Report No.: 46297-LS Lesotho Sharing Growth by Reducing Inequality and Vulnerability: Choices for Change A Poverty, Gender, and Social Assessment June 2010 Poverty Reduction and Economic Management Africa Region Document of the World Bank ISIC International Standard Industrial Classification KILM Key Indicators of the Labour Market Programme KoL Kingdom of Lesotho LDHS Lesotho Demographic Health Survey LEC Lesotho Electricity Company LENEPLW,HA Lesotho Network of People Living with HIV and AIDS LHWP Lesotho Highlands Water Project LRAP Livelihoods Recovery through Agriculture Programme LVAC Lesotho Vulnerability Assessment Committee MCA Millenium Challenge Account MCC Millennium Challenge Corporation MoE Ministry of Education (Lesotho) MOET Ministry of Education and Training MoGYSR Ministry of Gender, Youth, Sports and Recreation MOHSW Ministry of Health and Social Welfare mt Megaton NAC National AIDS Commission NGO Nongovernmental organization NMDS National Manpower Development Secretariat OLS Ordinary Least Square OSISA Open Society Initiative for Southern Africa ovc Orphans and vulnerable children PEP Post-exposure prophylaxis PEPFAR President’s Emergency Plan for AIDS Relief PIH Partners in Health PLWHA People Living with HIV and AIDS PMTCT Prevention of Mother-to-Child Transmission of HIV PPP Public-private partnership PPP Purchasing power parity PRS Poverty Reduction Strategy PRSP Poverty Reduction Strategy Paper PSI Population Services International SD Standard deviation? SNA System of National Accounts Std. Dev. Standard deviation TVET Technical and vocational education and training UNAIDS Joint United Nations Program on HIV/AIDS UNDP United Nations Development Programme UNGASS United Nations General Assembly Special Session on HIV and AIDS UNICEF United Nations Children’s Fund VCT Voluntary Counseling and Testing WFP World Food Programme WHO World Health Organization Vice President - Obiageli Ezekwesili Country Director - Ruth Kagia Sector Director - SudhirShetiy Task Manager - Louise Fox ABSTRACT L esotho began a structural economic transformation in the early 1990s. The transformation has brought higher, more secure incomes to households while the government succeeded in dramatically improving access to services such as education, health, water, and transportation. Yet today, Lesotho faces a number of serious development challenges, including a high rate of chronic poverty, entrenched income inequality, and-most troubling-ne of the highest HN/AIDS prevalence rates in the world. This report focuses on three main areas: livelihood patterns among Lesotho’s households and how these correlate with opportunity and exclusion; how the government could make access to public services and overall social development more equitable; and how the economic and social vulnerabilities of households, including HIV/AIDS, could be alleviated in order to reduce poverty. As of 2002/03, approximately 37 percent of households in Lesotho lived on less than $1 per day and about half of households lived below the national poverty line. Nearly one in four adults carries HIV. Inequality plays out in a variety of ways, from the much higher rate of child mortality among the poorest quintile of the population than among the richest to public spending on education that is dramatically skewed toward the richest quintile. In general, however, inadequacy of data on household demographics, assets, livelihoods, and access to social services, among other areas, compounds policies challenges in Lesotho. Data is collected too infrequently and is not sufficiently disaggregated to provide a clear picture of how economic and social policies and programs impact the people of Lesotho. From vulnerability and inequality perspective, several groups within Lesotho’s population are suffering from inadequate public resources. Particularly vulnerable groups include women (who have long had an important role in the economy of Lesotho but who are still severely under- recognized from a legal and cultural perspective), the large swath of the adult population battling HIV/AIDS, children orphaned by HIV/AIDS, and people who are chronically food insecure. As a result of uneven structural economic transformation across the country, the divide between rural areas (where 76 percent of the population resided as of the last household survey) and urban areas in Lesotho remains wide. Overall, people living in rural areas have lower incomes, higher rates of poverty, and less years of education than those in urban areas, and the rural population remains too dependent on subsistence agriculture and remittances. Efforts to restructure Lesotho’s economy have been successful in certain sectors, including manufacturing and natural resources extraction, but not successful in others, such as agriculture and services. Supporting these lagging sectors, along with ensuring that public policies and expenditures are pro-growth and pro-poor, would create needed employment and income opportunities for the population. Analysis in this report concludes that it is possible for Lesotho to reduce poverty and to continue its transition to an economically diverse middle-income country by undertaking three key strategies: continuing to develop the investment climate for labor-intensive production; implementing programs to support commercial agriculture and reduce land degradation in rural areas; and developing a strategy to support socioeconomic and geographic mobility of workers into higher-productivity sectors. Public funds should be better targeted towards assisting the poor to build human capital and manage risks. Other recommendations to improve the plight of the people of Lesotho include creating a more equitable and inclusive society, especially for women, and developing a more effective campaign against HIV/Aids. TABLE OF CONTENTS ACKNOWLEDGEMENT ............................................................................................................................ V 1 . SUMMARY AND CONCLUSIONS ................................................................................................. 1 A. B A C K G R O ................................................................................................................................ ~D 1 B. THELANDSCAPE OF POVERTY, MEQUALITY, AND WELFARE IN LESOTHO ....................................... 2 c. STRUCTURAL TRANSFORMATION, LIVELIHOODS AND INCOME POVERTY REDUCTION ..................... 4 D. PURSUING SHARED GROWTH THROUGH STRUCTURAL TRANSFORMATION: ...................................... 7 E. USING PUBLIC POLICY TO REDUCE POVERTY AND INEQUALITY ....................................................... 9 F. TOWARD A MORE EQUITABLE SOCIETY ......................................................................................... 14 G. FIGHTING HIV/AIDS AND REDUCING VULNERABILITY ................................................................... 16 H. PUBLIC SECTOR ACTIONS TO FIGHT HIV/AIDS AND REDUCE VULNERABILITY ................................ 20 2 . DIMENSIONS OF POVERTY IN LESOTHO .............................................................................. 23 A. INTRODUCTIONTO THE REPORT ................................................................................................... 23 B. POVERTYAND LIVING CONDITIONS ............................................................................................. 24 C. INEQUALITYAND GROWTH .......................................................................................................... 26 D. CAUSESOF POVERTY ................................................................................................................... 27 E. POVERTYTRENDS OVER TWE..................................................................................................... 30 F. CHARACTERISTICS OF THE POOR: TRENDS OVER T W E : ................................................................ 35 3 . IMPACT OF GROWTH ON POVERTY ....................................................................................... 50 A. STRUCTURAL CHANGE M LESOTHO: ENVIRONMENT FOR POVERTY REDUCTION, 1994-2005 ......50 B. S LESOTHO ..................................................................................... LABORMARKET ~ u T c ~ M E IN 53 C. LABORMOBILITY AND STRUCTURAL CHANGE .............................................................................. 57 D. GROWTH, EMPLOYMENT, AND POVERTY REDUCTION ................................................................... 61 4 . AGRICULTURE, FOOD, AND NUTRITION SECURITY ......................................................... 84 A. FOOD NATIONAL SECURITY ......................................................................................................... 85 B . HOUSEHOLD FOODSECURITY ...................................................................................................... 94 C . NUTRITIONAL OUTCOMES ............................................................................................................ 96 D . . POLICY OPTIONSFOR FUTURE FOOD SECURITY ........................................................................... 98 5 . RURAL ECONOMY, EMERGING LIVELIHOODS, AND SOCIAL DYNAMICS ...............106 A. INCOMES AND ASSETS OF RURAL HOUSEHOLDS .......................................................................... 106 B. SOCIALCHANGE ........................................................................................................................ 111 c. POLICIES AND INSTITUTIONAL ARRANGEMENTS ........................................................................ 113 6. PUBLIC SERVICE DELIVERY, GOVERNMENT EXPENDITURES AND POVERTY REDUCTION ................................................................................................................................. 123 A. GOVERNMENT EXPENDITURES ................................................................................................... 123 B. HEALTH CARE ............................................................................................................................ 124 C. EDUCATION................................................................................................................................ 130 D. INFRASTRUCTURE ...................................................................................................................... 135 7. THE CHALLENGE OF HIV/AIDS FOR POVERTY REDUCTION ....................................... 149 A. INTRODUCTION .......................................................................................................................... 149 B. W H O IS MFECTED WITH HIV/AIDS? ........................................................................................... 150 c. FACTORS DRIVING THE SPREAD OF HIV/AIDS .............................................................................. 151 D. ......................................................................................................................... CONSEQUENCES 153 E. RESPONDING TO THE CHALLENGE ............................................................................................... 156 F. IMPACT MITIGATION ................................................................................................................... 164 G. ASSESSMENT .............................................................................................................................. 166 1 8. LESOTHO'S LEGAL REFORM AND IMPLICATIONS FOR GENDER EQUALITY AND POVERTY REDUCTION ............................................................................................................. 183 A. GENDER-RESPONSIVEDEVELOPMENTS M LESOTHO'SLEGAL FRAMEWORK ............................. 184 B. BARRIERSTO THE IMPLEMENTATION AND ENFORCEMENT OF A GENDER RESPONSIVE LEGAL FRAMEWORK ...................................................................................................................... 186 C. NECESSARY IMPLEMENTATIONSTEPS ........................................................................................ 191 TECHNICAL APPENDIX 1 .................................................................................................................... 199 LISTOF BOXES Box 1.1: Strategic Options: Livelihoods And Income Poverty Reduction ..................................................... 8 Box 1.2: Strategic options toward a more equitable society ........................................................................ 16 Box 1.3: Strategic Options to Reduce Vulnerability .................................................................................... 21 Box 2.1: Main data sets used in the report ................................................................................................... 24 Box 2.2: Poverty and inequality, 2002/03 .................................................................................................... 27 Box 2.3: Key causes of poverty according to village interviews, 2002 ....................................................... 28 Box 2.4: Available data sources to track poverty over time .......................................................................... 31 Box 7.1 : Male circumcision as a part of the HIV prevention strategy ....................................................... 153 Box 7.2: A plethora of assistance needs Government coordination ........................................................... 158 Box 8.1: Customary and statutory law: How are their boundaries determined? ........................................ 187 Box 8.2: Impact of HIV/AIDS epidemic on women's customary rights to land ........................................ 188 Box 8.3: Internationaltreaties concerning women's rights ratified by Lesotho .......................................... 191 LISTOF FIGURES Figure 1.1: Demography of poverty . in Lesotho. most of the poor are adults .............................................. 3 Figure Declining importance of farming and remittance income for households. .................................. 5 1.2: Figure Net Enrollment by quintile. 2004 .............................................................................................. 11 1.3: Figure The op quintile by wealth receives 42 percent of public expenditures on education. while the 1.4: lowest quintile receives only 9 percent .................................................................................... 12 Figure 1.5: Service problems in public health clinics and hospitals, 2005 ................................................... 13 Figure 1.6: Adult mortality by residency and gender, 1975-2004............................................................... 18 Figure 2.1: Relationship between household characteristics and consumption per adult equiv ................... 30 Figure 3.1 : Growth in overall real GDP. but no growth in agriculture ........................................................ 51 Figure 3.2: Declining private consumption and remittances ........................................................................ 51 Figure 3.3: Education Determines Job Type ................................................................................................ 56 Figure 3.4: Men and women working as farmer and in wage jobs by age, 2002-03 (%) ........................... 58 Figure 3.5: Men are more likely than women to migrate internally to find work P A migrant by age cohort) ................................................................................................................................................. 61 Figure 3.6: Declining reliance on farming and nonlabor income, 1994-95 and 2002-03 ........................... 62 Figure 3.7: Household main source of income by region, 2002-03 ............................................................. 63 Figure 4.1: Cereal Production, 1968-2006 And 1980-2006 ........................................................................ 86 Figure 4.2: Key grains as share of cereal production, 1980 - 2006 (tons) ................................................... 86 Figure 4.3: Maize and wheat yields, 1980-2006 .......................................................................................... 87 Figure 4.4 Maize yields in Lesotho and neighboring South Africa, 2007 (tha) .......................................... 89 Figure 4.5: Maize monthly spot price, SAFEX, 2000-2007 (Wt) ............................................................... 93 Figure 4.6: Commercial imports in 2007-08 ................................................................................................ 94 Figure 4.7: Availability of cereals (maize, sorghum, wheat and rice) - total ('000 Tomes) and per capita (Kg/year) ,................................................................................................................................. 94 Figure 4.8: Percent food from own production, 2004 (%) ........................................................................... 95 Figure 5.1: Rural land distribution by consumption quintiles, 2002/03 ..................................................... 108 Figure 6.1 : Popular approval of GoL handling of policy issues, 2005 PA) ................................................ 124 Figure 6.2: High government health expenditures per capita, 2004 ........................................................... 126 Figure 6.3: Basotho seek treatment when having symptoms ..................................................................... 127 ii Figure 6.4: Regional comparison of ratio of poorest to richest quintile of % seeking treatment when .................................................... ems in public health clinics. 2005 (%) ............................... 129 Figure 6.6: International comparison ofpublic expenditure per pupil as % ofper capita GDP. 2005 ....... 131 Figure 6.7: Net secondary enrollment rate across Africa. 2005 ........... Figure 6.8: Number of electricity connections by LEC . 1972-2007 .... Figure 7.1: HIV prevalence by age and gender .......................................................................................... 151 Figure 7.2: Adult mortality by gender, 1975-2004. ........ ..................... 154 Figure 7.3: Adult mortality by age, 1975-2004 ...... Figure 7.4: Adult mortality by educational background, 1975-2004 ......................................................... 155 LISTOF TABLES Table 2.1 : Poverty and Inequality. 2002/03 ................................................................................................. 37 Table 2.2: Inequality in Sub-Saharan Africa ................................................................................................ 37 Table 2.3: Characteristics of population by consumption quintile, 2002/03 ................................................ 38 Table 2.4: International comparison of health indicators ............................................................................. 39 Table 2.5: Health outcomes across wealth, location and gender, 2004 ........................................................ 39 Table 2.6: HIV/AIDS prevalence and treatment: International comparison ................................................ 40 Table 2.7: Educational attendance rates across wealth, location, and gender, 2004 .................................... 40 Table 2.8: Reduction in poverty as function of inequality and growth ........................................................ 41 Table 2.9: Key causes of poverty from village interviews, 2002 ................................................................ 41 Table 2.10: Stock theft 2000/01-2003/04 ..................................................................................................... 41 Table 2.11 : Regressions of household characteristics and consumption per adult equivalent, 2002/03 .......42 Table 2.12: Description of regression variables: Urban, 2002103 ................................................................ 43 Table 2.13: Description of regression variables: Rural, 2002/03 .................................................................. 44 Table 2.14: Poverty by geographic region in 199495 and 2002/03 (%) ............................... Table 2.15: Poverty transitions over time (55) ..................................................................... Table 2.16: Educational characteristics of population by wealth quintiles and location, .............................. 46 Table 2.17: Education of heads of households and spouses, 1994-95 and 2002-03 (%) ............ Table 2.18: Health and nutritional outcomes, 1976-2007 ............................................................................. 47 Table 2.19: Poverty across demographics, 2002/03 .................................... ............................................. 47 Table 3.1 : Real GDP by Sectors ................................................................. ............................................. 67 Table 3.2: Activity by sex and age before and after reclassification PA)..................................................... 68 Table 3.3: Employment and unemployment by gender and location, 2002-03 ( ! A ) .................................... 69 Table 3.4: Household businesses by agricultural and nonagricultural, rural and urban (%) ........................ 69 Table 3.5: Distribution of salary workers across industries by sex and location .......................................... 70 Table 3.6a: Education level of the employed (%) ........................................................................................ 70 Table 3.7: Main source of income, 1994-95 and 2002-03 .......................................................................... 71 Table 3.8: Main source of income by quintile, 2002-03 ( 96 ) ...................................................................... 72 Table 3.9: Employment by sex, age, and location ........................................................................................ 73 Table 3.10: Employment by sex and age before and after reclassification (%) ............................................ 74 Table 3.1 1: Distribution of salary workers across sectors by age and sex (%) ............................................. 75 Table 3.12: Activity by sex, age, and location .............................................................................................. 76 Table 3.13: Migrants by age and sex, 2002/03 (% of domestic population) ................................................. 77 Table 3.14: Activity by rural and urban consumption quintiles ( 99) ............................................................. 78 Table 3.15: Migration destination and consumption quintiles (%) .................................... Table 3.16: Characteristics of households with and without textile wage workers (96) ............................... 79 Table 3.17: Textile workers across quintiles of consumption, 2002-03 ....................................................... 80 Table 3.18: Miners across quintiles of consumption, 2002-03 (%) ............................................................. 80 Table 3.19: Characteristics of households with and without miners PA)..................................................... 81 Table 4.1 : Agro-ecological conditions and farming systems ..................................................................... 102 Table 4.2: Potential average cost of government subsidy for input costs (2007108) on 180,000 ha summer crops (vS$) ............................................................................................................................ 103 Table 4.3: Calories from own production by wealth and location, 2004 ................................................... 103 Table 4.4: Nutritional outcomes, 1992-2007 ............................................................................................. 103 iii Table 4.5: Regional nutritional outcomes. 2002-06 .................................................................................. 103 Table 4.6: Lesotho Stunting Relative to GNI Compared to other SSA countries . Table 4.7: Anthropometric measures using old and tandards (Year 2004) .......... Table 4.8: Household Livelihood Sources. by distr ............................................................. 105 Table 5.1 : Rural household main source of income ion quintiles, 2002103 (%) ...................121 Table 5.2: Distribution of land by characteristics of head of household, 2002103 (%) 121 Table 5.3: Available labor in land-owning rural households, 2002-03 (%) ................................................ 121 Table 5.4: Characteristics of households by amount of available labor for working the land, 2002-03 ..... 122 Table 5.5: Households with farm implements, 2002-03 (%) ........ Table 5.6: Ownership of livestock across consumption quintiles, Table 5.7: Households with livestock, 1994/95 and 2002/03 (%) ....................... Table 5.8: Increase in widowed and never married 1994/95 to 200 Table 6.1 : Government current expenditure by ministry, 1999/00 Table 6.2: Approval rate of governments handling of political issues, 2005 ( A ) ...................................... 140 Table 6.3: Children with symptom in the last two weeks, 2004 (%) ................................. Table 6.4: Persons seeking treatment for illness, 2004 (%) .............................................. Table 6.5: Facilities visited for treatment of children, 2004 ( A ) ................................................................ 141 Table 6.6: Obstacles to access to health care: %thinking obstacle is a big problem, 2004 ....................... 142 Table 6.7: Distance to nearest health facility, 2004 ( A ) ............................................................................ 143 Table 6.8: Means of transportation usually used to access health care, 2004 ( A ) ..................................... 143 Table 6.9: Reason for not seeking treatment when ill, 2004 (%) ............................................................... 144 Table 6.10: Primary and secondary attendance ratios, 2004 (%) ...... ................................................... 145 Table 6.1 1: Trends in school characteristics, 1999-2006 (%) .................................................................... 145 Table 6.12: Survival rates from grade 1 to 5 and 7 for cohorts starting in 1999 and 2000 (%) .................. 146 Table 6.13: Repetition and dropout rates in primary by gender and wealth quintile (A) ............................ 146 Table 6.14: Main reason for men below age 24 for not attending school, 2004 (%) .................................. 146 Table 6.15: Benefit incidence of educational expenditures, 2004-05 (%) .................................................. 147 Table 6.16: Time to fetch water, 2004 ( A ) ................................................................................................. 147 Table 6.17: Access to water by source, 2004 P A )....................................................................................... 148 Table 6.18: Access to piped water and electricity decomposition into demand- and supply-side factors, 2002-03 ( 99) .......................................................................................................................... 148 Table 7.1: HIV prevalence by selected characteristics, 2004 (%) .............................................................. 172 Table 7.2: Determinants of HIV prevalence in Lesotho, 2004 ................................................................... 173 Table 7.3: Determinants of HIV prevalence, including an indicator for being a miner, 2004 ................... 174 Table 7.4: Determinants of HIV prevalence in Lesotho: Subgroup analysis, 2004 ................................... 175 Table 7.5: Percentage and characteristics of orphans among children age 0- 17, 2004 .............................. 176 Table 7.6: Determinants of school enrollment and orphanhood, 2004 ....................................................... 177 Table 7.7: Determinant of years of schooling with focus on Orphanhood by age .................................... 