training College UK United Kingdom UMATI National Family Planning Organization UN United Nations UNDP United Nations Development Program UNFPA United Nations Population Fund UPE Universal Primary Education US$ United States Dollar xii  Acronyms and Abbreviations (continued) USAID United States Agency for International Development VAD Vitamin A Deficiency WDR World Development Report WHO World Health Organization WTP Willingness to Pay xiii  ABSTRACT This report was prepared by The World Bank cooperatively with the Government of Tanzania to provide a comprehensive review of Tanzania's social sector activities, including education, health, water and sanitation, nutrition, and family planning. It has contributed materially to The World Bank's social sector policy dialogue with the Government of Tanzania, precipitated the restructuring of the World Bank's portfolio in health and education, and led to improved cooperation in the social sectors both within the donor community and between donors and the Government. The report is based on the Human Resources Development Survey of 1993/94, qualitative surveys, community case studies, and other population-based surveys, such as the Tanzania Demographic and Health Survey 1991/92 and the Tanzania Knowledge, Attitudes and Practices Survey 1994. The report describes trends in the social sectors and analyzes the factors that influence these trends, including the performance of the economy, government social sector spending and policy, household behavior and incomes, and demographic trends. Findings from the surveys and case studies are synthesized to provide a comprehensive picture of the demand for and supply of education, health, water and sanitation, nutrition, and family planning services. Tanzania's social indicators remain among the poorest in the world. The report reviews these trends, presents issues, assesses past government policies in the social sectors, and provided both the data and framework for the development and implementation of the Government's Social Sector Strategy. This document is part of a series of analytical pieces on poverty and the social sectors in Tanzania, including Tanzania: A Poverty Profile, Report No. 12298-TA (1993); Tanzania: The Challenge of Reforms: Growth, Incomes, and Welfare, Report No. 14982-TA (1996); and Technical and Higher Education and Training in Tanzania: Investments, Returns, and Future Opportunities, Report No. 15327-TA (1997). xiv  ACKNOWLEDGMENTS This report was prepared by a team under the guidance of Charles C. Griffin. Members included Anil Deolalikar, Nicholas Kessy, Faustin Mukyanuzi, Issa M. Omari, Perran Penrose, Liberatus L. Shirima and Dale Whittington (consultants); Michael Kiernan and Jens Pedersen (DANIDA); Larry Forgy (USAID); Margaret Grosh; Emmanuel Malangalila; R. Paul Shaw; Sverrir Sigurdsson; Maria Luisa Ferreira; Young Hoy Kimaro; Ruth E. Levine; and Lynn Tsoflias. Amit Dar and Ruth Levine assembled and analyzed most of the information on post-secondary education. Lynn Tsoflias took primary responsibility for assembling information on the social sectors in Zanzibar. Tamara Fox assisted with the collection and cataloguing of background materials. The authors are grateful for the many background papers that were prepared for this report. The papers are listed in the bibliography. The authors are also grateful for the thoughtful comments on earlier drafts. In particular, we thank Elizabeth Morris-Hughes, Mark Blackden and Chitra Bhanu for a thorough gender review of the previous draft. Ruth Levine took primary responsibility to pull the work of many different authors together into a cohesive report. R. Marisol Ravicz and Andrew Follmer updated the budget numbers for this publication. Oscar Picazo, under Ruth Kagia's direction, pushed the report to production, with final editing by Suzanne Gnaegy. The report was prepared under the general supervision of Jacob van Lutsenburg Maas, Division Chief, AF2PH, and James W. Adams, Director for Tanzania and Uganda. Motoo Konishi and Ronald Brigish, Resident Representatives, provided support throughout the process of developing this report. The peer reviewers were Jeffrey Hammer (EAPVP) and William McGreevey (PHN), and the Lead Economist was Shahid Yusuf. The household survey data used for this report were collected as part of a collaborative project of the World Bank with the University of Dar es Salaam and the President's Office Planning Commission, partially financed by the Government of the United Kingdom. The survey forms, training materials, and data from the Human Resources Development Survey 1993/94 are available in electronic form at The World Bank's web site, http://www.worldbank.org/html/prdph/lsms/guide/select.html. The authors would like to acknowledge, with thanks, the support provided by Yordi Seium, Lourdes Cuadro-Meliotes, and Helen Taddese at World Bank Headquarters, and Betty Sakaya, Maryanne Mwakangale, Rosalie Ferrao, Lorraine James and Mercy Sabai at the World Bank Resident Mission, Dar es Salaam. The authors would also like to express gratitude for the leadership provided by the President's Office Planning Commission in overseeing this work. In particular, we would like to thank F. Kazaura, Principal Secretary; J. Kipokola, Deputy Principal Secretary; and J. Zayumba, Director, Social Services. The Steering Committee and Technical Subcommittee provided guidance at all stages. Members