178 Table 7.8: Determinants of school enrollment with focus on type of orphans by age, 2004 .................... 179 Table 7.9: Determinants of school enrollment with focus on living arrangement of orphans and age ......180 Table 7.10: Determinants of school enrollment with focus on living arrangement of orphans and age ..... 181 Table 7.1 1: Determinants of height for age for children younger than 6 years old ..................................... 182 iv ACKWOWLEDGEMENT T his report was prepared by a team led by Louise Fox (TTL, Lead Economist). The core Bank team included Thomas Pave Sohnesen (poverty profile and public seivice delivery and expenditures), Elena Bardasi (gender and impact of growth), Lynn Brown (agriculture and nutrition), Trond Vedeld (rural economy and livelihoods), Damien de Walque (HN/AIDS), Linda Nxuma and Tersesa Marciori (gender and legal issues). Research support was provided by Lucia Corn0 and Rachel Kline (consultant), while Edmund Motlatsi Motseki (AFMLS), Theresia Rasethunts'a (AFMLS) provided much appreciated support in Maseru. The team is thankful for comments and feedback that ,benefitted the report from: Gero Carletto, Shimwaayi Muntemba who also contributed to the chapter on HN/AIDS), Jacomina P. de Regt, Preeti Arora, Husam Abudagga, Aidan Mulkeen, Edgardo Favaro, Harold Alderman, and Oscar Picazo. Support was also provided by managers: Sudhir Shetty, John Panzer, and Ritva 'S. Reinikka. The team is grateful for the collaboration with Lesotho Bureau of Statistics and the Ministry of Finance, Planning and Economic Development which made this repot possible. V 1. SUMMARY AND CONCLUSIONS A. BACKGROUND 1.1. Lesotho is a small, mostly mountainous country of 1.9 million people completely surrounded by South Africa. Despite having a per capita income of nearly $1,000 (Lesotho is a lower-middle-income country), about 37 percent of households lived below $1 per day in 2002/03 and approximately half lived in poverty according to the national poverty line. The main reason for the high poverty rate is not slow economic growth but high inequality, which means that the benefits of growth do not easily reach the poor. This inequality has been a structural feature of Lesotho for decades; it is not the result of the recent growth pattern. Indeed, inequality may have declined over the past decade. Inequality is high in both urban and rural areas. 1.2. Structural change in Lesotho’s economy began more than 15 years ago and has involved a shift in the drivers of production and consumption from subsistence agriculture and receipt of remittances toward modern natural resource and industrial exports and the service sector. The main growth engine has been the industrial sector, followed by the public sector, which has used increased nontax revenues to expand access to public services. For many households, this structural shift has brought higher and more secure incomes. And thanks to an impressive effort by the government to expand access to basic social services, most of the population now has much better access to social and economic services such as education, health, water, and transport than they had 15 years ago. 1.3. It appears that a substantial minority of households, mostly in rural areas, have not benefitted much from economic growth and structural change. For some households, the loss of income from remittances combined with the decreasing natural resources base for agriculture - which contributes to volatile crop production, low rural incomes, and hunger and poor nutrition - has increased poverty, vulnerability, and exclusion. By far the most important livelihood and income risk, however, is the high prevalence of HIV/AIDS. At 26 percent of females and 19 percent of males, Lesotho’s adult HIV/AIDS prevalence is one of the highest in the world, and has caused life expectancy to plummet. Reversing the increase in morbidity and premature mortality is Lesotho’s most important development challenge. 1.4. Lesotho’s fight against poverty and exclusion is hampered by a lack of adequate- quality data or disaggregated data on household demographics, assets, livelihoods, earnings and income, shocks, such as deaths or other major life events access to services, and social indicators. As a result, the impact of growth and public policy is hard to identify. After extensive national consultations, Lesotho published its first national Poverty Reduction Strategy (PRS) in 2005, based on inconsistent and inconclusive data. In the years since that document was prepared, two new data sets have been released by the National Bureau of Statistics: the Household Budget Survey of 2002/03 (HBS) and the Demographic and Health Survey of 2004 (DHS). The analysis in this report has been undertaken using these new data. The objective is to inform the revision and updating of this strategy. But the new data suffer from the same quality and comparability issues as those used in constructing the PRS; as a result, it is still very difficult to identify trends. 1.5. In spite of the data weaknesses, policy makers can benefit from taking stock of past accomplishments and disappointments and learning from what worked and what did not. The analysis in this report focuses on the following key questions: 1 What are the livelihood patterns, or the types of the economic activities of households, in rural and urban Lesotho, and how do these correlate with opportunity and exclusion? Who are the winners and losers in Lesotho’s economy today, and what are the key challenges facing policy makers in raising the incomes of the poorest to ensure shared growth? 0 What are the causes and consequences of inequality and exclusion for shared growth and poverty reduction in Lesotho, especially gender exclusion? Why have some social indicators improved over the last decade, while others, such as infant mortality and child nutrition, remain stuck at incongruously high levels given Lesotho’s overall income level? Has past public spending benefited the poor, or has it increased inequality? What measures would be needed to ensure greater access to publicly- funded services by poor households and excluded groups? 0 What are the economic vulnerabilities of Basotho households and communities, and how can these be mitigated to reduce chronic poverty? How does widespread HIV infectionthe most important vulnerability today-affect poverty reduction, and what more can the government do to reverse the negative impact of this epidemic on poverty households and communities? 1.6. Our analysis concludes that Lesotho can continue its transformation to a prosperous, economically diverse middle-income country while reducing poverty substantially by continuing policies and investments that have worked in leading sectors such as manufacturing and natural resource extraction, while extending these successful approaches to lagging sectors such as agriculture and services. But for this approach to be effective, Lesotho needs to fight the HIV/AIDS epidemic aggressively and effectively. To create a more productive and equitable society, Lesotho needs to continue putting the proceeds from selling natural resources into programs which are both pro-growth and pro-poor. Data needs to be collected regularly and used in monitoring systems to ensure that expenditures reach the intended beneficiaries and that programs achieve the expected outcomes. These recommendations are fleshed out below. B. THELANDSCAPE OF POVERTY, INEQUALITY, AND WELFARE IN LESOTHO 1.7. Roughly half of Lesotho’s population lived below the national poverty line in 2002/03. Inequality in Lesotho is among the highest of all African countries. Though trends in poverty cannot be reliably estimated due to data limitations, data on assets suggests a sluggish reduction in income poverty since 1995. 1.8. Income poverty is related to location within Lesotho, with the highest incidence in rural areas. As of 2002/03, 76 percent of the population lived in rural areas, and poverty was 50 percent higher in rural areas than in urban areas.’ Within rural livelihood zones, poverty rates differ considerably, from a high of 63 percent in the Rural Foothill region to a low of 54 percent the Rural Senqu Valley. Only 32 percent of households in the Maseru region, by comparison, were poor in 2002/03. Inequality within both rural and urban areas is high, and both rich and poor can be found in all areas. Demographically, poor households are larger than richer households and as a group, children are slightly more likely to be poor but owing to Lesotho’s progression along the demographic transition, children do not represent the largest subset of the population ’Areas are coded as rural based on administrative criteria, not size or population density (the University of Lesotho, for example, is in an area classified as rural). This helps explain the high rural inequality picked up in the data. Lesotho is also divided statistically into six “ecological zones” which represent a useful geographical division for this report. 2 living in poverty: 40 percent of the poor were children in 2002/03, compared with 54 percent adults Figure 1.1: Demography of poverty - in Lesotho, most of the poor are adults Elder 64 and above 6% Children aged 0- 15 40% Adults aged 1 64 54% Source: HBS 2002/03 and authors’ calculations. 1.1 1 Poverty and location are both correlated with worse health and nutrition outcomes. For the poorest quintile of the population, the under-5 mortality rate was 114 in 2002/03, compared to 82 for the richest quintile. Variation in mortality rates between the Maseru region and other parts of the country were as large as the spread across quintiles. Anthropometric indicators such as wasting and stunting are also poor in rural areas, and are much worse for the poorest quintiles. Data on nutrition and health indicators for the poorest households have shown virtually no improvement in the last 15 years. 1.13 Livelihood patterns explain much of the poverty and inequality in Lesotho. There are major differences between urban and rural areas and between men and women in economic opportunities, livelihood patterns, and what households gain from economic activities. Urban households draw most of their income from wages (60 percent) and household businesses (13 percent), not farming, although about three in ten households and one in five labor force participants in urban areas are engaged in farming activities (either subsistence or commercial). Remittances were the main source of income for only 7 percent of urban households in 2002/03. Urban residents have more years of education, which allows them access to wage and salary jobs. The few wage workers in urban areas with little or no education are overwhelmingly male. 0 Rural livelihoods vary by ecological zone. What distinguishes rural Lesotho most of all is heterogeneity in livelihoods and the lack of correlation between household characteristics and outcomes. In the Rural Foothill region, the poorest region in 2002/03, 47 percent of households indicated that farming was their main source of income, 25 percent indicated wages, and 12 percent indicated remittances-the latter down from 30 percent in 1994/95. In the Rural Senqu River Valley, wage employment is as important as farming in providing household income, while 15 percent of households still count remittances as their main source of income. Incomes are 12 percent higher than in the Rural Foothill region, controlling for the characteristics of the household. This shows the 3 importance of diversification of sources of income for improving incomes in rural areas. The Rural Mountain region shows a contrasting story. Controlling for household characteristics, incomes are 20 percent higher than in the Rural Foothill region, yet this area has the highest dependence on agriculture as a source of income. Households that are dependent on agriculture are likely to be engaged in livestock production, which has been more profitable than grain production in recent years. 1.9. Multiple income sources characterize rural livelihoods. While 80 percent of rural households farm or raise livestock and 68 percent of adults in rural areas report working in farming, less than half of rural households draw most of their income from agriculture. Nearly one-quarter of rural households report their main source of income is wage employment, and 13 percent of rural households report that most of their income is in the form of remittances. Men and women in Lesotho are equally likely to farm, but men are more likely to have wage income or to migrate outside the country for work. 1.10. A diverse portfolio of activities creates higher incomes for rural households, while those without multiple activities tend to be poor. Although some rural households are able to make a very good living be concentrating on agriculture and livestock, these are exceptional, as the majority of poorest households are primarily farmers, with no outside income. Having multiple sources of income (e.g., farm and nonfarm) is important to the welfare of rural households, even those whose main activity is farming, because a second income provides funding for the farm enterprise and a hedge against the risk of farming. If the head of a rural household is a wage worker, the household has on average a one-third higher income than other rural households, controlling for education and other factors. Having earnings from abroad leads to a 20 percent higher income on average. 1.1 1. Education creates livelihood options, and households whose adult, members are poorly educated have a higher chance of ending up in poverty. This trend has strengthened since the 1990s, especially in urban areas. Put simply, richer households have more education and draw most of their income from wages, while poorer households farm. Farmers, in particular, have low levels of education-nly 35 percent have completed primary school, compared with 45 percent of those who work in other jobs. Education is especially important in urban areas, where completing primary education produces a 36 percent higher household income, on average, than having no education, and completing secondary education a 70 percent higher income. 1.12. Improvements in access to education and safe water are two areas of public sector success in Lesotho, and they have raised the welfare of rural households. Lesotho now has a highly literate population, the educational attainment of children has been improving, and net enrollment in primary school is among the highest in Sub-Saharan Africa. In 2004, surveys showed no difference in average enrollment between rural and urban areas, and only a 5 percentage-point difference in enrollment between the poorest and the richest quintiles. Access to safe water is virtually the same across quintiles-about 86 percent of people in Lesotho live in households with access to improved water sources, a high percentage for a Sub-Saharan African country. c. STRUCTURAL TRANSFORMATION, LIVELIHOODS AND INCOME POVERTY REDUCTION 1.13. The differences between livelihood patterns and poverty in rural and urban areas reflect the incomplete structural transformation of livelihoods. Since 1993, Lesotho has been engaged in a major structural transformation of its economy, and this has brought growth and (probably) poverty reduction. Boosted by strong growth in the construction, mining, and textile sectors, the industrial sector has grown at an average of more than 4 percent annually since 1993, becoming the largest sector in the economy in 2005. During the same period agriculture grew at anly half that rate, falling to 15 percent of GDP. Meanwhile, remittances as a share of GDP fell by half, as earnings from domestic employment became more important in supporting consumption (see the figure below for trends in the source of household income). Since 1990, an estimated 65,000 South African mining wage jobs were lost, while about 40,000 new jobs were created in the public sector and the textile sector. Many more jobs were created in other sectors. Figure 1.2: Declining importance of farming and remittance income for households, 1994/95 and 2002/03 -- , Farming Wages Household Pension Remittances Other Business I [H 1994/95 02002/03 Source: HBS 2002/03 and authors' calculations. Note: Figure shows main source of household income in 1994/95and 2002/03. 1.14. In urban areas, structural changes in the economy have brought new opportunities in the labor market-public and private sector wage jobs (which account for more than half of employment in urban areas) and household businesses. These urban households are the winners from economic growth. An unknown number of these households are migrants from rural areas, drawn by the better opportunities. In 2002103, 12-13 percent of the labor force under the age of 35 reported that they had migrated in seek of employment. 1.15. Opportunities in the urban labor market have been especially important for women. The textile sector has been instrumental in providing opportunities for young women migrating from rural areas. Women living in urban areas working in the textile sector tend to have higher incomes than those living in rural areas, and their children under the age of 14 are more likely to be in school. Yet the HBS data report that textile workers attached to a rural household (but working in the factory in an urban area) tend to be poorer than other textile workers, either because of lack of other earnings in the rural household or because these workers were not able to send back much income to the rural household in the form of remittances. 1.16. In rural areas, structural transformation has been limited. The lack of structural transformation in rural areas reflects the history of a rural economy supported from the outside, e.g., from mining remittances from South Africa. Despite poor soil and rainfall, a large portion of Lesotho's households are located in rural areas. Decades ago, these poor ecological conditions were less important, as the rural economy was supported almost entirely by remittances from the 5 South African mining sector. Since then, the importance of mining income in rural areas has declined, and there has been little growth in other sources to replace it. In some regions, wage income in the construction sector or mining was available, but in others, such as the Rural Mountain region, few households report income from this source. 1.17. Commercial agriculture has not replaced mining as a key income source in rural areas. Agricultural crop production, especially of grains, has been declining over the past ten years. The sector remains undeveloped - marketing and credit facilities are inadequate, modem technology is inaccessible and irrelevant given the lack of markets, and local institutions are weak. Overall, however, not much of crop output is marketed (less than 5 percent of rural households sold any grain in the 2006-07 season) because grain production is not commercially viable; costs of production, including milling and transportation, are high compared to those in South Africa. Farmers have not shifted into higher value crops owing to risk associated with lack of markets. Unlike other countries in the Eastern and Southern Africa, subsistence farming in Lesotho is associated with higher incomes: the poorest households grow less of their own food than do higher-income households. Other factors that normally are correlated with high rural income or wealth are not in rural Lesotho. For example, ownership of livestock is not concentrated in the richer quintiles, except for poultry, as richer households have about 20 percent more per head, on average, than the poorest households. Assets such as land ownership are not correlated with income. This result may reflect declining land fertility, but it also reflects an active sharecropping market for land. Households with older heads tend to rent out their land to younger households, and to households with female heads who do not have access to land. 1.18. Access to remittance income from abroad remains important in rural areas. Households receiving remittances have some of the highest incomes and best outcomes in the rural economy. Having a household member working abroad (often in mining) raises rural household income by as much as if the household had a working member with a primary education. In fact, 70 percent of households containing a person working in the mining sector are in the top two quintiles of welfare distribution, and only 8 percent of households with a miner rely on farming as their main source of income. Miners are more likely to have their children enrolled in school, especially those children aged 15-24 where the ability of the household to pay secondary school fees is an important determinant of access. 1.19. Basotho households today are overdependent on farming given the development of the economy and the income level. This overdependence reflects several factors: (a) cultural and social traditions-retrenched migrants and the families of rural and urban migrants stay in rural areas, where they try to combine subsistence farming with other activities, and (b) low levels of education among rural residents, which makes them uncompetitive in urban labor markets. Low education carries a bigger penalty in urban areas, and 81 percent of heads of households in the Rural Mountain region have no education. Simply put, without education, opportunities to find wage-paying jobs in urban areas are limited, particularly for women. But the effect of too many people trying to make a living in rural areas is increasing landholdings to small to be commercially viable, limited investment in higher value production systems, land degradation, and a worsening cycle of poverty. 1.20. Even young households with a relatively high level of education seem reluctant to migrate permanently to urban areas, despite widespread complaints about the stress of limited opportunities in the rural areas. Instead, these households either send one member to South Africa (often illegally) in search of jobs, or send one member into Lesotho’s urban areas to find work. This is an adaptation of the previous external migration strategy to internal migration, as the rest of the household is left behind. Evidence of the effectiveness of these strategies is 6 mixed. Controlling for other factors, rural households with an internal migrant are 21 percent poorer than the average rural household (those with an external migrant, on the other hand, are 14 percent richer). At the same time, urban households with a textile worker (most of whom are migrants) are better off. It is possible that internal migrants who move without their families are less likely to send money back to rural areas than, for example, miners, where a portion of their pay is sent to Lesotho by the mining companies. Youth who migrate illegally for seasonal jobs report earning good wages abroad, though they face the risk of losing them in fines at the border. 1.21. New migration patterns are having social effects. In the traditional Basotho society with tightly prescribed roles for men and women in production and reproduction, increased poverty of households with internal migrants frays village social structures and informal safety nets. As men used to die faster than women from HIV/AIDS in rural areas (the epidemic in Lesotho originated with mining workers, and thus men have had the disease longer), the number of female-headed households is large. In rural areas, these households are 10 percent more likely than other to be poor. Youth in rural areas, especially males, report that they are discouraged by limited options in the rural economy and that they do not see pathways out of poverty, either domestic or external. D. PURSUING SHARED GROWTH THROUGH STRUCTURAL TRANSFORMATION 1.22. Lesotho can continue its structural transformation to, a middle-income country while achieving shared growth by creating better-paying, more secure livelihoods in the private sector. Three strategies will be key to the successful continuation of the transformation: (a) continuing to develop the investment climate for labor-intensive production for export; (b) a rural development strategy that supports the commercialization of agriculture as well as programs to reduce land degrad.ation and support the subsistence sector; and (c) a strategy to support socioeconomic and geographic mobility into higher-productivity sectors. 1.23. Lesotho’s economic growth will reduce poverty as long as stable jobs for those with moderate education are created in the urban private sector. The manufacturing sector is an example in which public investment in infrastructure has crowded in private investment, yielding a high social and economic rate of return for the country. Lesotho’s labor force is growing slowly, but the educational level is rising, so it should not be as difficult as in other Southern African countries to create adequate employment opportunities. Continuing the’ public policies and investments that brought about private investment in labor-intensive export sectors will be the anchor of a successful public-private growth strategy going forward. 1.24. Studies have shown that structural transformation of livelihoods in the form of rapid urbanization is necessary for low-income countries to transition to middle-income status. This is not surprising-a concentration of business (industry or services) brings economies of scope in production, and the informal sector can easily develop beside large firms offering goods and services that workers need. Often overlooked are similar economies of scope in the provision of public and private services to households. 7 Box 1.1: Strategic Options: Livelihoods And Income Poverty Reduction Lesotho’s economic transformation has brought many households more stable and secure income, but it has left some households behind, especially in rural areas. Lesotho needs to continue structural transformation to raise household incomes by creating better-paying, more secure employment in the rural and urban private sector. Internal and external migration should be supported. Recommended Strategies Create goodjobs by staying ahead of the export game: Lesotho must continue its program to improve investment climate for labor-intensive export industries and services in urban areas (e.g., provision of electricity, water, and factory shells on a sustainable basis and simplification of permit procedures, among other things). Support urbanization and urban small and medium enterprise (SME) development, Efforts in this area should include improving urban land tenure regulation and registration so that urban land market transactions that are currently informal become smooth and efficient and urban land can be used for collateral; studying constraints to setting up and maintaining a household business and acting on the results; and supporting migration of households from rural areas. Improve the investment climate for commercial agriculture. This includes improving marketing infrastructure, studying constraints to contract farming and other more profitable farming systems, studying the shareholding and rural land tenure system to develop options to encourage investment and reduce degradation, and acting on the results of these studies. The private sector should be included in planning efforts, through the development of effective local institutions (producer coops, etc.). Work with South Africa to increase remittance income by developing a medium term strategy to facilitate legal migration for employment for low-skill workers who now migrate illegally. 1.25. As with the industrial sector, the public sector has a role to play in setting up the infrastructure for households to relocate to locations where new jobs are being created. Qualitative surveys have highlighted poor access to land caused by the lack of an efficient, nontransparent land market as being an obstacle to rural-to-urban migration of households. Another factor that discourages households from migrating is the lack of microbusiness opportunities. Lesotho has a very small share of the urban labor force working in household enterprises. A special study on constraints to micro and small business development could be helpful. 1.26. Sustainable rural development will continue to be an important pillar of Lesotho’s poverty reduction strategy. This means combining measures to improve the profitability of crop agriculture-those for which Lesotho has an advantage versus South Africa-and livestock activities with measures to reduce or reverse environmental degradation and improve food security in the subsistence sector (such as no-till farming methods). At the same time, incentives should support the movement of much of the rural labor force out of agriculture. 1.27. Existing analysis of commercial agriculture potential suggests that investment to support horticulture would have the highest payoff. Expansion of livestock activity may be possible as well. Achieving a higher growth rate in agriculture will require: public investment in infrastructure and agricultural extension services (to get a high rate of return, public investments and programs need to be undertaken where they attract private investment, not replace it); and development of financial mechanisms to support farmers in managing risks. Regulatory reform is needed as well to increase competition and support the development of market supporting institutions. Data needs to be collected so that economic and financial analysis of investment programs can be done prior to implementation. 8 1.28. It is unclear whether the current customary law covering land ownership and transactions is a binding constraint to investment in agriculture. Analysis of this policy, including the determinants of sharecropping arrangements and the social and economic consequences of the current law and customs, would be helpful. 1.29. The government may wish to look into the feasibility of measures to increase the remittances from South Africa in the economy given their poverty-reducing impact. Countries such as the Philippines have developed training programs specifically for the external market, in collaboration with the receiving economies. It might be possible to undertake such agreements with South Africa. E. USING PUBLIC POLICY TO REDUCE POVERTY AND INEQUALITY 1.30. Economic growth has less effect on poverty in Lesotho than in other countries because of high monetary inequality. Many people in Lesotho have incomes very far from the poverty line-the average consumption per capita of the poorest quintile is only one-quarter that of the middle quintile (where the poverty line sits) and only 6 percent of the average income of the richest quintile. High inequality means that when growth occurs, it is not widely shared, so poverty reduction is slower than when inequality is lower. Worldwide, high levels of inequality are also associated with lower overall growth. 1.31. Inequality has a long history, and is difficult to address with economic policies alone, but there is a role for the public sector. Public expenditure can reduce the inequality in welfare and opportunities by creating productive assets for the poor-roads, water supply, and other infrastructure-and ensuring equal access to education and health services so that poor households can build up human capital. Public policy can ensure that all citizens have equal opportunities, do not suffer from discrimination, and are able to exercise economic and political rights. ,1.32. Public expenditure accounts for nearly 50 percent of GDP, a high percentage for a lower-middle-income country. The high level of public expenditure is essentially the distribution of two sorts of rental incomes, Southern African Customs Union (SACU) tax revenues and receipts from selling water to South Africa, back to the population. Public spending is a key tool of the government in poverty reduction. Four of the eight goals in the PRS-develop infrastructure, improve access to health care and social welfare, improve quality of and access to education, and improve public service delivery+xplicitly relate to improved public service. Many subcomponents of the PRS, such as improving safety and agricultural extension and combating HIV/AIDS, also are key public service delivery areas. Whether the high level of public spending has benefitted the poor, however, is debatable. Three views on this situation are: 1) yes, the poor have benefited in absolute terms because they receive more public services than before; 2) no, the poor have not benefited in relative terms-poverty-targeted spending and spending on services the poor can access is a low share of the budget, indicating that with a small shift in spending, the poor could benefit a lot more; and 3) it is unclear whether the poor have benefited because the data on expenditure and beneficiaries is so poor. A case can be made for all three viewpoints. 9 1.33. Data show a major expansion of social and economic infrastructure, much of it in rural and underserved areas, resulting in substantial improvements in access for everyone, but especially those in poor areas. In the most recent Afrobarometer survey, the government’s performance in providing public services was rated higher than its performance in managing the economy. Total public spending social service delivery in value terms is high compared to other Southern African countries. As a share o f total expenditures, however, spending on service delivery is quite low by regional standards, indicating scope for more poverty-targeted spending.2 1.34. Public financial management is weak and nontransparent, which hides inequities and inefficiencies in public service delivery and investment. Reliable information to monitor budget execution is generally not available. Data do not allow mapping of expenditures to priority programs in the government’s PRS, or to income levels of the beneficiaries. An ad hoc, line-by- line review exercise for 2004-05 showed that only 34 percent of government spending was directly poverty related. Administrative data on access and outcomes in education has improved dramatically in the last few years, but utilization and service delivery in health yields confusing and contradictory results. Consequently, it is difficult to assess the contribution of the public health system to persistently low health and nutrition indicators, and how the system could function more efficiently to reach the poor and excluded. Education Spending-Who Benefits? 1.35. Primary education is generally viewed as a success story in Lesotho. Between 1994 ‘and 2002, there was a 16 percent fall in the number of adult males and a 19 percent fall in the number of adult females with no education. The Free Primary Education (FPE) policy initiated in 2000, combined with increased expenditures on facilities, teachers and teacher training, and inputs, brought net enrollment ratios to 8 1 percent of boys and 87 percent of girls in 2004. One of the main reasons the ratio is below 90 percent is the tendency for children to start school late (by age 9, however, enrollment is well over 90 percent). Data from 2004 show that only in the Rural Mountain region is there a serious problem with primary school enrollment, and this is partly because of “herd boy” syndrome, where boys are needed to tend livestock. Special programs to address this issue are being initiated. 1.36. Once the analysis is disaggregated by wealth, the story changes. Lesotho has well- organized data on educational expenditures, allowing a comprehensive analysis of spending and poverty. The 2004 survey shows that dropout rates are five times higher in the poorer quintiles than in the richest quintile, and repetition .rates are 50 percent higher. Poor orphans are particularly likely to have low educational attainment given their age, even after controlling for location and gender. Even worse, fees are creeping back in as a response to flat funding levels, despite greater numbers of students and more facilities. In 2007, 94 percent of primary schools charged some fee; fees are higher on average in public schools than in community and private schools. Many social service facilities-90 percent of schools and 50 percent of health facilities-are owned by faith-based NGOs such as churches. The government pays these entities to offer the services. In education, payments cover, for example, teacher salaries, educational materials, and other recurrent costs. 10 Figure 1.3: Net Enrollment by quintile, 2004 100 80 60 40 20 0 1 I I I Boys Girls Boys 1 Girls , Primary net attendance rate Poorest Second Secondq net attendance rate Fourth IRichest I 1.37. The FPE policy has not successfully addressed issues of efficiency. Late entrance and high student repetition and dropout rates are major problems, and problems that hit poor students the hardest. The net enrollment rate is consistently low due partly to a high number of students entering primary at an older age than they are supposed to. In 2006, 55 percent of students starting grade 1 were above 6 years old; 23 percent were older than 7; and 4 percent older than 10. Each year, nearly 20 percent of students must repeat classes. High dropout rates mean that only 44 percent of boys and 60 percent of girls who started grade 1 in 2000 (the first year of the FPE policy) completed grade 5 by 2007. 1.38. Access to secondary and tertiary education remains low, particularly for the poor. Secondary education is not free, and net enrollment rates are a mere 2 percent for boys and 8 percent of girls from the lowest wealth quintile compared to 38 percent of boys and 55 percent of girls in the highest wealth quintile. Only 1 percent of tertiary-level students are from the poorest quintile, compared with 66 percent from the highest. More than 60 percent of males under 24 years old who are not attending school cite too-high fees as the reason for not being in school. Even after controlling for location and household wealth, orphans ages 14-17 have half a year less education as nonorphans of the same age. 1.39. Benefit incidence analysis shows that the distribution of expenditure on education in Lesotho is among the most regressive in the world. The richest wealth quintile receives by far more public education resources than all other quintiles (see Figure above).The lack of access of poor children to secondary education contributes strongly to the overall regressivity. In a addition, secondary education receives the smallest share of the education budget, just 19 percent. Most of the remainder of secondary education costs must be paid by households, and thus most of the students who complete secondary education are from households at the richer end of the wealth spectrum. At this point in the education progression, however, the purse strings of the government budget open wide. On average, the government spends about 20 times as much per student on tertiary education as it spends for secondary education. Expenditures per student on tertiary education are three times higher than in neighboring Botswana, for example, a much richer country. 11 Figure 1.4: The top quintile by wealth receives 42 percent of public expenditures on education, while the.lowest quintile receives only 9 percent Poorest. go/. v Rich, 20% Sources: DHS 2004. Health expenditures deliver access but not better outcomes 1.40. Although the government spends about 7-8 percent of its budget on health and social welfare, the Basotho have poor, relatively stagnant health indicators. The health sector faces a number of challenges, including maintaining skilled staff and reaching out to a largely rural population. Some of these have been met, as the average Basotho is now likely to seek care in either a government or community hospital when sick, but service problems in public health facilities are common (see Figure below). Approximately 40 percent of Lesotho’s health system is owned and operated by church groups (CHAL). The government provides partial financing to these providers. 1.41, Poor households are less likely than richer ones to seek care, citing cost and distance as the major impediments. Half of the poorest two quintiles reported not having enough money to seek treatment in the community health care center in 2004. Poor women are particularly likely to cite cost as a factor. Distance also matters, as only 34 percent of the poorest households live within an hour of the nearest health facility (more than two-thirds of Basotho reported in 2004 that they walk to access health care). In fact, 25 percent of the poorest quintile and 25 percent of people who live in the Rural Mountain region have more than three hours to travel to nearest health facility. Distances to a health facility in the Rural Foothill region and Rural Senqu River Valley also are large-only half of the households in these areas live within an hour of the nearest facility, and many are more than four hours away. 1.42. The Basotho were less satisfied with the government’s efforts to improve basic health care than they were with efforts to improve education. According to an Afrobarometer survey of 2005, the approval rating for the government’s efforts to improve health care was less than 60 percent, compared to more than 90 percent for education services. The gap between the responses of the poorest households and the richest is high, at 5 1 percent compared to 64 percent. Households complained of long waiting time (even while administrative data show underutilization) and lack of supplies. Quality issues such as these may be one reason that spending has not produced significant results. In general, more reliable data are required to understand the destination of health spending the outputs of the health sector. 12 Figure 1.5: Service problems in public health clinics and hospitals, 2005 Source: Afrobarometer 2005. Note: Percent = share of people who experienced problems "a few times" or "often," as opposed to "never" or "once or twice" among those who answered. 1.43. After five years of negotiation, the government and CHAL reached a new financing agreement to improve access to health care for the poor. The agreement took effect in 2008. Under the agreement, both government and CHAL facilities will cease charging user fees for poor patients. Services with public goods dimensions, including those for tuberculosis and HIV/AIDS, will be provided free of charge. Fees for other services will continue to be charged but these will be standardized across government and CHAL providers of health care. Data is not yet available on the effects of these policy changes. Addressing gender inequality 1.44. 'Basotho women have long had an important but legally and culturally unrecognized economic role. As men went to work in South African mines, Basotho women ran households, grew food, collected water and firewood, and, in the process, got themselves and their children (especially their girls) educated. Under both customary and statutory law, however, once they married, women were considered minors of their husbands and thus unable to own property (or to inherit it), enter into contracts, make a formal complaint to the government or be party to a lawsuit, or get a passport on their own, among other things. Even when women left the household and migrated to Maseru to work in textile factories, for example, they still had no formal legal powers over their own lives or the money they earned. 1.45. This situation changed in 2006 when The Legal Capacity of Married Persons Act (LCMPA) was passed. The act repeals a host of discriminatory provisions in the formal legal system and represents a crucial advancement in women's legal position. Women can now access credit, improve their land (if they get access to land), invest their money, engage in entrepreneurial activities, and be the sole guardians of their children in the case of, for example, death of the father or abandonment. Moreover, the law is perceived as a powerful policy framework that sets the standard for the government's gender-related agenda of hrther legislative and policy changes to reduce women's social subordination and empower women to contribute fully to Lesotho's economic development and poverty reduction. 13 1.46. Implementation of the LCMPA is proving to be a challenge, however, due in large part to the dual-customary and statutory-legal and judicial systems. Not only has an attempt to harmonize the legal systems not yet commenced, but the customary system rarely upholds women’s rights. While there is some demarcation of applicability between the two systems, there is significant overlap in the areas of domestic law and property rights. In practice, there is a long-standing tradition of resorting to the customary law in these two areas, in part because proceedings under formal law are lengthy and cumbersome. Both systems are characterized by a lack of adequate gender sensitization and awareness of women’s rights. As a result, women continue to have difficulty accessing justice in either system. 1.47. The enactment of LCMPA is part of a positive reform trend in Lesotho’s legal system to support gender equity and create the legal space for the full participation of women in the economy and society. Other new legislation includes the Sexual Offences Act of 2003, which for the first time criminalizes rape-including marital rape-and mandates specific penalties. But implementation of this policy is proving a. challenge, and sexual violence toward women continues to be high, at least in urban areas. In a recent study of urban women, “25 percent of women surveyed reported ever being physically forced to have sex; 13 percent reported that forced sex was attempted; 3 1 percent said that they were touched against their will; and 11 percent reported being forced to touch a man’s genitals.” (Brown, et al, 2006) Often, perpetrators are boyfriends and known community members. 1.48. It is not surprising that a change in the legislation is not sufficient to reverse the long tradition of economic and social discrimination. Attitudes and behaviors change slowly. Many women are not even aware of their change in status with the enactment of the LCMPA, and some oppose it, finding these changes threatening to the customary social structure. Supportive institutions such as legal aid clinics for women are rare and typically organized by nongovernmental organizations (NGOs). Government programs to foster women’s awareness of their own rights and how to exercise them, an important part of the implementation of the gender equity agenda, are lacking in Lesotho. A further complication is that the police and judiciary are often not trained in how to handle gender issues and sexual violence. F. TOWARD A MORE EQUITABLE SOCIETY 1.49. Public spending remains a powerful tool for poverty reduction in Lesotho. Other developing countries with high income inequality have successfully used the public purse and the law to make up for the resources that the poor and excluded do not have and did not inherit, providing new opportunities and enforcing the rights of the excluded to participate. Expanding educational opportunities produces shared growth; limiting such opportunities, as Lesotho does now through the underfunding of primary and secondary education, produces unequal growth and lowers growth rates. The experience of the FPE policy in the education sector shows that it is possible to make a major difference in access and outcomes for the poor when a clear policy of universal access is implemented. But it also shows that it is important to monitor progress (inputs, outputs, and outcomes). Though Lesotho has the public resources to improve its poverty situation, more resources must to be allocated to improving the welfare of the poorest citizens. 1.50. While the FPE policy is a positive step, it did not lead to universal access to primary education. One reason for the shortfall is that the policy was allowed to lapse-even public schools charge fees. Another reason is that other troubles in the education sector-late starting age, high repetition rates, and high dropout rates among the poor-were not addressed. Additional funding is needed for primary education in order to reduce fees. Ideally, this funding would be tied to improvements in quality and completion. 14 1.5 1. Providing middle-class and rich children with grants for higher education in South Africa while denying the poor secondary education perpetuates inequality. It is important for growth, equity, and poverty reduction that access to secondary education be expanded, a goal that can be reached through a combination of increasing the capacity of secondary schools and lowering the cost of education to households. This can be financed by targeting the higher education subsidies (e. g. directing then to poorest households only) and forcing higher-income households to finance more of their children’s higher education, as is done in other countries. 1.52. A well-functioning, effective health care system is important not only for building and maintaining human capital through the reduction of mortality and morbidity, but also for improving the nutrition of young children and as the cornerstone of the fight against HIV/AIDS. In many cases, simple measures to improve the quality of the health care system would help. For example, in 2006, a survey showed that more than half of the local health centers where women go for antenatal care did not offer services for the prevention of mother-to-child transmission of HIV/AIDS. Nutritional counseling and supplements are usually not available in rural health centers. Thus, the recent policy measure to reduce fees for health care is helpful, though it should be financed in the national budget so that fees do not creep back in. The , government, however, cannot expect to get value for the money it invests in the health care system (including the CHAL) without adequate data to monitor results. Both regular and better- quality household survey data, as well as better-quality administrative data, are needed. 1.53. Passing the LMCPA is the most significant step the government has taken to date in its program to modernize Lesotho’s legal system promoting gender equity. The effect of this legal change, though, may be severely limited if it is not accompanied by effective implementation and other contextual conditions. An effective implementation program should include: 0 further legal reforms in inheritance rules, land rights, and corporate and labor issues; elimination of clauses exempting customary law from application of gender-responsive legislation; and clarification of the scope of application of statutory/common and customary law; public awareness campaigns; training of justice officials and professionals (including, but not limited to, financial institutions); and gender sensitization in schools and universities on women’s rights, including the right to be free from violence; creation of legal aid clinics and support services for women, especially those subjected to violence or abuse; 0 development of programs to monitor results, including systematic collection and publication of gender-disaggregated data. 15 Box 1.2: Strategic options toward a more equitable society High income inequality reduces the potential for growth to reduce poverty, in addition to undermining growth itself. But in Lesotho, public expenditure on social programs goes disproportionately to the rich, creating rather than reducing inequality. Health and education sector budgets need to be rebalanced so that poor households get a higher share of benefits. Women need to be freed from the fear and reality of debilitating and degrading sexual violence, and be able to claim their newly-awarded economic rights. Recommended strategies (a) Ensure universal primary education for poor and middle-class children regardless of where they live. This includes eliminating all fees in primary education, rewarding schools for results (completion rates and exam results), monitoring school enrollment to make sure that “voluntary” charges do not stop poor children from completing primary schools, implementing programs targeted at herd boys, and evaluating results. (b) Increase access of poor and middle class children to secondary education. This includes increasing expenditure on secondary schools so that fees and boarding charges can be reduced; increasing the capacity of secondary schools (number of places) especially those accessible to rural children; and ensuring an adequate supply of secondary school scholarships so that all poor and middle-class children who pass the primary school exam can find a place in a secondary school. (c) Reduce share of budget for higher education. This includes requiring middle-class and upper-class households to bear a higher share of the cost of tertiary education tuition than they do currently and providing opportunities for loan forgiveness in return for service in remote, poor areas. (d) Implement health and nutrition programs targeted at rural needs. This includes ensuring that rural health centers get their fair share of inputs such as drugs and supplies; using demand-side accountability tools to ensure better service for poor households (e.g., reduced waiting times); and developing nutrition programs for young children and pregnant mothers in rural areas that combine supplements with training for mothers and effective monitoring of results. (e) Enforce laws that protect the rights of women. This includes educating both women and men about women’s legal rights and the relationship of those rights to socioeconomic development, enforcing laws against rape and other forms of sexual aggression against women and supporting victims in their efforts to press charges against aggressors, and creating legal aid programs for women claiming their rights. (f) Continue legal reform by harmonizing the customary and statutoly systems. This includes revising the customary and traditional laws (known as the laws of Lerotholi) to end gender discrimination, eliminating clauses allowing customary law to uphold gender-discriminatory practices, and training legal and law enforcement professionals in both systems on the application of newly enacted statutes. ~~ ~ G. FIGHTING HIV/AIDS AND REDUCING VULNERABILITY 1.54. The recent spike in food prices highlighted Lesotho’s continuing vulnerability-as a nation, as communities, and as households - to food insecurity, hunger, and malnutrition. Lesotho has long been food insecure, though, owing to the ecological conditions (tendency for drought or rains at the wrong time) and the failures of past policies to develop a commercial agricultural system. More than 15,000 tons of food aid was required in eight out of 15 years between 1990 and 2004, and in only two years was food aid less than 5,000 tons. Nutritional evidence suggests that not all households are able to bounce back from repeated food insecurity. In addition, households are also vulnerable to other shocks, such as the illness or death of a family member. These risks have also increased in recent years. Health indicators have not improved while the unchecked HIV/AIDS epidemic has led to skyrocketing mortality. Success in poverty reduction requires a systematic effort at the national level, to reduce the risks, and at the national and community level, to help households cope. 16 HIVIAIDS: Lesotho’s Biggest Development Challenge 1.55. Lesotho’s high prevalence of HIV/AIDS-23.2 percent of the total adult population in 2005, and more than 40 percent of the population ages 3 0 4 0 4 s the most important growth and poverty reduction threat. It is a grim fact that Lesotho has the third-highest HIV prevalence rate in the world. The percentage of HIV-positive women is greater than the percentage of infected men in almost all age groups except men 40 years and older, and prevalence is considerably higher among young women than among young men. By age 22, 30 percent of young women will have been infected, a statistic that jumps to almost 40 percent by age 24 and peaks at almost 45 percent as women approach their late 30s. Men’s prevalence rates peaks later, at about age 40. 