included: J. Kipokola, M. Konishi, A. Chiduo, P. Hiza, H. Kamote, E. Malangalila, R. Man'genya, F. Mrisho, L.K. Msaki, M. Mvungi, H.K. Mwenisongole, F. Ndaba, G. Upunda, and J. Zayumba. xv  PREFACE The core material for the report was drawn from the 1994 Human Resources Development Survey and supplemented with background reviews of specific sector issues. The report discusses the situation and trends in the social sector and identifies ways that the government, the private sector, and households can use their respective resources most effectively. Chapter 1 provides the background and objectives of the social sector review. Chapter 2 traces the evolution of the social sector in Tanzania and the issues engendered by different policy regimes since the country gained independence. Chapter 3 presents the characteristics of the Tanzanian household; from this characterization, the report draws innovative household-focused, demand-driven policy recommendations. The succeeding chapters review the performance and issues in specific subsectors (Chapter 4 on education, 5 on health, 6 on family planning, 7 on nutrition, and 8 on water and sanitation) and draws conclusions and appropriate responses to deal with these issues. Chapter 9 reviews these same sectors for Zanzibar. The enthusiasm by which the results of this report have been embraced by the government and its partners is demonstrated by the quick adoption of its prescriptions. In education, two recent initiatives - the community education fund and the scholarship for girls in secondary schools - illustrate the way the study's recommendations are being implemented. Community Education Fund (CEF) - This is a program designed to help communities improve their primary schools. A participating school community conducts an annual fund-raising campaign and deposits the resources generated in the bank. The government provides a matching grant to the community-generated resources on a ratio of 1:1 or 1:1.5 for poorer cdmmunities. The community, through a village committee and the schoolteacher, plans and executes the use of these combined resources to improve school instruction, based on an agreed list of allowable expenditures. Under the roll-out program, by 1999 a total of 357 schools are expected to have established functioning CEFs. Girls Secondary Education Support (GSES) Program - GSES is a pilot program which provides bursaries for selected secondary-school girls. It aims to increase the enrollment rates of currently under- represented girls in secondary schools by targeting bright students coming from poor families who would otherwise not be able to continue their studies beyond Standard VII. Girls' education is an important investment in Tanzania, as this Social Sector Review shows, for it has important linkages in the improvement of household income, reduction in fertility, and improvement in women's health status. The GSES pilot sites now cover nine districts in six regions. More than a thousand girls are currently enrolled, with the first batch expected to graduate in December 1999. Since the results of the Social Sector Review were disseminated, the Ministry of Health has revitalized its fee-based cost sharing program, pretested the concept of community health funds, initiated a drug capitalization program for hospitals, reexamined its health policies in light of its burden of disease and cost-effectiveness review, and is launching a health insurance program for civil servants with the enabling Act of Parliament passed in April 1999. xvi  Drug Capitalization Program - Under this program, the Medical Stores Department procures drugs for government health facilities and supplies hospital phannacies with drugs that they manage as revolving funds based on patient fees. Twenty-three hospitals have been selected for the initial set of participating hospitals and capitalized with an initial stock of 2- to 4-months' supply of drugs. The fee and demand-driven elements of the program are intended to provide a more sustainable arrangement and to inculcate discipline among hospital managers who are now mandated to determine their actual pharmaceutical needs and not reduced to being passive recipients of supplies. National Health Insurance - Tanzania is in the process of launching a health insurance program that will cover initially 92,000 civil servants or 552,000 persons including dependents. The program will be funded by a mandatory contribution (3% of salary) from the employee and an equal contribution from the government as employer. The program was motivated by the need to increase financial resources to the health sector; to facilitate private financing of curative care; and by allowing members' choice of accredited providers, to introduce demand elements into the historically supply-driven and ill-funded government health facilities. The program is expected to assist the further development of the Tanzanian health care market. Cost Sharing Program - This user-fee program, formally launched in July 1993, underpins the other initiatives in the health sector. One of the key results of the Social Sector Review was the policy focus on user fees at the hospital level, to be accompanied by a program for dispensaries and clinics that strengthens community participation and moves towards a risk-sharing system (see Community Health Funds below). The Cost Sharing Program has since been expanded to cover all hospitals and in 1998, national policy also allowed dispensaries and health centers to charge modest fees. The program now contributes around 13% of total recurrent costs of the Ministry of Health budget, from a modest contribution of under 1% in 1993/94. There are anecdotal evidence suggesting a variety of efficiency improvements (more reliable supplies at hospitals, better facility upkeep, improved morale of providers, etc.), but there continues to be issues with respect to waivers and exemptions that need to be resolved and for the system to be improved so as not to prevent access to care by the truly indigent. Community Health Fund (CHF) - This is a village prepayment scheme for health services pretested in Igunga District in 1996; it has since attracted national attention for its ability to mobilize communities to generate revenues for health, identify service needs, and plan for their funding and delivery. Under a typical scheme, a participating household voluntarily prepays Tsh 5,000 a year for the benefit of consultations and drugs provided at the dispensary, health center, or the district hospital outpatient department. Non-CHF members pay the usual Tsh 500 per visit at the dispensary or health center and Tsh 1,000 at the district hospital outpatient department. Those without capacity to pay are protected through community-determined exemptions. As of November 1998, CHFs in five pretest districts already cover some 269,000 household members who have contributed Tsh 131 million in the form of prepayments and user fees. The government plans to replicate these schemes in other districts nationwide. At the macro level, the Government of Tanzania has also taken major initiatives supportive of the social sector prescriptions made in this study. The budget reform program, through the medium-term expenditure framework, seeks to improve the predictability, sustainability, allocative efficiency, transparency, and accountability of the budget (especially education and health) through a variety of public expenditure management mechanisms such as better sector programming, stricter enforcement of budgetary ceilings, expenditure tracking, and greater performance or results orientation. xvii  The Local Government Reform Program is intended to devolve education and health services to local councils. The implementation will be carried out in three phases beginning January 1999, each phase comprising one third (around 35) of all districts and urban councils. The devolution process is expected to involve a complex set of activities including organizational restructuring, staff retrenchments and hiring under the Civil Service Reform Program, design and use of new fiscal instruments to provide funds to devolved district services, installation of new financial and personnel management systems, and capacity building at central, regional and district levels. It certainly would take some time before the results of these and other policy reforms in the social sector would be fully realized. And there is always an initial cost involved in the transitory period. But this Social Sector Review has taken a close and candid look at the past and the past left much to be desired. The review has catalyzed the process of change. Perhaps more important than the document itself is the critical catalyzing role that the participatory process played in helping key actors in the government to review the situation, identify key bottlenecks in the system, and take action in innovative ways to implement key changes in the system of financing and providing social services. Now five years into the implementation process, much progress has been made, but the reforms themselves have opened new challenges. xviii  EXECUTIVE SUMMARY THE MESSAGES The two main messages of this review echo throughout the document: * Message 1: Basic education is fundamental to Tanzania's efforts to speed up economic growth and distribute the benefits of growth widely throughout the society. Basic education is a catalyst that increases the impact of other investments in health, nutrition, family planning, and water. * Message 2: Women in Tanzania are the primary agents of human capital investment. The litmus test of an effective social investment is whether it improves the ability of women to carry out this task. Put these two messages together, and three recommendations result: * Recommendation 1: A renewed commitment in Tanzania to getting every child through primary school and improving the quality of the schooling. * Recommendation 2: A new commitment to getting a large share of the next generation of girls through secondary school. * Recommendation 3: That other interventions in health, water, nutrition, and family planning be carefully targeted, and if possible, should support these two initiatives. No institution or country can do everything. Thus our recommendations are selective and focus on a few key commitments by the public sector to human capital investments whose impacts will ripple through the economy and improve the lives of every Tanzanian. In order to meet the challenges posed by these recommendations, the Government must solve the problems that inhibit the education sector and have existed in Tanzania for at least a decade. These include under-performance, inequitable subsidy patterns, inefficient production, lack of accountability, inadequate choices for users, poor resource mobilization, insufficient