1.56. The principle mode of transmission of HIV is heterosexual contact, mostly through multiple concurrent partnerships, often intergenerational ones. The latter factor leads to a higher rate of infection among younger females than males. After age 40, the rate of infection among men is higher. High risk groups include: 0 miners (who are 9 percent more likely than non-miners to contract H IV controlling for all other risk factors) and sexual partners of miners (who are 6 percent more likely than other women to contract HIV, controlling for other risk factors); 0 adults with low levels of education, especially females currently married males and formerly married females; and 0 people living in middle income households, and those living in urban areas, especially females.’ 1.57. Factors driving the spread of W / A I D S include information gaps. The 2004 DHS, for example, showed a shocking lack of knowledge about AIDS among youth ages 18-24 and the persistence of sociocultural norms including the tolerance of sex-based violence and ‘multiple partnerships (the latter of which is often encouraged by geographical mobility). Both men and women report practicing unsafe sex, both within and outside the marriage. 1.58. The rise of HIV/AIDS means that the probability of death among Basotho adults has increased tremendously. Adult mortality rose from less than 2 percent per five-year period in the 1970s to more than 8 percent for males and 12 percent for females in the first four years of the current decade. In general, the older the age group, the higher the mortality rate, although for the 2000-04 period mortality was higher among those ages 3 5 4 4 than among the 45-54 group. Without treatment, there is an 8-10-year median interval between HIV infection and death. It is hoped that increasing availability of antiretroviral drugs will reduce premature mortality. This analysis is based on 2004 DHS data, which included a sero-survey. It is possible, for example that married women are now at as high a risk as unmarried women, depending on how behavior has evolved. This underscores the need for more frequent data collection. 17 Figure 1.6: Adult mortality by residency and gender, 1975-2004 0.14 7 010 i t 008 e $ 006 I 0 0 4 0 02 0.004 1975-79 1980-84 1985-89 1990-94 1995-99 2000-04 Period /+rural males +urban males +rural females - D u r b a n females] Source: DHS 2004 and authors calculations, 1.59. As a consequence of the steep increase in mortality, the number of orphans is rising rapidly, and faster than in other African countries. In 2004, 30 percent of all children had lost one or both parents. The long-term effects of orphanhood are not well known. Fears of “child- headed households” had not materialized by 2004, as both the DHS and the HBS survey found very few of these, but our analysis shows that orphans are less likely to progress normally through the school system and less likely to enroll in secondary school (where the fees are high). Double orphans, maternal orphans, and orphans who are not living with relatives are the most disadvantaged in terms of educational outcomes, suggesting that efforts should be made to place orphans with family members. 1.60. Lesotho is more than 15 years behind the leading countries in Africa (e.g., Uganda) in preventing and coping with the HIV/AIDS epidemic and thus faces the most severe crisis on the continent. Over the past three years, however, Lesotho has taken a number of important steps to reverse this situation, including the formation in 2005 of the National AIDS Commission (NAC), the body coordinating the national response, and the national strategic plan, which among other things identifies strategic areas of focus and assigns responsibilities among the myriad donors, NGOs, and government agencies and bodies active in combating the epidemic. The national strategic plan covers all key areas-prevention, treatment, and impact mitigation, including support for orphans-and is a country-wide, cross-sectoral strategy. While the plan is recognized as a major step forward, its effectiveness against the epidemic has been limited by several factors: 0 lack of clarity on roles and responsibilities between the districts and the national government, especially as the decentralization agenda moves forward; 0 long-standing human resource gaps in the health sector (for example, despite efforts at training, Lesotho’s nursing staff declines every year to external migration); 18 capacity constraints in ministries, reflected in the inability to spend allocated funds and carry out programs (for example, an estimated 7 million condoms procured in 2004 for distribution by the Ministry of Health were in storage for almost a year), inability of the NAC to develop a comprehensive information base about the epidemic and to monitor the response to efforts to combat the disease; few attempts within the government or its development partners to evaluate the results of interventions; and differing procedures and practices among multiple donors, which makes coordination and monitoring an enormous challenge. 1.61. Overarching all of these institutional constraints is the inability of the government to develop a response tailored to high-risk groups-young women and married women- possibly owing to an unwillingness to recognize that Lesotho has one of the highest rates of concurrent partnerships in the world according to recent survey data. One in five men in Lesotho reported having two or more sexual partners in the last four weeks. Such concurrent partnerships without condom use are particularly dangerous because the risk of transmission of HIV is highest in the first three weeks following infection. The active use of condoms could reduce this transmission, but married women are particularly reluctant to request the use of condoms even though their knowledge of the risk of sexually transmitted diseases is quite high. This is an important source of vulnerability that needs to be addressed if the epidemic is to be halted. A particular problem is male miners, who not only may bring the virus home, but may be infected by a partner who herself is having a concurrent partnership while he is away. 1.62. Provision of prevention material through the school system has been delayed owing to an unwillingness to implement a curriculum that is explicit about sexual transmission and recommends the use of condoms. At this point, the sources of information for youth are after-school and summer programs run by NGOs. These programs seem to be working, though, as anecdotal evidence shows a higher degree of awareness, at least among teenagers, than in the 2004 survey. But overall, most prevention efforts are not targeted to, for example, young women at high risk of contracting HIV/AIDS from older men. Rather, they are very general in their message. 1.63. Lesotho has focused major efforts on increasing access to treatment. Despite the still- high prevalence rates, the government claims that more than 40 percent of adults who need treatment get it, with the first-class World Health Organization (WHO) drug regimen. This is a major achievement. Many organizations are involved, in part to help overcome the limitations of the public network, including the staffing shortages. But only a small number of the estimated 18,000 children with HIV/AIDS are getting treatment. Food security and nutrition 1.64. The agricultural sector has not performed as an effective food security system for households, communities, or the country, and there is little evidence that it will be able to in the future. Lesotho’s mainly rural residents have long relied on subsistence agriculture for most of their food. Over the past ten years, though, many rural households have not been able to meet their food needs from their own production, especially in years in which growing conditions are bad, and they have not had enough cash from earnings in the good years to buy food in local markets. In 2004, the poorest households in the Rural Lowland and Rural Foothill regions reported that they produced only 25 percent of their food; the amounts are even lower in the Rural 19 Mountain region. As a result, rural communities are food insecure and rely heavily on food aid from donor organizations. Aside from the difficulties faced by donors in raising funds and distributing food aid as often as it is needed (roughly every other year), this approach provides negative incentives to the development of a sustainable domestic market for food. Alternatives such as providing cash (possibly through public works employment) would strengthen the local market. 1.65. Owing to food insecurity and to weakness in the public health system, anthropometric indicators for children under the age of 5 are very poor, and have been so for over ten years. Not only are the national anthropometric indicators bad given Lesotho’s income level, but the wide gaps between children in the richest and the poorest households are also surprising. Poor nutrition in the first years of a child’s life leads to poor cognitive function later on, thus making education investments, for example, less effective. 1.66. Standard packages of comprehensive nutritional interventions are available to address malnutrition of mothers and children, but these do not appear to be in widespread use in Lesotho. Such behavioral change programs combine nutritional supplements with education designed to teach mothers how to nourish their children effectively at a low cost. The first two years of a child’s life represent a key window of opportunity, so these programs normally start with supplements for pregnant mothers and information on well-baby care. Toward A New Public-Private Safety Net 1.67. Lesotho has a number of cash transfer and social welfare programs targeted at specific groups, including care for disabled children and scholarships for orphans to attend secondary school. A review by the NGO CARE in 2006 identified more than 30 official programs, along with several ad hoc programs run by local governments and NGOs. But existing survey data do not allow an analysis of who is adequately covered, who is missed, and who receives benefits but is not poor. And while CARE was not able to track the cost per beneficiary for many public programs, qualitative research does show evidence that traditional village and family networks are increasingly strained by the new social vulnerabilities, and it seems that the piecemeal social protection approach currently in effect excludes part of the population in need of assistance. In addition, threats such as nutrition deficiencies and HIV/AIDS require targeted interventions. H. PUBLIC SECTOR ACTIONS TO FIGHT HIVIAIDS AND REDUCE VULNERABILITY 1.68. Reducing excess mortality and morbidity from HN/AIDS is the best way to reduce poverty and improve welfare. Lesotho’s focus on scaling up the delivery of antiretroviral treatment in all areas of the country is key to support families with HIV/Aids. Antiretroviral treatment, though, extends life only if other conditions are present, such as good nutrition. A successful national treatment program must also be complemented by bold and efficient HIV prevention efforts, which should be explicit, well targeted, and comprehensively evaluated. HIV/AIDS programs should go beyond general messages and address upfront the specific behavioral characteristics contributing to the spread of the disehse in Lesotho, including migration of men to the mining sector in South Africa, internal migration of young girls to garment factories, widespread stigmatization of people with the disease, and residual denial of the problem. The government should invest in efforts to better understand the specificities of the epidemic and how to create effective messages to combat it within the Basotho context. 20 1.69. Lesotho’s food security policy needs to move away from reliance on subsistence production of grains towards creating secure income sources so that households can buy food. The fact that richer households grow more of their own food than poor households clearly indicates that reducing reliance on agriculture will not support the poor. Lesotho’s high dependence on food aid may be creating negative incentives from production activities and marketing chains; this needs to be analyzed. 1.70. Government policy needs to be informed by a careful review of the plethora of active social protection programs in Lesotho. Such a review should be based on administrative, budgetary, and household survey data of sufficient quality to allow detailed analysis and policy formulation. The focus of safety net efforts should shift from reactive emergency programs launched in the case of crisis (e.g., food aid) to proactive risk management that protects and support livelihoods. Given the scope of the challenge ahead, the government should give this shift high priority. Lesotho needs to rework its mix of public and private approaches to face these threats, as well as develop adequate systems to monitor poverty and vulnerability, assess the causes and consequences of such vulnerabilities, and evaluate the results of public and private interventions to combat them. ~~ Box 1.3: Strategic Options to Reduce Vulnerability The livelihoods of households-both farm and non-farm-can collapse in the face of a weather shock or a reduction in the earning power of the household owing to illness or death of a working adult. In Lesotho, the probability of these risks has increased with climate change, increased dependence on volatile international markets, soil degradation, globalization, and new diseases such as HIV/AIDS. Traditional village safety net systems and family networks are strained. Lesotho needs to develop a new mix of public and private approaches in order to aggressively prevent HIV/AIDS transmission and mitigate the worst effects of risks on households without bringing negative incentives for productive activities. Adequate data is needed for policy design and implementation in all areas. Lack of data damages household welfare and kills people because it supports ineffective policies. Recommended strategies (a) Know the problem. This involves contracting a private company to collect LSMS data every three years, beginning in 2008, in order to analyze household response to government programs, determine which households and individuals are at risk, and ascertain who benefits from the many different social programs now in existence. (b) Consolidate and reduce the number of programs to improve their effectiveness and affordability, and increase targeting. This includes using cash transfers, which are an effective way to fight hunger, malnutrition, and poverty (though they rapidly become expensive if not well targeted at, for example, the poorest children or low-income orphans at risk of dropping out of school); using public works programs to mitigate risk as they are self-targeting and flexible so can temporarily increase in size when a covariate risk occurs; and develop specific program aimed at underemployed rural youth. (c) Ramp up HIV/AIDS prevention programs, targeting high-risk groups for behavioral change. Lesotho needs specific prevention programs targeted at youth and women, including middle-class married women, especially those whose partners migrate out of the country for work. These programs should be evaluated for effectiveness. 21 In every section of this report, the weaknesses of current household survey data, administrative data from ministries and public sector delivery units, and data on public sector spending has been highlighted. While it is impossible to quantify the costs of weak data in terms of growth, poverty reduction, and waste and inefficiency in the public sector, it is certain that the situation allows ineffective programs to continue and households to suffer. For example, the latest survey data on HIV/AIDS prevalence and mortality is from 2004-a lag in monitoring which is difficult to justify in a country where one in four adults are estimated to be afflicted. The latest household survey, undertaken in 2002/03, lacked much of the data needed to track employment, the Millennium Development Goals, poverty reduction, or other development key outcomes at the household level. This area of public sector management has been neglected for quite a while, so that now, although Lesotho is a lower-middle-income country, the information base for policy formulation and implementation is weaker than in other African countries that are poorer and have a lower level of literacy. 1.71. The government has taken some actions to improve the information base. A new financial management system currently being implemented, for example, is expected to allow much better tracking of expenditures and objectives. The Ministry of Education and Training used an international consulting firm to upgrade its administrative systems, so that up-to-date data on educational outcomes by district, type of school, gender of student, and other measures is now available. In addition, the Lesotho Bureau of Statistics is preparing ,a new strategic plan that may help develop the consensus necessary for long-term reform, despite the fact that it is not likely to enable the creation of new, reliable databases in the next few years. Efforts such as these to improve the information base should be continued. 1.72. More action is needed, though, if Lesotho’s statistical database is to be brought up to the level required to manage a diversified middle-income economy with a large public sector. The government should consider the following initiatives in order to improve the information base: Seek donor funding this year and contract a foreign firm to undertake the rapid production of an integrated Living Standard Measurement Survey (LSMS). By following standardized methodologies that have been internationally tested and are used by statistical agencies in many developing countries, high-quality microeconomic statistics could be produced in Lesotho. A user group of researchers and analysts could be formed to guide the survey, to make sure that data are collected on key interventions. Request and implement a new DHS. Given Lesotho’s high HlV/AIDS prevalence, putting the DHS on a three-year rotation would help monitor and evaluate the vast number of HIV/AIDS interventions carried out by donors, NGOs, and other organizations, as well as track the effectiveness of government programs. Conducting the DHS more frequently would also support the development and or evaluation of policies and programs to address many of the others issues identified above. Use the national budget as a tool to develop a results-based system of expenditure allocation. Within this system, a high level of outcome and impact evaluation should also be undertaken. See the technical appendix to this report for a full discussion of the weaknesses of Lesotho’s household survey data. 22 2. DIMENSIONS OF POVERTY IN LESOTHO A. INTRODUCTION TO THE REPORT 2.1. Despite a decade of economic growth and structural change, Lesotho still faces a substantial poverty challenge. The Government is committed to using public policy to create shared growth and reduced vulnerability of households in the future. After extensive national consultations, Lesotho published in 2005 its first national Poverty Reduction Strategy (PRS). This strategy was based on limited data. Since that document was prepared, two new data sets have been released by the National Bureau of Statistics - the Household Budget survey of 2002/3 (HBS) and the Demographic and health Survey of 2004 (DHS). The analysis in this report has been undertaken using these new data. The objective of this report is to inform the revision and updating of this strategy, using an integrated poverty, gender and social lens to reflect poverty’s many dimensions in Lesotho. 2.2. The analysis starts with this chapter, which frames the report using multiple quantitative and qualitative indicators to describe levels of and changes in opportunities and outcomes for households in Lesotho during this period of structural change. Chapter 3 explores the structural changes in the economy in more depth, and analyzes how changes at the macro and meso levels affected household livelihoods, and how households, especially poor households, responded. There were important differences in how males and females responded to these changes, owing to cultural norms and occupation segregation, and this is highlighted in this chapter. 2.3. The poorest and most vulnerable households in Lesotho are mostly found in rural areas, and the next two chapters of the report are devoted to this topic. The importance of rural development and profitable rural livelihoods for fighting poverty is the subject of Chapter 4. One of the key vulnerabilities in rural areas is food security, which results in levels of malnutrition which have persisted at surprising high levels despite consistent food aid support from donors. Chapter 5 completes the analysis of rural space by examining strategies that would produce both growth and a reduction in vulnerability and malnutrition in rural areas. 2.4. Improving access to basic services such as health and education is one of the key tools Government has used to build human capital and improve lives. Chapter 6 covers the past achievements and the challenges which remain in ensuring that the poorest households are healthy, they have access to infrastructure such as water, roads and electricity, and their children have access to education. 2.5. Chapters 7 and 8 cover major vulnerabilities of Lesotho’s growth and poverty reduction efforts - the challenges of exclusion owing to HIV and gender. Lesotho has a very high HIV infection rate, and as the progression of the disease moves to the stages of morbidity and mortality, family and communities are challenged economically and socially. Chapter 7 explores the implications of this pandemic for poverty reduction in Lesotho. Until very recently, married women were treated as minors in the formal legal system, despite the sizable economic role that they played in household livelihoods. Chapter 8 discusses the economic implications of the recent legal changes and the steps that are needed to insure that this changes contributes to shared growth in Lesotho. 2.6. Lesotho’s fight against poverty and exclusion is hampered by a lack of data of adequate quality describing household demographics, assets, livelihoods, levels of living, shocks, access to services, and social indicators, disaggregated into policy-meaningful 23 categories. Scarcity of data limits the analysis of why poverty is occurring and the dynamics that shape it, as well as are effective tools Government can use to reduce poverty. Except for the census data (not available at the time this report was completed), the Lesotho National Bureau of Statistics has not collected nationally representative data measuring poverty status and outcomes since 2004. As a result, most of the findings of this report are based on data which is 5 years old data or more. When data are available after this (e.g. from private surveys or sector reports), these are used as well. In spite of data weakness, policy makers can still benefit from taking stock of past accomplishments and disappointments. Box 2.1: Main data sets used in the report Data Set Collection Period Sample Size Representativeness Lesotho National, urban, rural Demographic and Sept. 2004 - Jan. 2005 8592 HHs and at district level Health survey I Household Budget Survey Nov. 2002 - Oct. 2003 5992 HHs National, urban, rural and at district level Household Budget June 1994 - May 1995 3787 HHs National, urban, rural Survey and at district level B. POVERTY AND LIVING CONDITIONS 2.7. In this chapter, we begin by reviewing the main indicators available who are the poor, where do they live, and what are their household characteristics according to monetary and non-monetary indicators such as health and education. We analyze inequality, finding that inequality plays a strong role in reducing the impact of growth on poverty in Lesotho. We use qualitative and quantitative analysis to identify the causes and correlates of poverty, including sources of income and vulnerabilities. We review available evidence on the trends in poverty over time, concluding that there is some evidence of improvements in levels of living over time, and education indicators have improved, but overall the trends are at best fuzzy, and at worst discouraging. 2.8. Poverty in Lesotho in 2002/03 was primarily rural, Using the national poverty line defined by the Bureau of Statistics (BoS) and a slightly redefined consumption aggregate5 the average national poverty head count in 2002-03 was 54, with 58 in rural areas and 40 in urban areas (Table 2.1). It should be noted that the national poverty line was high compared to the World Bank international $1 a day poverty line. Using the $1 a day poverty line produced a significantly lower average national head count of 37, with 41 in rural areas and 25 in urban areas. Variation existed among different urban areas and rural areas. The capital, Maseru, had the lowest percent of poor (20 percent) compared with the urban areas that had 28 percent. In rural areas, the rural Foothills had the highest level of poverty (47 percent) compared to 41 in the lowlands, 39 in the Mountains, and 36 in the Sengu River Valley. 2.9. Demographically, larger households with many children, headed by older and less educated heads were worse off, while households headed by migrating males were better off. Lesotho is farther along in the demographic transition, and as a result, the majority of the poor are adults, not children as in poorer African countries. Elderly (65 and above) have about the same poverty rates as children, slightly lower than adults, but make up only 6 percent of the poor. See Annex to chapter 2 on definition of household for measurement of poverty. The estimates in this report are different from those produced by the HBS because we excluded migrants (and the estimate made by HBS of their consumption) from estimate of household consumption per capita. 24 Figure 2.1 Poverty by demographic characteristics and location, 2002103 (“/o) 70% , 60% 50% 40% 30% 20% 10% 056 j LivelihoodZone 1 Educationof head Household size Source: HBS 2002/03 and authors’ calculations 2.10. Average household size also follow very closely the consumption q u i i l e . The average household size in the fifth quintile was 6.7 persons-significantly larger than 4.7 in the highest quintile. The average number of children showed the same pattern with 2.6 children in the first quintile and 1.8 in the fifth quintile. The poverty rate for household with less than 4 persons is well below the national rate. 2.1 1. Household heads in the first quintile also were older then household heads in other quintiles. The average age of a household head was 55 in the first quintile compared to 48 in the fifth quintile (Table 2.3). The households headed by males who were living and working somewhere else also were characterized by being less likely to be poor. Among the first and second quintile, 14 percent and 15 percent of households were headed by migrating males, compared to 27 percent and 28 percent among the fourth and fifth quintiles. Health Outcomes 2.12. Lesotho’s health outcomes are much worse for the poor than the rich. Iinfant mortality, under-5 mortality, wasting, and stunting outcomes were all worse for the poor. For instance, the under-5 mortality rate was 114 for the poorest quintile, compared to 82 for the richest quintile. Significant intra-regional variation also existed. The variation in mortality rates between the capital, Maseru, and other rural parts of the country were as large as the spread across wealth. Anthropometric indicators such as wasting and stunting seemed more related to wealth. 