local authority and responsibility. In one way or another, the same problems are faced by the other sectors that are reviewed in this  2 Executive Summyar document. There is no need to wait to act in these other sectors. However, improving access to, and quality of, primary education, and advancing girls farther through the system, are the top priorities, and will require the concerted efforts of the best minds and managers in Tanzania. Simply allowing parents and communities to take greater responsibility for decisions will go far towards improving the management and efficacy of the limited resources that are available. THE SOCIAL SECTOR REVIEW The Tanzania Social Sector Review is a product of the Government of Tanzania and The World Bank, plus experts and other interested parties from Tanzania, NGOs, and other bilateral and mult.ilateral donors. The process of producing it has been led by a Government Steering Committee and Technical Committee that worked closely with Tanzanian consultants and Bank staff. Consultants and products were managed by The World Bank. However imperfect, this process has been uniquely successful in advancing the Government's policy process in the social sectors. In late November 1994, the Government prepared a Social Sector Strategy, a direct result of the background work underlying this Review. The Strategy was presented to donors at the Consultative Group Meeting in Paris in March 1995. In addition, Bank staff have been working with the government to develop pilot projects to test new approaches to human capital investment that follow directly from its Strategy. The main purpose of this report is to focus on social sector outcomes (as opposed to inputs to sector ministries) from a human capital investment perspective. Thus the Review is oriented to the demand for social services by households, where most decisions affecting the formation of human capital are made. Treating households as the main producers of human capital, rather than as passive consumers of social services provided by sector ministries, provides an opportunity to consider how inputs from social services combine to produce a healthy, knowledgeable, and skilled population. For comprehensive coverage, however, information on the supply of social services is also provided, and the report is organized Growth Rate in Real GDP along sectoral lines. The reader who 10 persists to the end of the document will find an exhaustive review of the 8 education, health, water, nutrition, Adjustment Begins Population Growth Rate and family planning sectors. 6 ECONOMIC GROWTH AND POVERTY 41 Economic Growth Since 1980 2 Figure S.l shows annual 0 N 1111 growth rates in GDP for the Tanzanian economy since 1980. -2 Since structural adjustment began in 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 1984, there has been steady positive Year growth in the economy, with a large Figure S.1: Tanzania's Record of Growth Since 1980 surge in the late 1980s and a period Source: World Bank Data  Executive Summary 3 of slower growth in the first half of the 1990s. If population growth of about 2.8 percent per year is taken into account, there was substantial improvement in per capita income in the last half of the 1980s, with much slower growth prevailing in the 1990s. A nationally representative survey conducted for this Review, the Tanzania Human Resource Development Survey of 1993/94 (HRDS), asked detailed questions about expenditures to estimate household welfare levels'. In addition to information on overall household spending, the survey provides estimates of use and spending patterns for health, education, family planning, and water/sanitation services in Tanzania. The survey is used throughout this report to describe household behavior. Estimates of household Dollars Per Capita expenditures from the survey suggest 52 538 that income in Tanzania is probably 500 substantially higher than is measured by national accounts. Figure S.2 400 compares estimated GDP per capita 289 from the national accounts (the left- 300 _ - most bar) with expenditure estimates 200 0 from the survey (the remaining bars). 144 ---- For the nation as a whole, cash 100 110 expenditures are about a third higher than GDP estimates. If nonmarket 0 _7_j _- thansGDpestimate s. i fc o GNP All Tanzania Rural Urban Dar es Salaam consumption IS included, consumption per capita is about Cash 1 All Sources double the estimate of GDP per capita2. The bars for rural, urban Figure S.2: Estimated Consumption per Capita Compared to GDP (other than Dar es Salaam), and Dar Estimate for 1993 es Salaam households show the Source: HRDS 1993194 much higher per capita expenditure levels and greater share of cash in expenditures in urban areas. Poverty in Tanzania The Tanzania Poverty Profile, based on household data from 1991 (it is therefore comparable to the HRDS survey results, not to the GDP estimates), shows that 51 percent of the population had incomes ' The survey covered approximately 5,000 households from the National Master Sample and was carried out during late 1993 and early 1994 by the Faculty of Economics at the University of Dar es Salaam, the Planning Commission, and the World Bank. It was financed by The World Bank, ODA, and the Govemrnment of Japan. 