2.13. Lesotho’s health outcomes were surprisingly weak in 2004 given the level of income (per capita GDP). Lesotho’s infant mortality rate ranked 13 of 28 among Sub-Saharan countries and 14 in severe stunting. This is disappointing considering Lesotho had a higher GDP per capita than most, Sub-Saharan countries. When comparing inequality in health outcomes, measured as the ratio between the health outcome in the lowest and highest quintile, Lesotho generally fared a little better than the level. Lesotho’s low to high ratio ranked 8 out of 28 in infant mortality, 7 in stunting, and 9 in female malnutrition (Table 2.4). These results are consistent with the co- dependence on location and wealth seen in Table 2.5. 25 2.14. Lesotho has one of the highest HIV infection rates in the world. Among Lesotho?s neighbors, only Swaziland and Botswana had higher prevalence rates. Even though HIV/AIDS infection rates were not worse for the poor, the impacts may be worse. The DHS 2004 surveys showed that urban areas, particularly women, had the highest rates. In urban areas, 33 percent of females and 22 percent of males were HIV positive. In rural areas, 22 percent of females and 18 percent of males were HIV positive (Table 2.5). 2.15. The HIV/AIDS pandemic caused a staggering high number of orphans. 4 1 percent of households in Lesotho were inhabited by at least 1 orphan in 2004. The burden of taking care of orphans (single and double) fell disproportionately on poorer households, in particular on female- headed households. This is partly because the HIV/Aids epidemic appears to be creating female headed households, as men were the first to be infected and the first to die. One out five children in Lesotho had lost their father in 2004. As a result, sixty-seven percent of all female-headed households had orphans, compared to only 26 of male-headed households. Female headed households in rural areas are more likely to .be poor (see section 2 below), so orphans are more likely to be in poor households. The rural areas, particularly the Sengu River Valley, had a very high number of households with orphans: almost every second household (48 percent) had at least 1 orphan. Thirty-two percent of households in the wealthiest quintile had an orphan, in comparison to 44 percent in the poorest quintile. Note that very few ?child-headed? households were found in the DHs data - it seems that orphans are still being cared for. Chapter 7 of this report provides a detailed picture of the infection rates, risk behavior, causes, and impacts HIV/AIDS epidemic, including or orphans, in Lesotho. Education 2.16. For adults, having less than complete primary education is highly correlated with poverty. Households with heads and spouses with no or little education were much more likely to be poor than better educated heads and spouses. In 2002-03, in the first quintile, 72 percent of heads of household had no completed education, compared to the 48 percent in the richest quintile. Similarly, 22 percent of household heads in the fifth quintile had secondary education or above, while only 4 percent in the lowest quintile had that much education (Table 2.3). 2.1 7. Wealth and school enrollment were strongly correlated in 2004, particularly for boys, suggesting that an intergenerational transmission of poverty is likely to occur. Lesotho made great progress in primary enrollment after it initiated its Free Primary policy in 2000. However, by 2004 the net enrollment rate was only 66 percent for boys and 82 for girls in the poorest quintile, compared to 88 percent and 87 percent for the richest quintiles (Table 2.7). The net primary enrollment rates were particularly low in Maseru and in the Mountains. 2.1 8. Enrollment in secondary education is still largely a privilege of the wealthy. In 2004 only 2 percent of boys and 8 percent of girls from the poorest quintile attended secondary education, while 38 and 55 percent of boys and girls respectively attended in the richest quintile. Secondary attendance rate was particular low for the Mountains (Table 2.7). c. INEQUALITY AND GROWTH 2.19. Inequality was high throughout the country in 2002/3. Lesotho?s Gini coefficient of inequality was a high 0.53. Decomposing the Theil Index, which also expresses inequality, showed that only 4 percent of overall inequality can be explained by the rural-urban divide. The remaining 96 percent was explained by intra-urban and intra-rural inequality. Looking at average consumption per adult equivalents, the inequality also could be viewed by the difference between 26 quintiles. Taking, for example, average consumption per adult equivalent per month of 33 and 557 in the first and fifth quintile, respectively, consumption in the fifth quintile was 17 times higher then the first quintile (Table 2.2). Box 2.2: Poverty and inequality, 2002/03 Poverty line Inequality $1 a day National Head Squared Theil’s Head count Gini count ga P ~~ Index National 0.37 0.54 0.17 0.53 0.53 Urbadrural Rural 0.4 1 0.58 0.19 0.5 1 0.49 Urban 0.25 0.40 0.1 1 0.52 0.54 Within-group inequality 95.6% Between-group 4.4% inequality Sources: HBS 2002/03 and authors’calculations. The calculations use a consumption aggregate per adult equivalent in which external migrants are excluded, and the poverty line constructed by BoS (BoS 2007). 2.20. The effect of economic growth impact on poverty is reduced by Lesotho’s high inequality. Lesotho has a very high level of consumption inequality-higher than most in the region (Table 2.2). As a general rule, growth will lead to a larger reduction in poverty if inequality is lower. Table 2.8 shows the a mathematical relationship between growth, inequality, and poverty reduction as calculated by Lopez and Serven (2006) with a poverty line similar to Lesotho’s (e.g. with 50 percent of the country in poverty). For instance, with the most recent (2002/03) estimate of the Gini coefficient at approximiately 0.5, Lesotho can expect to reduce the poverty head count by 1 percent in response to a 1 percent growth rate; a 3 percent reduction in poverty with 3 percent growth; and a 7.1 percent reduction with 7 percent growth. Earlier studies indicated a Gini coefficient in Lesotho of approximately 0.6. In that case, the 1 percent growth would have led to 0.7 percent reduction in poverty, and 7 percent growth to a 4.6 percent reduction in poverty. 2.21. If Lesotho had a lower level of inequality, for instance, a Gini of 0.4, then 1 percent growth would lead to 1.6 percent reduction in poverty. At this level of inequality, 3 percent growth would lead to 4.8 percent reduction in poverty, and 7 percent growth would lead to 11.2 percent poverty reduction.6 The examples show that too many people are simply too far away from the poverty line to escape poverty even if they were to gain as much as everybody else from growth. To sign8cantly reduce the poverty headcount, growth needs to be absolutely pro-poor in that it benejts the poor more than the average or the rich. Pro-poor growth’ in this case also would reduce the level of inequality, so it would have a more rapid effect on poverty reduction. D. CAUSES OF POVERTY 1 2.22. When asked, Basotho most often quoted agricultural problems such as drought and harsh weather conditions, unemployment (including retrenchment), and theft as the main causes of poverty (Table 2.9). As important inputs to the Poverty Reduction Strategy (PRS) and National Vision, extensive qualitative interviews were conducted in 200 communities throughout An important assumption in the examples is that the level of inequality is unaffected by growth. ’In the literature, several definitions of pro-poor growth exist. In this context absolute pro-poor growth is positive growth, with higher growth rates for the poor, which cause a redistributive effect. See for instance Ravallion (2004). 27 rural Lesotho.* Based on these interviews, a ranking of individual causes of poverty showed that theft was the most frequently mentioned cause, followed by unemployment, drought, and the retrenchment of mine workers. The two categories of “unemployment” and “retrenchment of mine workers” show the clear focus on lack of paid employment as a key cause of poverty. Repeated references to drought and harsh weather such as frost, snow and hail, as well as “poor farming” and “poor range management,” all point to the extent to which Basotho still perceive their livelihoods as agricultural and the causes of poverty as therefore rooted in problems of the agricultural sector. The role of political instability also is seen as significant. It was mentioned in over 25 percent of the consultations. (Leboela and Turner 2003). Box 2.3: Key causes of poverty according to village interviews, 2002 Agricultural problems 1 Employment, problems in the economy 2 Health, sanitation, water supply 3 Politics and governance 4 Crime, poor law enforcement 5 Poverty, hunger 6 Poor natural resource management, land 7 degradation Welfare, social problems 8 Education 9 Lack of equipment 10 Source: Leboela, and Turner, 2003 Notes: Answers have been merged and ranked according to most fremient RnSwer 2.23. In 2005 perception data indicated that the risk of theft diminished but that the availability of jobs had worsened. When asked about crime and violence, 57 percent said the situation was better in 2005 compared to a few years before; 18 percent thought the situation was the same, while 22 percent thought the situation has become worse. When asked about the availability of job opportunities, 90 percent said the situation was worse in 2005 compared to a few years before (71 percent answered much worse); 3 percent thought the situation was the same, whereas 6 percent thought the situation had improved (Afrobarometer 2005). 2.24. Statistics confirmed the trend of declining levels of livestock theft. From 2000-01 to 2003-04, there was a steady decline in number of animals reported stolen (Table 2.10). Livestock theft was described as heavily syndicated, with operatives being unemployed youth and retrenched migrant workers, who, it was argued, were driven by poverty. The 1999 poverty survey reported that 11 percent of households nationally, but a very high 29 percent in the Mountains, had lost animals to theft. The livestock syndicates operated across the border with South Africa and routinely were violent. Livestock often were taken at gunpoint, resulting in loss of life (Dzimba and Matooane 2005, Leduka and Gay 2007, Turner 2001). Multivariate Analysis 2.25. Multivariate analysis has the advantage over bivariate analysis that the former enables analysis of several variables at the same time to determine which variables are most correlated with consumption per adult equivalent once one also controls for the other variables. So, to more clearly identify the causes of poverty, we ran a regression on consumption per capita of a set of * Almost 300 visits took place, but only approximately 200 were complete enough to be used for analysis. Of the 300 visits, only 6 were in urban areas. 28 demographic variables and variables describing the economic activity of household members (the construction of the latter variables is described in chapter 3 on the labor force). The regressions are shown in Table 2.1 1, the significant variables in Figure 2.2, and the variables are described in tables 2.12 and 2.13. As we used a semi-log specification, the regression coefficients can be interpreted as the percent change in consumption which would occur from a 1 percent change in the variable. 2.26. The multivariate regression showed that employment-particularly nonfarming- was a key way out of poverty. Living in a rural household in which the head was employed with a salary or was an employer, raised consumption expenditure by 34 percent, compared to being a farmer. In contrast, the head being unemployed lowered consumption expenditure by 24 percent, compared to a farmer. In urban areas, the head being employed increased consumption expenditure by 28 percent compared to being a farmer. Having several sources of income also was important. Households in rural areas that had a second self-employed member had 27 percent higher consumption expenditure. Urban households that had either a salaried or a self-employed worker had 17 percent and 15 percent higher consumption, respectively. It is noteworthy that once we controlled for the employment of the head, education, and household size, having another family member working as a farmer did not measurably increase the consumption per adult equivalent. 2.27. Migration abroad seemed to be a successful livelihood strategy, while internal migration did not. Households that had a family member usually living abroad (most external migrants live in South Africa) had 14 percent higher consumption per adult equivalent in rural areas and 9 higher percent in urban areas, compared to households with no migrants. In rural areas, these numbers were in stark contrast to the households with internal migrants, who had 21 percent lower consumption compared to households that did not have any migrants. It may be that migrants come from lower income households, and the negative coefficient is picking up this effect. But it can still indicate that internal migration was not a very successful livelihood strategy that substantially increased the overall consumption level of the household. 2.28. The importance of education and the negative impact of a larger household size also were found in the multivariate analysis. Education was strongly correlated with consumption. The higher the education level, the higher the consumption. The advantage of being educated was greater in the urban areas then in the rural. For instance, rural households with heads who had completed primary education had 14 percent higher consumption than those whose heads could not readwrite. A corresponding urban household whose head had primary education had 36 percent higher consumption. Households heads with secondary education had 30 percent and 70 percent higher consumption per capita in rural and urban areas, respectively (compared to households with heads who could not readwrite). The negative impact of larger household size was a little larger in urban areas than in rural areas. 2.29. The “traditional household”-headed by a married man-was better off-at least compared to households headed by unmarried men. In rural areas, households with a male married head did better than female unmarried-headed households. These results also were found in the qualitative literature, in which widowed men and women (the majority of the unmarried) were considered a particularly vulnerable group. The number of widows and widowers is .~ increasing owing to HIV/Aids m ~ r t a l i t yWomen often do not remarry; especially as seen in the falling marriage rates. HBS data showed that in 1994-95, 10% of adults 15 years of age or above were widowed; in 2002-03, 12% were widowed. 29 Figure 2.1: Relationship between household characteristics and consumption per adult equiv. Male head - not married r Female head - not married I Household size I I Squared household size Head has no education I I Head has primary __ t I Head has secondary or above I Head Self-employedhonfarm I I Head employed with salary 1 Head not employed I HH had non head employed with salary I HH had internal migrant I I HH had external migrant Urban area other than Maseru I Rural Mountain 1 Rural Sengu River Valley 1 Source: HBS and authors calculations. Each bar shows the size of the estimated coefficient in the multivariate regression shown in Table 2.11. The dependent variable is the log of consumption per capita. Only significant variables are included in the figure. E. POVERTY TRENDS OVER TIME 2.30. Poverty can be measured in different ways. This section looks at the available data on consumption, income, and assets. Unfortunately we are unable to reach a clear conclusion on the trend, since there is a dire lack of time-consistent and high-quality data. 2.31. A trend in poverty measured by consumption is hard to establish as design differences make direct comparison of the two most recent H B S s impossible. The survey design changed in several ways between the two survey years, 1994-95 and 2002-03. Changes included the way products were coded in the diary and recall periods for several products. Appendix 1 details the differences between the two surveys and why they are not comparable. Since there are several differences, it also is not possible to tell whether the changes bias one survey up or down compared to the other. Experience from other countries has shown that even minor changes in survey design can significantly affect measured consumption and poverty (Scott and Amenuvegbe 1990, Deaton and Grosh in World Bank 2000a, Deaton and Kozel, 2005). 30 Box 2.4: Available data sources to track poverty over time Monetary poverty is usually measured by consumption or income. Two sets of surveys over the 1990s and early 2000s with income and consumption expenditure data at national level are available in Lesotho. They are the HBSs (1994-95 and 2002-03) from Lesotho Bureau of Statistics (BoS) and household surveys implemented by Sechaba Consultants in 1993 and 1999. This box briefly describes the quality and issues of comparability and sampling in these four surveys. It concludes that there yet are no data available which allow a confident comparison of poverty over time in Lesotho. Income Income data collected in developing countries is often of rather poor quality, in many cases so bad that the usefulness for tracking of poverty is limited - Lesotho is no exemption. Both the HBSs and the Sechaba surveys show indications of weak income data, with many households having no or very limited income that is far below their level of consumption. (World Bank 2000, Appendix 1). Consumption Expenditure Even minor changes in survey design have been shown to jeopardize comparability of consumption expenditure over time. Consumption expenditures in the Sechaba surveys were recorded by recall for the last 6 months for major expenditures on furniture, animals, clothing, school, health, building, gifts and feast. This measure is a very crude and limited measure as it only allow for eight categories and does not include expenditures on food. Food expenditures are normally a large part of consumption and a crucial element in the link between poverty - the ability to cover basic needs - and consumption expenditures. Hence the Sechaba surveys are also severely limited in their ability to track poverty over time, They are also not consistent or comparable with the HBS surveys, where comparable data was not collected as well (see Appendix). Sampling The HBSs are a stratified random sample, while Sechaba surveys used a mixed strategy. The Sechaba sample is based on a random sample of 130 enumeration areas chosen for a 1990 study using the 1986 census. These same 130 enumeration areas were revisited in 1993 and in 1999. Within these enumeration areas a new sample of households were chosen by going to the center of a selected village and then randomly choosing one of the four cardinal directions and interviewing households by a predetermined rule - for instance every fifth household. This kind of strategy is known as a random walk and avoids having to list the entire area. Ideally the starting point of the walk should also be random. The sample size of each survey can be seen below: Available surveys for tracking of poverty Survey Survey period Sample size HBS June 1994-May 1995 3184 HBS Nov. 2002-0ct. 2003 5992 Sechaba 1993 1700 Sechaba March-June 1999 3280 T h e Lesotho Vulnerability Assessment Committee (LVAC) releases a report every year on food security and vulnerability monitoring. The key components builds on the Household Economy Approach which based on location and wealth profiles try to estimate the food shortages that poor households will face, given shocks and livelihood responses. Though estimating food shortages every year the LVAC reports do not report time series on any data. This might be because the applied methodology has undergone several improvements over the recent years as change of boundaries and re-estimation of wealth groupings. For estimation of poverty on a national scale the methodology has limited scope as the definition of poor households varies with locations (to take into account local livelihood patterns) and the sample is limited (36 villages in 2007). 31 2.32. Ranking areas by poverty status might be more robust against design change; in ranking livelihood zones the HBSs showed relatively less poverty in the Mountains and the Sengu River Valley. The change in survey design certainly affected the level of measured consumption. For instance, education expenditure was measured with a four-week diary in 2002- 03, but as 12-month recall in 1994-95. Comparing education expenditure over time cannot be done (even by multiplying 1 year by 12), since the change in survey instrument would bias the result. Despite the two surveys not being directly comparable, the relative ranking of areas across the country might be more reliable for establishing a poverty trend. For example, is likely that the change in recall period did not impact the ranking of household education expenditure to the same degree, since each survey was internally consistent, because the same survey instrument was applied throughout the country. This is not to say that the change in instrument could not bias the ranking over time, it certainly could, but we believe that the bias would less and the ranking in each year therefore more robust than the comparison over time. 2.33. Ranking the ecological zones shows that the Rural Mountains and the Rural Sengu River Valley changed from being the poorest regions to being better off than the Rural Lowlands and Foothills (Table 2.14). The strong poverty reduction in period between the two surveys in the Mountains and Sengu River Valley areas can be explained by the massive investment in the Lesotho Highlands Water Project (LHWP). It provided both temporal employment income and much improved access to and from remote areas though new roads. Road construction for the Mohale Dam started in 1998 and was completed in 2002. The road construction therefore overlapped with the 1994195-2002/03 period covered by the surveys. The dam is estimated to have created the equivalent of 20,000 person-years of employment in an area populated with approximately 500,000 people. The dam thus led to an estimated M294 million flowing into the local economy through associated wage earnings. In addition to the large quantity of temporal work possibilities, the project included a number of compensation schemes10 and investment in infrastructure-particularly roads but also schools, clinics, and water and sanitation. 2.34. Privately prepared national surveys measuring income, consumption expenditure, and wealth show mixed trends from 1993 to 1999.” The surveys found decreasing levels of poverty when measuring poverty by income,l2 marginal change in poverty when measuring poverty by consumption, and indication of increasing poverty as wealth measured by assets decreased. Income measured by recall showed a 14 percent increase in average real income per capita between 1993 and 1999. Using a fixed poverty line of M80 per capita in 1999, 65 percent of the population was considered poor, compared to 71 percent in 1993. The poverty line is taken from a 1986 study, was corrected for inflation,l3 and was established as the income needed to cover basic food and nonfood needs. Consumption expenditure was recorded by recall for major expenditures on furniture, animals, clothing, school, health, building, gifts, and feasts for the last six months. Using this measure, which does not include food, and the same M80 per capita fixed poverty line, the surveys found that 91 percent of the population consumed under M80 in 1993 lo The compensation schemes had a number of different programs including individual and communal food and cash compensation, infrastructure investment in schools, clinics, water and sanitation, land compensation, and training programs. ” See additional information about the survey in box 2.2, “Available data sources to track poverty.” ’*The 1993 expenditure data was incomplete and covered only a limited number of expenditure items. Some households did not report any expenditure. The income data also was incomplete, and the average household income was much lower than expected (World Bank 2000). l3 The poverty line of M80 measured in 1999 prices corresponded to M106 in 2002-03 prices, which is in between the M84 food poverty line and M150 food and nonfood poverty line established in the HBS 2002-03. Correction for inflation was done using the IMF’s 2006 World Economic Outlook database. 32 and 92 percent consumed under M80 in 1999. Finally, the surveys also recorded wealth by ownership of the following assets: ploughs, planters, scotch carts, tractors, cars, babies, television sets, radios, sewing machines, and stoves valued at current retail value. Using this wealth measure, the survey showed that 23 percent of households had assets valued at less than M320 in 1993 (in 1999 prices), whereas almost twice as m a n y 4 4 percent of households-had values less than M320 in 1999 (Gay and Hall 2002). 2.35. Measures of wealth improved sluggishly between 1994 and 2004 (Figures 2.3 and 2.4). Over this period, up to nine surveys measured indicators of wealth that are roughly comparable. Based on these data, housing characteristics such as improved roofing and access to piped water on the premises both showed a slow and marginal improvement. This trend also is seen in ownership of assets such as bicycles and refrigerators. Other assets including radios and televisions indicated a downward trend in average ownership. Access to toilet facilities was the only household characteristic included here that showed consistent improvement over time. Figure 2.3 Trend in Housing Characteristics 1994 - 2004 70% rn ~ __--- 80% - _ -_ _ _ _ - - A ! - -_ _ _ -_ ___-_ --______------ A --A 1 A I I I 1 rn Improved Roof l o Toilet A ACCESS l ox: 30% *Piped water on premls 20% - 0 10% ..... 0 ................................. .............~ . . . . - . ” - “ ~ ’ 0 0 0% 1993 1985 1997 1999 2001 2003 2005 Year Sources: 1993 (Sechaba 2000); 1994 (HBS 1994/95 and authors’ calculations); 1996 (BoS 1996); 1999 (Sechaba 2000); 2000 (BoS 2000); 2001 (BoS 2001); 2002 (BoS 2002); 2003 (HBS 2002/03 and authors’ calculations); 2004 (DHS 2004). Notes: The question asked varied by survey. Some statistics were household-level while others were population. The trend lines are linear. 33 Figure 2.4 Trend in household ownership of assets, 1994-2004 (Oh) 80.0% - 16.0% Trend line Radio (leflaxis) 70.0% \ -- 14.0% A 60.0% -- 12.0% 50.0% -- 10.0% Trend line Refrigerator (right axis) 40.0% -- 8.0% Trend line TV (right axis) 30.0% \ -- 6.0% 20.0% 10.0% 0.0% 0.0% Sources: 1993 (Sechaba 2000); 1994 (HBS 1994/95 and authors' calculations); 1996 (BoS 1996); 1999 (Sechaba 2000); 2000 (BoS 2000); 2001 (BoS 2001); 2002 (BoS 2002); 2003 (HBS 2002/03 and authors' calculations); 2004 (DHS 2004). Notes: The figure shows percent ownership of asset.The question asked varied by survey. Some statistics were household-level while others were population. The trend lines are moving averages based on two periods. 2.36. Studies suggest that traditional sharing mechanisms and informal safety nets were coming under pressure and becoming more exclusive, but that they were still holding up and did not allow absolute destitution (Turner 2003, Magrath 2004, Boehm 2004, Leduka and Gay 2007). The demand from the various vulnerable groups was continuously increasing. However, the conclusions from different village studies and reviews seem to be that some degree of mutual social protection remained; complete social exclusion was uncommon; and "absolute destitution is not allowed to occur" (Turner 2004:54, Leduka and Gay 2007). Turner suggests, however, that then-current sharing mechanisms to some degree depended on past mining earnings, savings, and pensions, which in the medium and longer term could become exhausted. Hence, a new and more serious level of vulnerability could arise among many groups and localities. 