2 In each pair of bars from the survey, the left bar is cash expenditures, and the right bar is cash plus noncash expenditures, including consumption from own production. GDP estimates are adjusted for estimated own-consumption, so the right bar of each pair, which includes cash plus noncash expenditures, is probably more comparable to GDP estimates. As consumption expenditures are only one of the components of GDP (which also includes investment, government expenditures, and exports minus imports), there is little question that the national accounts substantially underestimate per capita GDP.  4 Executive Summary of less than $1 per day per person in 1991. Forty-two percent had incomes of less than $0.75 per day3. The Poverty Profile suggests that poverty in Tanzania is almost exclusively a rural phenomenon. 'While 59 percent of the population live in rural areas, about 85 percent of the below-$ 1-a-day poor, and 90 percent of the below-$0.75-a-day poor, live in the countryside. A comparison of the 1991 data with another survey from 1983, just before the first structural adjustment program, shows that the resumption of economic growth over the past decade has substantially improved the lot of the poor in Tanzania. In 1983, 65 percent of the population was living on less than $1 a day equivalent, compared to 51 percent in 1991, using prices in 1991 for both years. In terms of the lower poverty line of $0.75 per day, the percent of the population in poverty fell from 51 percent in 1983 to 42 percent in 1991. As a result, there are nearly 2 million fewer people living below $0.75 per day today than in 1983. If there had been no growth in the economy, population growth alone would have swelled the ranks of the poor from 11 million in 1983 to at least 14 million in 1991. Tanzania's success in poverty reduction through growth is an impressive achievement. In fact, the analysis probably understates the improvement that actually took place (Ferreira, 1994). There are three problems with this picture of general economic improvement: * Poverty in Tanzania remains primarily a problem of growth. To say the least, it is disturbing to have 50 percent of the population living on less than $1 per day, and the only way to change this situation is to accelerate the rate of economic growth far beyond current levels. * Growth has been accompanied by greater inequality. The really poor -the bottom 1 0 percent or so of the income distribution - appear to have fallen behind. They need special attention. Income inequality between cities and the countryside, and even within the countryside, has increased. These problems are probably related to the inequitable distribution of human capital and uneven access to markets, rather than to a maldistribution of other productive assets. * Tanzania missed a golden opportunity during this period of growth to do a much better job in the social sectors. Children who should be in primary school are not enrolled. The gross primary school enrollment rate should be well over 100 by now (it reached nearly 1 00 percent in 1979-80), but it has dropped to less than 70 percent4. The average age of entry to primary school is far too old, 9 years for girls and almost 10 years for boys in 1993. Tanzania's secondary-school enrollment rate is among the lowest in the entire world. The infant mortality rate has hardly changed over the last decade (Figure S.3). These poverty lines are based on income, but they are adjusted for purchasing power parity and family composition (children require less food than adults to sustain themselves, for example). Poverty lines based on consumption, a more reliable measure ol welfare, put 51 percent of the population below the equivalent of $1 per day and 36 percent below the equivalent of $0.75 per day (World Bank 1993, Tanzania: A Poverty Profile, Report No. 12298, December). 4 The gross primary enrollment rate could exceed 100 percent because it is the total number of children in primary school divided by the number of children in the normal primary school age groups.  Executive Summay 5 These problems cannot be solved by increased per capita Percent / Deaths Per 1000 Live Births income alone. But sustained growth 120 in the economy gives Tanzania some 100 breathing room to begin addressing . these problems in the social sectors. 80 --. ---- Solving them will add to the growth 60s potential of the economy and will - ----- contribute directly to improved 40 . - -- welfare and an improved 20 distribution of income. These concerns are the starting point for 0 -- ---- this review, and comprise the Primary Enrollment Infant Mortality principal reason for the 1982 1991 government's dedication to finding a new approach in social sector ProIisChilfdren of Primary School Age development. (World Bank Data) Figure S.3: Reversal in Education Enrollment and Stagnation in Health HUMAN CAPITAL AND ECONOMIC Indicators DEVELOPMENT For individuals and households, it is now clear that there are very high private rates of return to investments in schooling and health. The private internal rate of return to schooling in Sub-Saharan Africa is estimated to be 41 percent to primary, 27 percent to secondary, and 28 percent to higher education. For countries, the social rate of return to schooling in Sub-Saharan Africa is estimated to be 24, 18, and I percent to primary, secondary, and higher education, respectively. There are few investments that governments or parents could make that could compete with the high returns from primary and secondary schooling. Much of the success