2.37. Limited panel data from 1993 to 2002 show that 26 percent of rural households were chronically income poor. Unfortunately, very few panel data sets exist in Lesotho and none of them came from nationally representative surveys. One of the few existing panel studies used a small, nonrandom rural panel and found that 42 percent of households changed poverty status from 1993 to 2002: 14 changed from poor to non-poor whereas double that number (28) changed 34 from non-poor to poor. (Wason and Hall 2004).14That panel study also showed that, over the 9 years, 58 percent stayed at the same level: 26 percent were chronically poor, while 32 percent had never been poor. In comparison, 24 percent were found to be chronically poor in Uganda over 4 years, and 22 percent were chronically poor in South Africa over 5 years (Table 2.15) F. CHARACTERISTICS OF THE POOR: TRENDS OVER TIME: 2.38. Substantially more Basotho had completed primary education in 2002-03 than in 1994-95. In 1994-95, 10 percent of household heads and 24 percent of spouses had completed primary. In 2002-03, 25 percent of heads and 41 percent of spouses had completed primary, corresponding to a 15 percentage point and 17 percent point increase in the share of heads and spouses with primary education, respectively. The increase in heads and spouses with completed primary took place in both urban and rural areas (Table 2.16).15 Unfortunately, there was limited progress in the share of heads and spouses with completed secondary or above. Both heads and spouses saw a 1 percentage point increase in the share with secondary or above (Table 2.17). 2.39. With rising levels of education, households with heads with no education were even more concentrated in the bottom quintiles. In 1994-95,39 percent and 86 percent of urban and rural heads of households, respectively, had no education. By 2002-03, 33 percent and 74 percent of urban and rural heads of households had no education. This shows a reduction of 6 and 12 percentage points of household heads with no education in urban and rural areas, respectively. At the same time, 43 percent of rural household heads with no education in 1994-95 were found in the 2 lowest quintiles. In 2002-03, 51 percent of these households were found in the 2 lowest quintiles. In urban areas, the concentration of uneducated households at the bottom of the distribution was even stronger: 57 percent were found in the bottom 2 quintiles in 1994-95, compared to 67 percent in 2002-03 (Figure 2.5). 2.40. Health outcomes show only setbacks over time. Nutritional outcomes measured through stunting and wasting over 1976-2007 show no progress.'6 Other health indicators including under-5 and infant mortality show a steady increase from 1992 to 2004. In 1992 infant mortality and under-5 mortality were 72 and 86, respectively; by 2004 they had increased to 91 and 113 (Table 2.18). I 4 The study found that 66 percent of the intended sample and likely suffered from some biases because the most remote households were not interviewed in the second round, and no effort was made to track individuals or households that moved. I5 Close reading of table 2.16 reveals that the share of heads and spouses with completed secondary or above in both urban and rural areas fell between 1994-95 and 2002-03. These numbers, however, should not lead to the conclusion that fewer heads and spouses in the population completed secondary or above. There were approximately 1 percentage point more heads and spouses with completed secondary or above in 2002-03 than in 1994-95 (table 2.17). The result from table 2.16 in fact results from migration from rural to urban areas. Migration of heads and spouses both above and below secondary level means changes in both the denominator and numerator, which in this case led to lower ratios in both urban and rural areas. I 6 See more detail on anthropometric indicators in chapter 4 on food security. 35 Figure 2.5 Distribution of uneducated across quintiles, 1994/94 and 2002/03 (YO) Rural Urban 7" 35 *" 25 20 15 10 5 0 q1 92 93 q4 q5 10 1994194 120021031 /01994194 120021031 Sources: HBSs and authors' calculations. Summing Up 2.41. Poverty in Lesotho is characterized by low income and worse health and education outcomes. Geographical isolation is a key variable in both monetary poverty and level of well being in the household. The profile of the poor showed a familiar pattern: larger households and households with little or marginal relationship to the labor market were worse off.. The main causes of poverty seemed primarily to be-related to economics and included lack of earners with full time work, lack of human capital among the adults in the household - which usually resulted in lower incomes - and shocks such as droughts. These issues are explored in more detail in the next chapters. 2.42. Better public policy needs better data on baseline levels of living and livelihoods, and trends in household behavior, reactions to events, and the consequences of these decisions. The most recent, and highest quality data is in the DHS survey. This survey was funded by donors and implemented by a private contractor. This could be a useful model for the next multi-purpose household survey. 2.43. The lack of comparable consumption o r income data over time severely limits the analysis of trends in, and the determinants of monetary poverty. Nonmonetary indicators such as assets, health, and education show divergent trends. Health indicators and ownership of assets do not seem to have improved. In contrast, education indicators such as enrollment and percentage of people with primary education increased. 36 Tables Chapter I1 Table 2.1: Poverty and Inequality, 2002/03 Poverty line Inequality $1 a day National rnrint Squared National 0.37 0.54 0.17 0.53 0.53 UrbadRural Rural 0.4 1 0.58 0.19 0.51 0.49 Urban 0.25 0.40 0.1 1 0.52 0.54 Within-group inequality 95.6% Between-group inequality 4.4% Region Maseru Urban 0.20 0.32 0.08 Other Urban 0.28 0.44 0.12 Rural Lowlands 0.41 0.59 0.19 Rural Foothills 0.47 0.63 0.22 Rural Mountain 0.39 0.54 0.18 Rural Senqu Valley 0.36 0.54 0.17 Sources: HBS 2002-03 and authors’ calculations. The calculations use a consumption aggregate per adult equivalent that excludes external migrants and the poverty line constructed by BoS (BoS 2007). Table 2.2: Inequality in Sub-Saharan Africa Country Year Richest 10%- poorest 10% Gini coefficient Ghana 1998-99 14 41 Kenya 1997 14 43 Uganda 1999 15 43 Nigeria 2003 18 44 Madagascar 200 1 19 48 Gambia 1998 20 50 Lesotho 2002-03 14 53 South Africa 2000 33 58 Swaziland , 1994 50 61 Botswana 1993 78 63 Namibia 1993 129 74 Sources: UNDP Human Development Report (HDR) and BoS Household Budget Survey (HBS), various years. 37 Table 2.3: Characteristics of population by consumption quintile, 2002/03 National Head of household (%) 41 42 43 44 45 laveragej Male head present 54 53 48 46 45 49 Male head away 14 15 21 27 28 21 Female 33 32 31 26 26 30 Demographics Household size 6.4 6.1 5.9 5.6 4.7 5.7 Average number of adults 3.9 3.7 3.7 3.5 3.1 3.6 Average number of adult excluding members abroad 3.5 3.4 3.2 3.1 2.7 3.2 Average number of children 2.6 2.5 2.4 2.3 1.8 2.3 Average number of elder 0.3 0.4 0.3 0.3 0.2 0.3 Age of head and spouse Average age of head present 55 52 50 50 48 52 Average age of heads abroad 46 46 45 44 42 44 Average age of spouse 44 42 40 40 37 40 Education o f head @ ) Heads with no education 72 71 69 58 48 Heads with primary 23 21 22 29 29 Heads with secondary or above 4 8 9 13 22 Education o f spouse (%) Spouses with no education 55 53 50 38 31 44 Spouses with primary 40 39 38 46 42 41 Spouses with secondary or above 5 8 12 16 27 14 Location - % population Rural 86 81 78 73 64 76 Maseru 4 6 8 10 17 9 Other Urban 10 13 14 17 20 15 Lowlands 45 40 43 38 29 39 Foothills 16 17 12 10 10 13 Mountain 21 21 20 21 19 20 Sengu River Valley 5 4 4 4 5 ' . 4 Living conditions - % with access to Toilet 46 53 57 61 63 56 Improved roof 62 66 71 72 77 70 Electricity 4 6 6 9 17 8 Improved water 86 83 86 87 85 86 Average consumption expenditure per adult equivalent 33 77 130 216 557 202 Average consumption expenditure per adult equivalent as % o f the third quintile ~~ 25 59 100 167 430 Sources: HBS 2002103 and authors' calculations. Children are aged 0-14, adults 15-65, and elder over 65. Members abroad are considered part of the household that lives abroad. The quintiles of consumption are based on consumption per adult equivalent, in which household members living abroad are excluded and their reported consumption also is excluded from the consumption expenditure aggregate. 38 Table 2.4: International comparison of health indicators Infant mortality Severe stunting Female malnutrition GDP per Low to capita National Low to National Low to National average high ratio average high ratio average high (PPP Year ratio 2005) Ghana 2003 65 1.1 9 3.8 9 3.3 1,225 Kenya 2003 76 1.5 11 2.3 12 4.6 1,359 Lesotho 2004 91 1.3 16 2.2 6 1.8 1,415 Namibia 2000 40 1.6 7 1.3 na na 4,547 Tanzania 2004 83 1.4 12 4.7 9 1.6 1,018 Uganda 2000/01 89 1.8 15 2.6 10 3.1 99 1 Zambia 2001 94 2 22 2.2 15 2 1,175 Source: World Bank, “Socio-Economic Differences in Health, Nutrition, and Population: Cross-country Report” 2007, Lesotho Demographic Health Survey, 2004 Notes: Low to high ratio is the value in the fifth quintile divided by the value in the first quintile. Infant mortality: Number of deaths to children under 12 months per 1,000 live births, based on experience during the preceding 10 years. Severe stunting (height-for-age): % of children aged approximately 3 years old (varies by survey) with a height-for-age Z- score of below -3 SD of the median reference standard for their ages. Female malnutrition: % of women aged 15-29 years with a Body Mass Index of less than 18.5, where BMI is defined as weight in kilograms divided by square of height in meters. LowesUhighest ratio is highest quintile/lowest quintile. GDP per capita ppp is from World Bank 2008. Table 2.5: Health outcomes across wealth, location and gender, 2004 H IV/AID S Infant . Under-5 Severe Severe Infection rate Orphans mortality mortality stunting wasting Male Female Wealth quintiles Poorest 83 109 17 1.3 18 20 44 Second ‘93 110 21 1.3 16 28 43 Third 97 116 16 1.2 23 25 45 Fourth 94 117 10 0.8 21 27 42 Richest 83 103 9 1.2 14 29 32 Urbadrural Urban 82 118 11 1.1 22 33 37 Rural 91 109.7 16 1.2 18 24 42 Region Maseru Urban 70 101 13 19 33 37 Other Urban 93 130 10 1.5 26 34 37 Rural Lowlands 91 105 11.2 1 19 25 42 Rural Foothills 75 82 17 1.1 17 24 40 Rural Mountain 98 122 18 1.4 17 23 42 Rural Senqu River 98 115 18 4 16 25 48 Valley Household head Female 92 124 13 1.4 15 32 67 Male ~ ~~~~ 90 107 15 1.4 20 23 26 Sources: DHS 2004 and authors’ calculations. Notes:All calculations regarding infant mortality and under-5 mortality refer to 2004. Severe stunting is height for age: YOof children (under 6) with a height for age Z score below 3 SD. Severe wasting is weight for height: % of children (under 6) with a weight for age Z score below 3 SD. Orphan: % households with father or mother orphan or both. 39 Table 2.6: HIV/AIDS prevalence and treatment: International comparison Adult (15-49 yrs) HIV positive O r p h a n s (0-17 yrs) Country prevalence (%) receiving A R T (%) due to AIDS Botswana 24.1 85 120,000 Namibia 19.6 35 85,000 Swaziland 33.4 31 63,000 Zambia 17.0 27 710,000 South Africa 18.8 21 1,200,000 Malawi 14.1 20 550,000 Lesotho 23.2 14 97,000 Mozambique 16.1 9 5 10,000 Zimbabwe 20.1 8 1, I 00,000 Angola ’ 3.7 6 160,000 Central African 10.7 3 140,000 ReDublic Sources: UNAIDS 2006. Note: A child under 17 is considered an orphan if at least 1 parent is deceased. According to administrative data the . percentage receiving ART has increased to 44percent in 2007 in Lesotho. Table 2.7: Educational attendance rates across wealth, location, and gender, 2004 Secondary net Primary net attendance rate attendance rate Boys Girls Boys Girls Wealth quintiles Poorest 66 82 2 8 Second 78 86 6 16 Third 87 86 15 28 Fourth 87 92 18 33 Richest 88 87 38 55 Urbadrural Urban 81 87 37 47 Rural 81 87 13 25 Region Maseru Urban 80 82 34 48 Other Urban 90 93 35 46 Rural Lowlands 87 88 16 30 Rural Foothills 84 87 12 20 Rural Mountain 77 84 5 16 Rural Senqu River Valley 87 88 12 27 Household head Male 82 86 17 31 Female 79 88 14 25 Sources: DHS 2004 and authors’ calculations. Notes: Net primary attendance rate = the proportion of children aged 6 to 12 attending primary. Secondary net attendance rate = children aged 13 to 17 attending secondary. 40 Table 2.8: Reduction in poverty as function of inequality and growth Growth rate Gini Coeflcient 1 3 5 7 0.3 -2.8 -8.4 -14.0 - 19.6 0.4 -1.6 -4.8 -8 .o -11.2 0.5 -1.0 -3 .O -5.1 -7.1 0.6 -0.7 . -2.0 -3.3 -4.6 Sources: Lopez and Serven (2006) and authors' calculations. Notes: Using the theoretical growth elasticities from Lopez and Serven and assuming a poverty line of half of mean consumption. This poverty line is equivalent to 100 maloti per adult equivalent . The $1 a day poverty line is equivalent to 98.94 and between Lesotho's food poverty line of 84.4 and food and nonfood poverty line of 149.9. Note that the model builds on the assumption of income or consumption being lognormal. Lopez and Serven tested this assumption on a large cross-country data set and found that per capita income is generally lognormally distributed, whereas consumption expenditure per capita is not. Table 2.9: Key causes of poverty from village interviews, 2002 No. of times issue Consultations issue mentioned mentioned (%) Theft 60 47.6 Unemployment 57 45.2 Drought 51 40.5 Retrenchment of mine workers 50 39.7 Harsh weather conditions 47 37.3 Poor farming 45 35.7 Political instability 35 27.8 Disease 24 19.0 Lack of improved farming inputs 21 16.7 Poor range management, overgrazing 20 15.9 Malnutrition 19 15.1 Lack of concern for others 18 14.3 High population growth 16 12.7 Lack of agricultural extension 14 11.1 Lack of equipment 14 ' 11.1 Lack of money 14 11.1 Source: Leboela and Turner 2003. Table 2.10: Stock theft 2000/01-2003/04 Year Number of animals stolen Number of animals recovered Recovery (?A) 2000-0 1 33,950 11,150 33 2001-02 30,105 11,074 36 2002-03 26,678 13,369 50 2003-04 18,442 7,847 43 Total 109,268 41,862 38 Source: Dzimba and Matooane 2005. 41 Table 2.11: Regressions of household characteristics and consumption per adult equivalent, 2002/03 Rural Urban Gender and marital status o f houseliold head (ase is male head - married) Male head - not married -0.27 -0.23 (4.21) * * * (4.13) *** Female head - married -0.07 -0.05 (0.42) (0.66) Female head - not married -0.09 -0.04 (1.95)* (1.09) Household size Household size - excl. migrants -0.33 -0.40 (13.20) * * * (16.24)*** Squared household size 0.02 0.02 (8.82)* * * (1 0.25) * * * Education of head (base is head with no education) 0.06 0.21 Head has no education, but can readwrite (1.45) (3.04)* * * Head has primary 0.14 0.36 (2;73)* * * (5.20)* * * Head has secondary or higher 0.30 0.70 (3.11)*** (9.66) ** * Activity of household head (base is farming) Employed with salary or employer 0.34 0.28 (6.45) * * * (4.81)* * * Self-employednonfarm 0.20 0.18 (2.05)* * (2.65)* * * 0.10 -0.06 Other nonfarm (1.54) (0.82) Nonemployed -0.24 0 (3.77)* * * (0.02) Activity of other household members (excluding head) Farming 0.04 0.07 (0 .87) (1.44) Employed with salary or employer 0.05 0.17 (1.07) (4.25)* * * Self-employedhonfarm 0.27 0.15 (2.41)* * (2.40) * * Other nonfarm -0.06 -0.04 (1.17) (0.84) Migration Household had members living periodically -0.02 -0.21 outside household in Lesotho (4.54) * * * (0.48) Household had members living periodically abroad 0.14 0.09 (2.96)* * * (1.87)* Location dummies (Maserufor urban and Foothills for rural excluded), Urban area other than Maseru -0.20 (5.09) * * * -0.02 Rural Lowlands (0.34) Rural Foothills Rural Mountain 0.21 (4.02)* * * Rural Sengu River Valley 0.12 42 Table 2.11: Regressions of household characteristics and consumption per adult equivalent, 2002/03 Rural Urban (1.78)" 5.75 6.10 Constant (66.18)* * * (59.69) * * * Number of observations 2930 3063 Adjusted R2 0.17 0.26 Sources: HBS and authors' calculations. Notes: The dependent variable for the OLS regressions is the consumption per adult equivalent, when external migrants are excluded. Absolute value o f t statistics in parentheses. * = significant at 10%; ** = significant at 5%; *** = significant at 1%. The excluded variables are: Male headed household-"married"; Household Head Education-"no education"; Household head activity - "Farming"; Livelihood Zone Dummies - "Urban Maseru'' for urban regressions, "Rural Foothills" for rural regression. Table 2.12: Description of regression variables: Urban, 2002/03 Mean Std. dev. Consumption expenditure ,per adult equivalent 5.43 1.05 Household head Male head - married 0.50 0.50 Male head - not married 0.12 0.32 Female Head - married 0.05 0.21 Female Head - not married 0.33 0.47 Education o f household head No education 0.06 0.24 No education but can readwrite 0.26 0.44 Has primary 0.35 0.48 Has secondary or above 0.32 0.47 Employment status o f household head Farming 0.12 0.32 Employed with salary or employer 0.52 0.50 Self-employednonfarm 0.12 0.33 Other nonfarm 0.09 0.29 Not employed 0.15 0.36 Household size Size 4.02 2.24 Size squared 21.14 22.87 Activity of other household members (excluding head) Farming 0.16 0.36 Employed with salary or employer 0.35 0.48 Self-employednonfarm 0.08 0.28 Other nonfarm 0.19 0.39 Not employed 0.67 0.47 Migration Household had members living periodically outside HH in Lesotho 0.17 0.38 Household had members living periodically abroad 0.15 0.36 Livelihood zone dummies Other urban 0.62 0.48 Maseru 0.38 0.48 ~~ Note: Mean and std. dev. are weighted according to survey. 43 Table 2.13: Description of regression variables: Rural, 2002/03 ~ ~ ~~~~ Independent Mean Std. dev. Consumption expenditure per adult equivalent 4.90 1.06 Gender and marital status o f household head Male head - married 0.56 0.50 Male head - not married 0.12 0.32 Female head - married 0.01 0.1 1 Female head - not married 0.3 1 0.46 Education of Head No education 0.29 0.45 No education but can readwrite 0.45 0.50 Has primary 0.22 0.4 1 Has secondary or above 0.05 0.21 Employment status o f head Farming 0.52 0.50 Employed with salary or employer 0.25 0.43 Self-employedhonfarm 0.04 0.19 Other nonfarm 0.09 0.29 Not employed 0.10 0.30 Household size Size 4.65 2.48 Size squared 27.80 29.0 1 Activity of other household members (excluding head) Farming 0.55 0.50 Employed with salary or employer 0.20 0.40 Self-employednonfarm 0.03 0.16 Other nonfarm 0.20 0.40 Not employed 0.64 0.48 Migration Household had members living periodically outside household in Lesotho 0.2 1 0.41 Household had members living periodically abroad 0.27 0.44 Livelihood zone dummies Rural Sengu River Valley 0.06 0.23 Rural Mountain 0.26 0.44 Rural Foothills 0.16 0.37 Rural Lowlands 0.52 0.50 Note: Mean and std. dev are weighted according to survey. 44 Table 2.14: Poverty by geographic region in 199495 and 2002103 (YO) Distribution of Poverty headcount rate Distribution of the poor population 199419 200210 Chang 199419 200210 chang 199419 200210 Chang 5 3 e 5 3 e 5 3 e Urbandrur al Urban 36.7 41.5 4.8 5.3 16.2 10.9 9.7 22.0 12.4 Rural 69.8 60.5 -9.2 94.7 83.8 -10.9 90.3 78.0 -12.4 Region Maseru 1.8 6.7 8.4 1.7 32.1 33.7 1.6 3.2 5.0 Urban Other 3.3 11.2 7.9 5.5 13.6 8.2 40.7 46.3 5.6 urban Rural 35.1 42.1 7.1 36.0 38.0 2.1 64.9 62.4 -2.5 Lowlands Rural -8.6 22.3 13.1 -9.3 71.1 65.8 -5.3 23.8 15.2 Foothills Rural 18.6 22.0 3.4 80.9 55.0 -25.9 22.6 21.5 -1.1 Mountain Rural Senqu 10.9 4.9 -6.0 72.7 57.0 -15.7 11.9 5 .O -7.0 River Valley Total 66.6 56.3 -10.2 100.0 100.0 0.0 100.0 100.0 0.0 Sources: HBSs and authors’ calculations. Notes: Calculations use the consumption aggregates and poverty lines constructed by BoS (BoS 2007). Table 2.15: Poverty transitions over time (%) Country Chronically poor Transient Nonpoor Years Kwazulu-Natal, South Africa 22 31 47 1993-1998 Uganda 24 20 56 1992193-1995196 Riral Lesotho 26 42 32 1993-2002 Sources: Okidi and Mugambe (2002) for Uganda, Carter and May (1999) for South Africa, and Wason and Hall (2004) for Lesotho. 45 Table 2.16: Educational characteristics of population by wealth quintiles and location, 1994-95 and 2002-03 (YO) Rural Q1 43 Q5 National 1995 2003 1995 2003 1995 2003 1995 2003 Heads education No education 98 87 91 73 62 57 86 74 Primary 1 12 8 25 13 31 8 21 Secondary or above 0 1 1 3 24 12 6 5 Spouse education No education 83 70 70 55 42 35 69 52 Primary 16 29 25 41 34 47 23 40 Secondary or above 2 1 5 4 25 18 8 8 Urban Q1 Q3 Q5 National 1995 2003 1995 2003 1995 2003 1995 2003 Heads education No education 47 57 38 28 31 11 39 33 Primary 27 34 17 40 21 24 21 35 Secondary or above 27 9 45 33 48 65 40 32 Spouse education No education 32 41 23 20 26 9 27 24 Primary 42 49 29 52 24 28 27 43 Secondary or above 26 9 49 28 52 63 47 34 Sources: HBS 1994-95 and 2002-03 and authors’ calculations. Members abroad are individuals are considered a part of the household but that lives abroad. Quintiles of wealth are based on asset indices calculated for the two surveys. Table 2.17: Education of heads of households and spouses, 1994-95 and 2002-03 (%) 1994-95 2002-03 Change Head no education 80 64 -16 Head primary 10 25 15 Head secondary or above 10 12 1 100 100 0 Spouse no education 65 46 -19 Spouse primary 24 41 17 Spouse secondary or above 12 14 1 100 100 0 Sources: HBS 1994-95 and 2002-03 and authors’ calculations. Table 2.18: Health and nutritional outcomes, 1976-2007 1976 1992 2000 2002 2004 2006 2007 Stunting 23 33 45 47 38 38 41 Underweight 22 16 18 22 20 20 20 Wasting 3 2 5 12 4 4 6 Infant mortality 72 79 91 Under-5 mortality 86 111 113 Sources: Nutritional outcome: 2000 EMICS, 2002 CWIQ, 2004 DHS, 2006 FPLVAC CHS draft report, simple average of figures for beneficiary and non beneficiary households, 2007 CHS WFP/LVAC/FNCO nutrition survey. Mortality: 1992 and 2004 DHS 2004. Table 2.19: Poverty across demographics, 2002/03 Poverty head count Severity of poverty Population share Gender Female headed households 0.58 0.19 0.30 Male headed households 0.52 0.17 0.70 All females 0.53 0.18 0.5 1 All males 0.55 0.17 0.49 Household size 1 0.26 0.05 0.03 2 0.34 0.10 0.05 3 0.40 0.1 1 0.10 4 0.49 0.14 0.15 5-7 0.53 0.17 0.32 7 and above 0.66 0.23 0.34 Age Children aged 0 - 15 0.56 0.18 0.38 Adults 16 - 64 0.52 0.17 0.56 Elder 64 and above 0.56 0.18 0.06 Marital status Never married 0.56 0.18 0.59 Married 0.50 0.16 0.29 Divorced 0.55 0.19 0.03 Widowed 0.54 0.17 0.08 Living together 0.43 0.05 0.00 National 0.54 0.17 100 Source: HBS 2002/03 and authors’ calculations Annex to Chapter 2: Definition of household members for Poverty estimation 2.44. The strong Basotho tradition and common practice of migrating to other parts of the country and abroad for extended periods makes establishing the true number of people sharing consumption in the household difficult. According to HBS 2002-03, approximately 13 percent of the population lived for some time during the year elsewhere than in their family households. Roughly 50 percent of these people lived within Lesotho, the other half abroad.17 2.45. HBS poverty estimates are based on consumption per adult equivalent. This measure is calculated as total consumption in the household divided by the number of adult equivalents in the household. In other words, it estimates how much each household member consumes while taking into account household composition and different needs of household members. 18 Thus, a key factor in estimating poverty is determining who is considered a part of the household: how many individuals share the household?s consumption? 2.46. The HBS establishes members of households as people ?who are sharing most of their economic resources.? The survey continues with directions to include household members who are temporarily absent and do not share their economy with another household. This definition is extended to include absent members living abroad. This definition is in line with one of the internationally guiding principles of surveying households: that any individual can belong to only one household so cannot be counted twice in the survey. However, this definition alone is not sufficient for defining a household, since long-absent household members usually do not consume in the household. Including these absent members in the denominator when calculating consumption per adult equivalent would give a negative bias to households that have absent family members. 2.47. In a situation where migration is important, questions that enable a more exact determination of whether members should be included in the household and whether household members significantly take part in household consumption are usually included. 2.48. HBS did not include these questions. They did included questions on how much members abroad consumed away from the household. While theoretically a plausible solution to establishing a true estimate of consumption per adult equivalent for the household, this approach was not particularly successful. 1. The question was asked only of households that had members living abroad, not of households that had absent members living within Lesotho. 2. The question is very difficult to answer exactly. The person consuming is not present, so other household members may know little about his or her consumption pattern. 3. Only 10 percent of households that had a member living abroad answered the question. 2.49. Table A . l shows the difference in estimates of poverty if we assume that household members not living in the household do not take part in the household?s consumption. The national head count falls to 53.6 if we exclude household members living abroad, and to 49.9 if ? The 2004 DHS asked a similar set of questions and found that approximately 13% of the population lived away from their households outside Lesotho. An additional 6% lived away from the household within Lesotho. ?*BoS 2007 explains the concept in more depth and shows how it is calculated. 48 we exclude household members living abroad and in other places in Lesotho.’’ This 4 percent difference shows that whether household members living elsewhere were considered to take part in normal everyday consumption in the household has a major effect on the estimates of poverty. For future household consumption surveys,.it is recommended to include the additional questions that would enable a more exact assessment of who consumes how often in each household. Table A.l Poverty headcount rates by definition of household size (%) Excluding household Excluding household members living outside members living outside All household members the household- within the household abroad Lesotho and abroad only National 56.3 49.9 53.6 Maseru Urban 33.7 24.9 27.8 Other Urban 46.3 39.7 42.8 Rural Lowlands 63 .O 56.1 59.5 Rural Foothills 66.8 58.9 63.1 Rural Mountain 57.8 50.3 55.7 Rural Senqu River 58.6 56.2 58.4 Valley Sources: HBS 2002-03 and authors’ calculations. l9 Removing members from the household can impact the estimation of the nonfood component of the poverty line, but minimally. Ten percent of households that had members abroad estimated the overall consumption of these members where they lived. These estimates were excluded from the consumption aggregate as they were a poor measure. Very few household estimated it in any case. 49 3. IMPACT OF GROWTH ON POVERTY 3.1. Structural development of the economy is fundamental for poverty reduction. Sustainable poverty reduction comes from shifts in the structure of production from low- productivity sectors units to those with higher productivity associated with labor mobility by the poor into these higher-productivity sectors. This production and employment shift should produce growth not only in aggregate income but also in private consumption reaching the poorest household. In other words, labor mobility brings to the poor changes in the structure of their household incomes which enable a higher, and more secure, level of monetary welfare (consumption), as well as investment in human capital and assets. 