of the "East Asia Miracle" in achieving unprecedented rates of growth over the past three decades has been attributed to heavy investment in primary and secondary education by both households and governments. Those economies reduced poverty at rates of up to 7 percent per year growth in per capita GDP. They employed huge pools of previously underemployed labor and spread the benefits of growth widely throughout society. They allowed people to seek internationally competitive returns to these human capital investments by following export-led growth strategies. Beyond the direct economic benefits, investment in human capital also improves the welfare of individuals and the society as a whole. Even single inputs, such as education, can create several types of benefits. Table S. 1 illustrates what we know about these impacts. It shows social services down the left side and human capital-related outcomes across the column headings. Each cell shows the impacts of social services on outcomes. Only the stronger "cross effects" are shaded. Additional education, for example, increases the ability of couples to space their children optimally, which invariably results in lower birthrates. Education also contributes to improved health, better nutrition, and the ability to take advantage of better water and sewerage services to create more healthful living conditions.  6 Executive Summary Table S.1: Interactions among Social Sector Investments and Human Capital Outcomes INPUTS FROM SOCIAL _________ IMPACTS ON HUMAN CAPITAL FORMATION SERVICES Knowledge Family Size and Child Health Nutritional Healthful Living SERVICES________ Knowledge_______ Quality Status Status Conditions Education [ | |T[ m 11|E1111H1|| 1 1Th[ Family planning | + Health + + lJlI[I41 + Nutrition + lIlIl Water and Sanitation _ + Note: The strongest cross-sector effects are shaded. The two cells surrounded by heavy lines with no shading indicate high potential for carefully targeted interventions. All of the diagonal cells could be shaded, of course, but in this table only cross-effects are considered. Source: This is the authors' summary of the general findings of the empirical literature. Similar cross effects are generated by other social sector services. Family planning investments improve the health of mothers and children. Health investments improve nutritional status because of the link between a high disease burden and nutritional failure. This link works in the other direction as well: improved nutrition raises the body's resistance to disease. A healthier environment resulting from better water and sanitation services can have positive health impacts and are complementary to many other investments, such as education and nutrition. In addition, targeted health and nutrition interventions before and during the primary school years have positive impacts on educational outcomes. Four messages emerge from this framework. First, education is the key producer of cross benefits. It has large impacts across the board for the individual receiving the investment in terms of higher earning potential, cross effects in generating additional human capital investments, and improved welfare for the individual. Second, educationfor women has large external or "multiplier" benefits. Because women bear children and care for them in the crucial early years, but also because of women's traditional role as cultivators in Tanzania, investments in schooling for women have the largest impact in creating additional human capital and in improving living conditions. Third, targeted health, nutrition, and sanitation investments can have high cross-benefits, especially for school children. Fourth, health is the major "user" of cross benefits. Almost all of the other sectors have strong impacts on improved health outcomes because better health is the result of many different inputs. Human capital investments provide a stream of returns for their whole lives even if they do not own other property. In fact, human capital is often the only asset the poor and women own. It cannot be sold or stolen from them once they get it. For women, schooling and good health provide a stream of returns not only for themselves but also for their families, and especially for their young children. Because education has strong cross effects, it is possible to get large payoffs across the board by emphasizing primary and secondary education, especially education for girls. THE TANZANIAN HOUSEHOLD The production of human capital is managed by the household, especially by mothers, and the money the government spends in the social sectors is absorbed by households in pursuit of their goals. What do Tanzanian households look like? How are they using social services? How much are they spending to accomplish their objectives in human capital formation? How are they using government  Executive Summarv 7 subsidies? How do they view the social services available to them? In this section, we start with a few general facts about households. Table S.2: Annual Household Expenditures in 1994/95 (Cash and Household Expenditures Consumed Production) in Tanzania (Tsh) Consumption of Rural Urbana DSM All purchased and home-produced Expenditures 664,116 910,905 1,487,092 776,604 goods and services in Tanzania Food 76.1% 66.4% 61.1% . 7155% is shown in Table S.2 in Tsh and Health 1.7% 2.2% 2.3% 1 9a% in Table S.3 inUS dollars5. A Education 1.2% 2.0% 1.2% 1 4