3.2. In the previous chapter, analysis showed that the livelihood pattern of the household was closely correlated with household welfare. In this chapter, the development of household livelihoods is related to Lesotho’s structural development processes. We begin at the macro level and then move to an analysis of the structure of the labor market, including labor demand and supply. In the final section, the implications of labor market outcome for household welfare are analyzed. 3.3. The analysis shows that the structural change in Lesotho brought about by the growth of domestic industry benefited a large share of Basotho households, especially in urban areas, by providing steady wage employment for at least one household member. But it did not contribute as much to rural livelihoods. Rural households with remittances are still well off, as are some households engaged primarily in agriculture. And the development of the highlands through the water project helped this area. But the backward linkages from the wage economy in urban areas to rural households appear weak based on available data. Opportunities for women in rural areas are still limited. A. STRUCTURAL CHANGE IN LESOTHO: ENVIRONMENT FOR POVERTY REDUCTION, 1994-2005 3.4. Since 1994, Lesotho’s domestic economy has grown rapidly-an estimated 3.5 percent per annum-and transformed, shifting value-added out of the agriculture and livestock sector (Figure 3.1). The growth engine has been primarily the industrial sector, which grew 40 percent faster than the economy as a whole owing to the development of the mining and textile sectors. In the service sector, transport, communication, and financial services also were growth drivers (Table 3.1). These sectors made up. for the fact that agriculture and livestock grew at less than 1 percent per annum over the period. Much of the growth in agriculture came from the livestock sector; market crop production is low (de Lange, 2008). Associated with this weak performance, from 2000, the food processing sector declined. As a result, by 2005, the share of agriculture in GDP fell to just less than 15 percent, while industry rose to take the largest share at 43 percent. Since 2003, the rate of growth is estimated to have slowed to less than 2 percent per year, but current estimates (since 2005) show an uptick, led by the mining sector. 50 Figure 3.1: Growth in overall real GDP, but no growth in agriculture 4500 5000 1 ,p---o 1 9% 4000 7% 3500 5% 3000 3% 2500 2000 1% 1500 -1% 1000 -3% 500 0 -5% 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 Sources: BoS and authors’ calculations. GDP = real GDP measured in constant 1995 millions of maloti. 3.5. While domestic production grew, the population did not. Population growth was less than 0.1 percent per year, so GDP growth virtually equaled GDP per capita growth. As a result, both consumption and savings grew rapidly. From 1995-2002, government consumption, which accounts for approximately one-third of total consumption, grew at 3 percent per annum. However, consistent with the slowdown in GDP growth, the growth of government consumption subsequently slowed to just under 1 percent per year. Government consumption growth has been financed primarily by nontax sources (SACU revenue transfers and revenue from water sales). 3.6. Private consumption is estimated to have been stagnant from 1994 until 2002. From then it is estimated to have grown at the same rate as government consumption, (0.8 percent) but below the rate of GDP. This rise in consumption implies that domestic savings grew substantially. Figure 3.2: Declining private consumption and remittances 3000 2500 2000 500 04 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 10 - - Private Consumption +Government Consumption +Investment -A- Remittances 1 Sources: BoS and authors’ calculations. Data are real numbers measured in constant 1995 millions of maloti. Remittances are fiscal-year data, while others are calendar-year data. 51 3.7. Overseas remittances traditionally have been a major source of financing for private consumption. Private consumption may have fallen because mining jobs and remittances from the South African mining sector fell steadily (Figure 3.2). In 2005 at 20 percent of GDP, these remittances would have financed 25 percent-30 percent of household consumption (the upper bound assumes that remittances financed no private investment). However, in 1994-05 they financed approximately 33 percent-40' percent of private consumption (allowing private consumption alone to be 10 percent greater than estimated GDP in that year). 3.8. How could poverty reduction take place if private consumption was declining? This is somewhat of a mystery. Indeed, the difference in the trends between the national accounts and the survey data is one element casting doubt on the observed poverty trend (which was, in any case, a very shaky observation, as noted in chapter 2). Part of the answer may lie in the estimated decline in inequalrty. Establishing trends over time in indicators measured at the household level in Lesotho is difficult, but it is (possibly) easier to establish relative trends . If inequality did decline, the poorest quintiles would have gotten a larger share of the pie, even if the pie were not growing. This theory would imply that, at the upper income levels, savings increased dramatically. 3.9. However, the distributional story is more complex than just rich and poor, because the change in the structure of production and the source of financing for household consumption changed the distribution of income between rural and urban areas. As domestic consumption became increasingly financed by earnings from domestic economic activity rather than remittances, and the growth in value added and jobs happened primarily in urban areas,'then distribution of the increased domestic income would be expected to favor urban areas. The growth of non-mining remittances affects the distribution of income and consumption as well. Overall, we would expect these trends to cause some households-xpected to be primarily rural- -to lose ground, and others to gain ground-expected to be primarily urban. Available data does not establish the extent to which this happened. While poverty in 2003 was estimated to be higher in rural than in urban areas, the rural urban gap is small and accounts for a small share of total inequality.*' 3.10. In sum, the evidence on the role of structural change in the macro economy-the decline in agricultural production and the growth of industry and of domestic labor income-on poverty reduction is ambiguous. The structural changes favored urban areas and wage workers, vs. farming and remittance household. Such changes are usually associated with widening inequality; yet the data show narrowing inequality. Furthermore, the ruralhrban gap in Lesotho is not an important determinant of inequality or poverty. Inconsistency between the micro and macro trends is not unusual; and given the low quality of the historical data, all of the above may be just noise in the data and not actual trends. 3.1 1. We now turn to the analysis of the labor market, the most important determinant of household livelihoods and welfare using microeconomic data, once again, analyzing patterns of structural change and their relationship with poverty. *' Recall that the definition of a rural area is administrative and does not relate directly to population density. This probably explains the conundrum. 52 B. MARKET OUTCOMES IN LESOTHO LABOR 3.12. At the aggregate level, the size of the labor force, the level of employment, and the human capital embedded in that labor force are important determinants of GDP. Equally important are the individual labor market outcomes: types of activity (sector), the products produced (per hour, day, week, or month), and the remuneration (wages, profits from the sale of goods, rents). In this section, we build up a the national labor market profile from the HBS household level data. We then disaggregate this aggregate picture by welfare level, rural and urban, to tease out the factors underlying the portrait of poverty in chapter 2. Normally, we would look at both employment and earnings as labor market outcomes. However, since HBS has no data on earnings at all we are limited to an analysis of employment. Structure of Employment, Unemployment, and Inactivity 3.13. In a primarily agricultural economy, members may work part or full time depending on the season, and undertake multiple economic activities within a week. This context makes measuring the gross amount of economic activity an individual undertakes, in which sector, and the returns to this effort complex. Deriving these indicators such that they are consistent across agricultural and nonagricultural activities and across types of households requires a more detailed employment and economic activity module than was used in the HBSs. Nonetheless, we did try to create these standard indicators from the data, and in the process we derived a richer picture of the labor force and employment than was possible using the indicators as they came off straight from the questionnaire. 3.14. Our goal was to create from the HBS data indicators of the labor force; and of those, the employed and unemployed according to the most commonly used definitions of these variables, agreed over time by statisticians and researchers under the leadership of the ILO.” Briefly, the definition of an employed person that we used is someone who was engaged in an economic activity for at least 1 hour in a 1-month period and was employed at least part of the 12-month period covered by the survey. An unemployed person therefore was someone who reported being unemployed and for whom we could find no evidence of having been in a job. To be consistent with standard definitions, we should have also required evidence on whether the labor force participant searched for a job. However, this question was not asked, so we simply assumed that self-reported unemployed had searched for a job. 3.15. To create these indicators, we made use of all of the survey data on the economic activity of individuals and of the household to estimate which household members are most likely to be economically active during at leastpart of the period covered by the survey, and in which type of activity.22We then estimated occupation, and from this, sector of activity. The latter is needed to link up household labor supply with national production. 3.16. Our profile of the labor force finds that 75 percent of the population over 15 was in the labor force in 2002/3.This rate compares with the approximately 50 percent participation 21 See Annex to this Chapter and h~o://~~vw.ilo.orglpublic/englishlburea~i/sta~do~~~nload/res/ecacooo.~d~ for the standard definitions, and annex 3.1 for details on how we applied them. The main difference between our analysis and that done by the BoS is that following international conventions, we do not allow a self-definition of unemployment. Rather, we check for any evidence of labor market activity, and if we find it, we classify the person as unemployed. 22 Owing to the seasonality of agricultural activities, days or weeks of part-time activity or idleness are to be expected. However, they are not the same as unemployment. Underemploynzentwas coined to describe labor force participants who (a) are not able to work as many hours in a week or weeks in a year as would be considered full time and (b) seek and are available for additional work. 53 rate picked up in the unadjusted data. The main difference in our methodology is that we considered most of the 51 percent of women in rural areas and 26 percent in urban areas who defined themselves to be housewives out of the labor force, to be working, because they lived in households with either a farm or nonfarm business’. Very plausibly these women engaged in economic activities by working in the household business over the reference period of the 12 months before the survey. The extremely high percentages (especially in rural areas) of women reporting themselves to be out of the labor force in poor households in which subsistence agriculture was the norm is implausible and clearly indicates the existence of a measurement bias that reflects a cultural bias. Women are more likely than men to consider themselv.es-and to be considered-inactive, even when they are very plausibly contributing to a family 3.17. Even after our reclassification, 12 percent of rural women and 9 percent of urban women remained “out of the labor force,” engaged solely in domestic chores without any farming responsibilities. Is this plausible? Women typically suffer from a gender-specific “burden of chores” that poses a heavy trade-off between time spent in productive activities and time spent in reproductive activities. Therefore, we expect female inactivity rates that are higher than males’. According to time-use data collected for a subsample of individuals interviewed in the 2002-03 HBS, women worked far longer hours than men in tasks such as cooking and domestic work. Women aged 15 and older spent approximately 3.7 hours per day in these unremunerated activities vs. 1.5 hours spent by men.24 3.18. The most important reason for being out of the labor force for those over 15 was to finish school, especially among young people in urban areas. A higher percent of males than females stayed in school beyond age 14 in both rural and urban areas (Table 3.2). Only a small fraction of those over 15 reported being out of the labor force due to disability or other reasons. 3.19. Of those in the labor force, a narrow definition of unemployment-ours after reclassification-finds approximately 5 percent unemployed compared with the H B S self- definition, which found a much higher unemployment rate of 24 percent. The broader definition of unemployment was self-described (how individuals perceived themselves). Numerous studies point out that, in rural Lesotho, the meaning of unemployment is lack of formal work, especially mine-work (Leduka and Gay, 2007), so it is not surprising that a high proportion of men would self-identify as ~nemployed.’~ 3.20. What do the economically active do? In Lesotho, as in most SSA countries, primarily they farm. Activities can be characterized by type (wage, self-employed), occupation (manager, clerical), and productive sector (agriculture, mining). Family farming is the one activity that is the same in all three classifications, and it is still the most important type of employment in Lesotho.26Overall, this activity occupies 58 percent of the labor force (Table 3.3), and 67 percent of households report having an agricultural business (Table 3.4). Even in urban areas, 21 percent of employed people are farmers. At least 1 member of the household usually is ” Interestingly, 23% percent of men aged 15+ define themselves as subsistence furrners vs. only 8 percent of women. However, it is extremely likely that housewives living together with male subsistence farmers are (at least occasionally) contributing family members. case given that in traditional society, crop agriculture included male and female jobs. 24 Lawson 2007. ” The 1999 labor market survey showed an unemployment rate of 3 1, whereas the 1997 labor market survey had had an unemployment rate of 59. This large gap indicates problems in generating believable time series data on unemployment (Lesotho Bureau of Statistics, Lesotho Labour Market Survey 1997 and 1999). 26 The exception is agricultural wage work that is not self-employed farming. 54 engaged in family farming in 24 percent of the households.?’ Farming shows up as an activity done equally by women and men in both urban and rural areas. However, this apparent result may emanate from our imputations. Agriculture/farming was often the default choice when it seemed as though a respondent was economically active but we had no additional information on the type of activity except that a farm business was run by someone in the household.** 3.21. Wage employment accounts for 21 percent of total employment-a fairly high share for a country at Lesotho’s income Over 16 percent of wage jobs are in agriculture, mining, or construction, and 5 percent are in transport and communication. Both d’these sectors are dominated by men (Table 35). Thirteen percent of wage jobs are in the textile sector, which is female-dominated, as is the education sector, which accounts for 8 percent of jobs (although the segregation here is fairly weak). Note that the authors were unable to classify 20 percent of the labor force by sector, because either no occupation was listed or an occupation was listed that was too generic to classify (such as clerical worker or me~hanic).~’ 3.22. Nonagricultural self-employment and small-scale (such as a family) business are important sources of income, especially in urban areas. Only 4.6 percent of labor force participants report being self-employed or employing others in a nonfarm business. However, slightly more than 14 percent of all households have either a self-employed or an employer among their members. This percentage is slightly lower than the percentage of households that report having a nonagricultural family business (17 percent) (Table 3.4). In urban areas, 25 percent of households report having such a business. As was the case for farm businesses (n. 5 ) , in urban areas, these percentages are somewhat discordant with those obtained from the individual employment status. Approximately 3 percent of the households without a nonfarm business have self-employed or employers among their members. 3.23. A more puzzling result was that approximately 41 percent of households with a nonfarm business had no member who was either self-employed or an employer. The authors could register only 1 activity type for each. person, and approximately 10 percent of households had both a farm and a nonfarm business. Therefore, it is possible that some people whom we classified as farmers had secondary activities as owners and operators of nonfarm businesses. However, in addition, it seems clear that at least some of the 13 percent of the labor force whom we could not classify by job type are owners/managers of such businesses. ” We compared the % of households with at least 1 farmer (after the authors’ reclassification) with the % of households with an agricultural business based on the corresponding HBS original variable. The results are concordant for rural areas (77% of households with at least 1 farmer and 79% of households with agricultural businesses), but slightly discordant for urban areas (24% and 3 1% of households, respectively). Overall, these rates suggests that our adjustments were not too far off. ’* Our assumption was that women living in households with a farm business are very likely helping their husbands (or other household members) in the business, but still identify themselves as housewives. For this reason, we reclassified them as unpaid workers in agriculture. See appendix 1 for the procedure that the authors followed. Regarding the invisibility of women in agriculture, see Safilios-Rothschild 1985,299-3 17. 29 For example, Ghana has roughly the same share ofthe labor force in agriculture but only 17% of the labor force in any kind of wage job. Note, however, that Ghana has a large commercial farm sector, though, while most of the crop farming in Lesotho is subsistence. ’’ The HBS asks only about the occupation, not the industry. Moreover, for occupation, it adopts a classification that is a mixture of occupational and industrial classifications. The authors tried to derive both an occupational classification and an industrial classification by mapping the HBS codes into ISCO and ISIC codes. However, some occupational categories (disproportionately low-skilled ones) did not find a corresponding industrial category. In sectors such as agriculture, textiles, or education, this inability to classify an occupation by sector probably did not affect the gender segregation finding. In these cases, it was easy to classify a respondent into any of these sectors if an occupation had been reported. However, for other sectors, the gender breakdown could be altered by the missing data. 55 Consequently, based on employment status, we probably underestimated to some extent engagement in nonfarm activities. Characteristics of Labor Force and Determinants of Employment 3.24. Lesotho is blessed with a well-educated labor force. Education levels are higher than in most African countries, especially for women. Nearly 50 percent of all workers aged 15+ have completed at least primary education, and more than 10 percent have completed at least secondary education (Table 3.6). Employed women are more educated than employed men (Figure 3.3). While 27 percent of male workers have no education at all and do not know how to read and write, the same is true for only 9 percent of female workers. This is one of the lowest levels of illiteracy in Africa. 3.25. Education to some extent determines job type. The education level of wage employees, employers, and self-employed is substantially higher than the education level of farmers and other non-farm workers (mostly family contributors and occasional workers), for both men and women (Figure 3.3). 3.26. Failure to complete primary education does not keep men from getting wage jobs, whereas women with no education are almost totally out of wage jobs. At low levels of education, women have a much lower likelihood than men to obtain wage jobs, but the situation is totally reversed at higher levels of education (vocational/technical education and above) (Table 3.6b). Thus, for example, only 2 percent of women with no education are in wage employment vs. 11 of men with no education. However, as many as 79 percent of women with a university degree are in wage employment, a much higher percentage than men with similar education (63 percent). Furthermore, many more women than men are employed in the public sector (especially in education and in health and social work), where education requirements are higher. Figure 3.3: Education Determines Job Type Women All Men Farmers Men Wage Women employees M~~ Women Employers Men Setf- Women employed Men Other non Women farm Men 0% 20% 40% 60% 80% 100% w None II Can readhvrite Primary 0 Secondary Vocational 0 Higher than sec. Sources: HBS 2002-03 and authors' calculations. Figure 3.3 includes the domestic labor force aged 15+ and excludes migrants living abroad. 56 3.27. Women have much better access to wage employment in urban than in rural areas, but occupational segregation exists in both urban and rural areas. In urban areas, where wage jobs are 48 percent of all jobs, wage jobs are almost equally divided between men and women. In contrast, in rural areas, the gap in wage employment between men and women is larger (1 8 percent of men have wage employment vs. 11 percent of women). Nevertheless, occupational segregation is as pervasive in urban areas as it is in rural areas. In urban areas, almost 30 percent of female wage employees are concentrated in the textile industry; in rural areas, 24 percent are in personal services (Table 3.5). Men tend to be distributed a little more evenly than women across occupations in urban areas. However, in rural areas, 33 percent of male wage employees are concentrated in agriculture and mining. While occupational segregation is often problematic, because it can signal women's exclusion from a whole range of jobs and accompanies wage discrimination against women, a more positive interpretation can be given to the Figures presented here. The textile industry, especially in urban areas, has offered women opportunities for wage employment that otherwise they would have had very little of, if any. c. LABOR MOBILITY AND STRUCTURAL CHANGE 3.28. Did growth translate into good jobs? As noted above, it difficult to be conclusive on this point given available data. Using trends on primary income source of the household and on employment by cohort, we find evidence that while growth did generate better jobs in urban areas, the structural transformation in employment is lagging, and this needs to be addressed in future poverty reduction strategies. 3.29. Agriculture is still an important sector of employment, even as it declines in importance in the economy. Structural change in value added was key driver of growth: agriculture and livestock are shrinking as a share of GDP, while mining, manufacturing, and other sectors are growing. By 2003, agriculture contributed only 15 percent of GDP, and much of this is in the form of subsistence (e.g. non-marketed production, undertaken with very limited capital and few modern input^).^' But in the same year, approximately 57 percent of the labor force still reported agriculture as their main economic activity. Even more surprising, the only time series on this appears to show that the percent of the population for whom farming is the main economic activity actually rose between 1994-05 and 2002-03. 32 But data on the main source of income for households showed a contrary trend:'the number of households actually relying on nonwage incomes and farming for their main sources of income has fallen (from 44 percent to 38 percent), and the percent of households relying on domestically earned wages has increased from 18 percent to 32 percent (Table 3.7). Most likely, household reliance on agriculture income declined, but households still wanted someone in the household working in this sector for food security, even though productivity of labor was higher elsewhere. Chapter 4 has a more detailed discussion of the structure of the agricultural sector. l2 As with other variables, our efforts to construct time trends are hampered by the fact that sector of activity was not asked at all in the two surveys, as well as lack of comparability between the two surveys on economic activity as it was measured. The question on main activity in HBS 1994-95 did not include an option for farming. Thus, farmers are believed to be found in the self-employed and unpaid family worker categories. In HBS 2002-03, a category for subsistence farming was included in addition to other self-employed and unpaid family worker. 57 3.30. I n both rural and urban areas, younger men and women are less involved in agriculture than older people, denoting labor mobility out of agriculture into faster growing sectors. (Figure 3.4). The percentage of men who are “farmers” decreases in line with the progression to younger birth cohorts. The decline goes from 71 percent of all employed men aged 55 to 64, to 60 percent Figure 3.4: Men and women working as farmer and in wage among 45 to 54 year-olds, jobs by age, 2002-03 (%) to 51 percent among 35 to 44 year-olds, to 49 percent 90T- I among 25 to 34 year-~lds.’~ (Table 3.9). Similarly, younger women are less 60 - likely to be engaged in 50 - agriculture than older women. The trend moves from women older than 55-for whom employment in agriculture is 70 percent or more of those 0-1 I employed-to women 25-34 35-44 45-54 55-64 65-74 younger than 55-for -A- Farmer Male -a- Farmer female whom employment in agriculture accounts for -A- Wage Employee Male -E- Wage h p l o y e e Female approximately 50 percent of Sources: HBS 2002-03 and authors’ calculations. Figure 3.4 includes the those employed. This domestic population and excludes migrants living abroad. percentage does not decrease further among women younger than 55 .34 3.3 1. The percentage of the labor force who are in wage jobs is higher for the younger than the older cohorts, especially among women (Figure 3.2). This suggests that the decline in agricultural employment is accompanied by higher engagement in high productivity activities. In urban areas, 58 percent of employed women aged 25 to 34 are wage employees-up from 54 percent for 35 to 44 year-olds, 48 percent for 45 to 54 year-olds, and 24 percent for 55 to 64 year-olds (Table 3.10). The increase for younger men is less dramatic. In particular, among 25 to 34-year-old urban men, there was a slight decrease of wage employees with respect to the next older cohort. According the 2002 H B S data, the percentage of employed young (aged 25-34) men and women living in urban areas who were in wage employment was about the same (approximately 60 percent) (Table 3.9). 3.32. Most salary jobs a r e in faster growing and high-productivity sectors. Public sector jobs absorb approximately 15 percent of all men and 23 percent of all women aged 15+ who are in wage employment. Apart from public sector jobs, men find wage jobs in personal services (mostly as protective service workers), but also in mining, construction, transport, 33 This is not an artifact of the authors’ reclassification. The percent of employed men aged 25 to 34 who are subsistence farmers according to the original HBS classification is 32% vs. 45% of 35 to 44 yeitr-olds, 56% of 45 to 54 year-olds, and 71% of 55 to 64 year-olds. 34 Note that the self classification and the authors reclassification show similar trends but different magnitudes. Based on the original definition of subsistence farmer, women are, overall, much less likely than men to define themselves as subsistence farmers. Moreover, the percentage of employed women who define themselves subsistence farmers dramatically declines among younger cohorts (table 3.10). However, based on the reclassification intofarmers of individuals who are “inactive” but live in a household with a farm business, there was an age-correlated drop in the percentage of employed women engaged in agriculture. 58 telecommunication, and as well as agriculture. Women are mostly found in the textile industry (approximately 25 percent of all wage jobs held by women 15+), personal services (mostly as helpers and housekeepers), and to a much lesser extent in trade, and hotels and restaurants (Table 3.5).35 3.33. Mining jobs attract older men while textile jobs attract younger women (Table 3.11). While approximately 20 percent of all male salary workers aged 45 to 54 were working in mining, less than 10 percent of those aged 25 to 34 were found in this sector. At the same time, new opportunities are opening up for young women in the textile sector. Twenty-nine percent of female salary workers aged 25 to 34 are being absorbed by the textile industry vs. only approximately 7 percent of female salary workers aged 45 to 54. If the textile industries prefer to hire younger women, then labor demand is driving this development. But supply side matters as well. Younger women are more geographically mobile, and they have more education, so they are likely to qualify for these jobs. ‘3.34. Employment rates among 25 to 34 year olds are lower than among older cohorts, especially for women. The employment rate (after reclassification) is 4 percentage points lower for men aged 25 to 34 with respect to men aged 45 to 54. The difference for women is 8.5 percentage points. This implies that the magnitude of the movement of the youngest cohorts into wage employment described above in terms of employment rates may well be less substantial when measured in absolute numbers of (wage) jobs created (Table 3.12). 3.35. It is unclear whether lower employment rates among the youngest cohorts are explained entirely by age factors (schooling, motherhood), or are (also) an indication of increased labor market instability. One reason that the 25 to 34 year-olds have lower employment rates than older cohorts; 3.5 percent of the former age group were still in school. For women, an additional explanation could be the presence of young children. However, when tabulating the employment rate for women who are spouses of the head of household and have no children aged 0 to 6 (admittedly a selected sample) to roughly control for the presence of young children, the difference between young and older cohorts persisted. Only 78.5 percent of women aged 25 to 34 are employed vs. 87.4 percent of women aged 45 to 54, suggesting that the presence of young children does not (fully) account for the lower employment rate of the younger cohorts. Whether younger generations (especially women) experience an increase or a decrease of employment opportunities is an important question that needs more and better data (e.g., panel data enabling separating age and cohort effects) (Table 3.12). 3.36. Younger cohorts suffer from an increasingly uncertain and unstable labor market, with potentially disruptive socioeconomic consequences. According to the self definition, as many as 27 percent of all men aged 25 to 34 and 12 percent of all women in the same age range defined themselves as unemployed over the previous year. These percentages were much higher than those among older cohorts (Table 3.12). The majority of these individuals are actually living in household with farm and nonfarm businesses, so they may not be technically unemployed given that it is very likely that they spent some time helping with the household business (hence the authors’ reclassification). As unemployment is correlated with income, the high number of people who see themselves as “unemployed” could have been caused by their expectations fuelled by increasing education levels, which were not matched by increased opportunities in the labor market. The important point is that a non-negligible part of the population, with 35Unfortunately, approximately 20% of all wage jobs held by women and approximately 25% of those held by men cannot be assigned to a specific industrial sector based on their occupations. 59 unemplo-vment rates that are dramatically higher among younger men and women, does not see itself as having a stable and clearly dejned livelihood. 3.37. Another worrying result is that nonagricultural activities other than wage employment represent a small and non-increasing share of all activities. Self-employment and engagement in entrepreneurial activities (including household members contributing to these activities) accounted for less than 20 percent of all jobs-a low percentage compared to other SSA countries. (Data weaknesses may be driving part of this result.) Self-employment is especially low in urban areas, where only 18 percent of men and 22 percent of women aged 15+ (28 percent of men and 34 percent of women among the employed population aged 15+) are either self-employed, or employers, or other nonfarm workers (tables 3.2). Moreover, younger generations do not seem to pick up more of these jobs than the older generations (Table 3.9). The picture is particularly gloomy for women, for whom self-employed activities could represent a unique opportunity to access income and diversify their activities outside agriculture. Female self-employment and non-farm activities rates decrease steadily as age decreases, especially in urban areas, from 30 percent for women aged 45 to 54, to 27 percent for women aged 34 to 45, and 18 percent for women aged 25 to 44 (Table 3.12). Moreover, it is worth noticing that most of the self-employed and nonfarm activities are concentrated in low productivity sectors, such as housekeeping and other personal service activities (more than 20 percent of all self-employed men and self-employed women). 3.38. One factor in the employment insecurity is the need to leave the family to migrate within Lesotho to get a job (Figure 3.4). Because many of the new jobs have been created in urban areas (especially in Maseru), many residents have had to migrate. According to the 2003 HBS data,36 approximately 11 percent of men and 9 percent of women have left behind a household and were working in another location (Table 3.14). Most of these are young workers. Fifteen percent of women 15-24, and 13 percent of men in the same age range, migrated. After age 45, people are much less likely to be internal migrants. Wage workers tend to be migrants;, 27 percent of female wage workers aged 15-24 and 24 percent of male wage workers in the same cohort are migrants (Table 3.14). Quite interestingly, young Basotho women migrating to urban areas tend to reproduce the same migratory patterns of male miners, who typically migrated individually leaving their family behind. While most women who migrate to urban areas of Lesotho are single, a non-negligible number are married, often leaving their husband and/or children in the rural areas where they are coming from. 36The HBS records only migrants who are still considered a part of the domestic household based on having a joint economy. Migrants who have set up new household structures are not included and thus could indicate an underestimation of the number of migrants. 60 status is an outcome; it may not reflect intentions. Many self- Figure 3.5: Men are more likely than women to migrate employed may have hoped to get a internally to find work (“Amigrant by age cohort) wage job, and probably still do. 16- Since 1 in 5 wage jobs within ,4.. \ Lesotho is taken by an internal ‘ , .. ... . .. . . . , .‘... ‘.... 12 - migrant, the incentive to migrate remains (Table 3.14). However, 10- among 25 to 34 year olds, more than 8. 1 in 3 self-employed males are 6- internal migrants; among 35 to 44 * year olds, almost 1 in 2 are. These 4 - figures suggest that self-employment z. and nonfarm opportunities are not created close to where people live. O1 15-24 25-34 3544 45-54 5564 The question is whether these two 1 .* Men +-Wornen] activities are Only in urban Sources: HBS 2002-03 and authors’ calculations. Figure 3.3 includes areas, or whether there are the domestic labor force aged 15+ and excludes migrants living opportunities in rural areas that are abroad. still unexploited. 3.40. In sum, structural transformation in the economy is working through the labor market with the creation of new wage jobs, but primarily in the urban areas and with gender biases. The Basotho labor force, especially younger workers, is migrating to get these new salary jobs. However, these are not being created as fast as the number of migrants need. The nonwage, nonagricultural sector is not growing as quickly and is not providing the secure employment that younger workers are seeking as they leave agriculture. In the next section, we trace the developments in the labor market through to household welfare and poverty reduction. D. GROWTH, EMPLOYMENT, AND POVERTY REDUCTION 3.41. Has structural change in the labor market supported poverty reduction? The most important income-earning asset that most households have-especially poor households-tends to be their “labor assets”: who works, and what they earn. Unfortunately, the HBS did not have data on earnings, so we were not able to estimate the returns that households got from deployment of these assets. We correlated (a) sources of income in the household with household welfare and (b) type of employment with household welfare. This allowed an investigation of how different types of employment status are associated with poverty, and how engagement in farm and nonfarm activities are associated with poverty. 3.42. Farming is still the main source of income for approximately 38 percent of * households. Nevertheless, the share relying on farming and on nonlabor income as their main sources has dropped substantially since 1994-05, reflecting the structural. changes (Table 3.7 and Figure 3.4). In 1994, 32 percent of households stated their main source of income to be either pensions, remittances, or non-labor sources, but by 2002-03, this number had fallen to less than 25 percent. Simultaneously, the share of households relying on wage income rose by 75 percent. While there were some regional differences based both on initial conditions and 61 developments during the period-for example, wage income households went up more in the areas near the construction of the dams-all areas gained in households dependent on wage income. Surprisingly, the share of households relying on nonfarm enterprise income as their main source stayed constant. bU - 50 - Faning Wages Household Pension Remittances Other Business 1.1994/95 02002/031 Sources: HBS 2002-03 and authors’ calculations. Figure shows household main source of income in 1994-95 and2002-03. 3.43. Strong differences remain between livelihood zones on the importance of farm income (Figure 3.5). These seem to be driven more by proximity to Maseru or South Africa and the opportunities there rather than the quality of the natural resource base. For example, the rural lowlands, where the best agricultural land is, has only 45 percent of households relying on farming as their main income despite having the best conditions, while the rural mountains have 60 percent of their households relying on in farming. This suggests that movement out of farming might have been driven more by whether opportunities were available (for example, in a nearby urban area) than land quality. However, it is noteworthy that the rural lowlands was the only area to show an increase in the number of households dependent on farm income-in all other areas, the loss of remittance income meant that households shifted into a nonagricultural activity. 62 F i g u r e 3.7: Household m a i n s o u r c e of income by region, 2002-03 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Maseru Other Urban Rural Rural Rural Rural SRV Lowlands Foothills Mountains Wages public sector Wages private sector 0 Farming Other HH busines 0 Pensions 0 Remittances El Other Sources: HBS 2002-03 and authors? calculations. 3.44. Being a wage employee (and, equally, living with a wage employee) is strongly correlated higher household income, especially in urban areas. Wage employees are disproportionately in the top of the consumption distribution. Table 3.15 shows the employment status composition by consumption quintile when only those permanently residing in the surveyed household are considered?? The positive association between consumption levels and being a wage employee is strong in urban areas, and somewhat stronger for men than for women. In rural areas, however, there is only a mild positive correlation. This may indicate that the quality of wage jobs in rural areas is not good enough to raise the position of wage employees and their households in the consumption distribution. 3.45. Self-employed are found through the distribution of welfare, suggesting that there is a large heterogeneity in nonfarm businesses and their success. In rural areas, self-employed, both men and women, are underrepresented in the bottom consumption quintile, but otherwise are pretty evenly distributed. For permanent residents in rural areas, living in a households with no business or only a nonfarm business is associated with a (mildly) higher probability of being in the bottom of the consumption distribution (Table 3.15). By contrast, those individuals living in households with only nonfarm businesses fare relatively better. Moreover, the combination of both a farm and nonfarm business represents another winning strategy. However, the picture is rather different in urban areas. Having a nonfarm business, or both a farm and nonfarm business, is not associated with higher consumption, but rather has the opposite effect. Actually, having or not having a business and the type of business owned do not appear to be crucial dimensions in explaining the level of household consumption in urban areas. In urban Lesotho, what matters most is having a wage job (Table 3.15). 3.46. Migration is still an effective livelihood strategy in sheltering households from poverty, but only when migration is directed outside the country and the migrants are men. 37Both external and internal migrants were excluded from the statistics to focus exclusively on the regular labor market activities of those continuously living in the household, to assess the extent that these activities correlate with their household?s poverty status. 63 As the shown in the regression in chapter 2, male migration to South Africa is associated with lower poverty of the household members left behind. However, this is the only type of migration that appears to bring some benefits to the rural sending household. By contrast, individuals living in household with male migrants within Lesotho, or with either internal or external female migrants, are overrepresented in the bottom of the consumption distribution. It is quite likely that the interpretations differ in the case of external male migrants and the rest of the migrants. Male migrants to South Africa are likely to be in good jobs that enable them to send home remittances that can improve the situation of the sending househl Id. In contrast, men migrating within Lesotho and women migrating (to either Lesotho or Sou~n Africa) could be more likely to do so because their household is poor. Or, these migrants may have higher living costs in their destination, so there is less money to send home. Whichever the direction of causality, this type of migration is much less effective in bringing benefits to the household. Old and new opportunities in wage employment: miners and textile workers 3.47. Approximately 5 percent of all households-2.5 percent of rural and 13.4 percent of urban household-have a textile wage worker among their members?* Textile workers represented 5.2 percent of all employed women aged 15+ in 2002/3. The large majority of them were wage employees (83.2 percent); the remaining were self-employed. Analyzing textile wage workers is one way to analyze to what extent the new employment opportunities in the (mostly urban) textile industries affect household livelihoods and poverty. The very large majority of textile wage workers are women (86 percent). Only 23 percent of all textile wage workers are recorded as “migrants” in the questionnaires. We expected a larger number of migrants among textile workers, but it may well have been that some textile wage workers residing in urban areas had migrated in the past and are now considered urban household members. Female textile wage workers are very young. More than 25 percent are younger than 25, and as many as 63 percent are younger than 35. 3.48. There is no evidence that households with textile wage workers fare better than the other households, particularly rural households declaring a textile worker as a member. Unlike miners, textile wage employees do not appear to be a crucial resource for the sending household. This is reflected, first, in the fact that very few of these workers are “mapped” into their household of origin in rural areas.39Second, only 6 percent of households with textile wage employees declare that remittances are their main source of income vs. 1 1.1 percent of household without textile wage employees, (Table 3.1 7).40Third, the rural households with textile workers are underrepresented in the top of the consumption distribution (3.17). Finally, regression analysis in chapter 2,showed that having an internal migrant was not associated with a higher income. 3.49. Indeed, households in rural areas counting a textile worker as a member were unlikely to be richer. Less than 35 percent of rural households with textile wage employees (representing 32 percent of individuals) are in the top 2 quintiles of the consumption distribution. This rate 38 Many textile workers recorded in urban households may have migrated from rural areas, It is not clear to what extent female textile workers who migrated from rural areas are recorded as migrant members of their rural household of origin or as a “new’’ urban household. It may be that, for cultural and economic reasons, male migrants are more likely than female migrants to be listed among the members of their rural household of origin, especially if they are heads of household, but this supposition cannot be verified in the data. 39 This statement means that of the total number of workers who listed wage employment in the textile sector,as their main employment, few were identified as migrants by the household that might have sent them. The workers themselves were not asked whether they were migrants. 40 It is true, however, that the majority (77%) of textile wage workers are not migrants. Thus, looking at the % of households that receive remittances from them as their main source of income is not as meaningful as it is for miners. 64 compares with more than 55 percent of urban households with textile wage employees-of which as many as 34 percent are in the very top quintile. However, quite interestingly, the urban households in the top quintile of the consumption distribution have on average only 3 members (2.7 if the authors exclude those who migrated abroad), instead of the 5-6 members that characterize the rest of the distribution. These results indicate that some of the young women who become textile wage employees in the urban areas do slightly better than the rest of the urban population if they live in a small household-maybe just on their own, with a partner and/or a child, or with a parent. Otherwise, they are no better off than the rest of the urban population. 3.50. The presence of a textile worker in the household does not have a sizable impact on the livelihood strategy of the household and the labor market behavior of the other members. Table 3.17 presents some characteristics of the household with and without textile wage workers from the HBS data. Having a textile worker made little difference on the probability of having a large remittance income (consistent with other evidence above.) 0 Household without textile workers were more likely to depend on farming as their main source of income 0 Household with textile workers were more likely to have their children in school, particularly youngsters aged 15 to 24: the percentages are higher in households with textile wage employees. A positive effect exists also for youngsters aged 15 to 24. 3.5 1. To sum up, based on the evidence presented in this section, the wage jobs created in the urban textile industry certainly represent new employment opportunities for women, especially younger ones, but they do not have a sizeable, positive impact on the sending rural household. They appear to have a positive impact on the “new” urban household of the women who take these jobs, whose consumption levels appear to be higher than average (this is especially true for smaller household). Moreover, there also are positive impacts on the education of children living with the textile wage worker. 4 1 3.52. Approximately 13 percent of all households-14 percent of rural households and 8 percent of urban households-have a miner among their members. Miners are overwhelmingly men (98 percent) and migrants (85 percent). Almost all miners who are migrants migrated to South Africa (96.3 percent); the remaining 3.7 percent are internal migrants. Therefore, of all miners, 82.3 percent work in South Africa and 17.7 in Lesotho. (This latter group is likely to increase in the future depending on the fortunes of the revival of the diamond mining industry in Lesotho). Miners tend to be older than other workers. More than 70 percent of them are older than 35, as compared with less than 50 percent for all the remaining workers, but very few are older than 55. 3.53. Households with miners (either in South Africa or in Lesotho) are richer than other households, particularly in rural areas. Almost 45 percent of rural households ( and one-third of urban households) with a miner among their members are in the top quintile of the consumption distribution (Table 3.19). (In terms of individuals living in those household, the percentage in the top quintile is slightly lower.) It is clearly the case that individuals supported by 41In this case, because we were interested primarily in the worker as a “labor asset” for the household, we compared urban household with and without textile workers (table 3.17). Clearly, other counterfactuals are possible, including at the individual level, for example, similar women (in urban areas) who are not employed in the textile industry and similar women who did not migrate to urban areas. 65 miners fare better than average. The majority of the individuals are in the top 2 quintiles (more than 50 percent in urban and almost 70 percent in rural areas are in the top 2 quintiles). 3.54. Households with miners heavily rely on remittances. Approximately 35 percent of households with miners indicate that remittances are the main source of household income. This compares with a mere 7 percent among households that do not have a miner (Table 3.20). By contrast, more than 40 percent of households without miners rely on farming as their main source of income, as compared with 8 percent of households with miners. The difference between the 2 types of households is even larger if one considers that these figures are not adjusted to take into account that the percentage of household living in rural areas is higher for households with miners (85 percent) than for households without miners (73 percent). 3.55. As with textile workers, households with miners use the income to send their children to school. Table 3.20 shows that in household with a miner, the percentage of youngsters going to school is approximately 15 points higher than in household without a miner. The difference is less pronounced than for the 15 to 24 year-old group, given that school attendance is generally very high in Lesotho. The difference is approximately 6 percentage points between the 2 groups of household (Table 3.20). 3.56. Quite interestingly, there were very small or no differences between households with miners working in Lesotho and households with miners working in South Africa. In both cases, the households tended to be richer. (About one-third of these households cited remittances as their main sources of income, and less than 10 percent cited farming as their main sources of income.) But the percentage of boys attending school, was higher in households with a miner working in South Africa than in households with a miner working in Lesotho. While 52.6 percent of all male household members aged 15 to 23 were going to school in households with a South Africa miner, only 37.8 percent in households with a Lesotho miner went to school in 2002103. 66 - C v; . r .- L +a. W Y P .e 8 2 8 2 8 0 8 3 0 * - m N 09 P " " 1 N W .8 . 0 y 0 0 r- r- a o w ~ n * K w P r n o - - d r i ~ rn mI N T 0 8 0 3 d N vr I 8 - 2 0 E 0 .- - I n a n 0 m .- I N 0 L s d N -3 N .u I I I d N P - x n B .- I h: E P .. m 2 w N ij F - a2 > 0 t- ? C .- 0 I . m u 3 F * ? - 0 w N 0 0 - $ u N u C m E vr m X u v) 3 QI A m P m CI e u $ 8 a - 0 .- c 0 E E u u u L w a M a w 3 C .- .I v1 .- -E v) 2 .. Y a m 3 2 2 z E .- > a 9) z 2 e, e 1 s 3 F g 4 a2 M 'E U z 9 F D .- 0 f % m 9 e, E n - C 5 8 N 0 0 N - .- Y C d .I a U + P B 8 .- L E 0 2 a u L a s .- C E .. : Q) z 3 z o m . p m m 0 t.- 4 1 b t. I m o=gwcro v) 41 3 z. d m I VI 8 ' p E: 7 m 8 W 2 p ' W I m 8 N 5 d m m I 8 d E: 7 m z W E: d W m I 8 c1 E: - e cd G E: m f 2: I n - ?'? m - O"."b. 8 I n N 2 ?-? c m w m 3 c d N 5 I vl -09 C 4 m m - 2 c E c 1 8 N $ -2 d .s b -2g Pc4 Table 3.15: Migration destination and consumption quintiles (%) Consumption quintiles 1 2 3 4 5 All Men from rural areas To Lesotho 36.4 25.9 16.7 13.4 7.6 100.0 To South Africa 15.6 15.3 22.5 22.0 24.7 100.0 Womenfrom rural areas To Lesotho 29.4 22.6 22.0 16.2 9.9 100.0 To South Africa 23.6 21.4 28.3 13.7 13.1 100.0 Menfrom urban areas To Lesotho 11.7 15.3 20.9 23.7 28.5 100.0 To South Africa 12.4 17.6 16.8 22.2 31.0 100.0 Womenfrom urban areas To Lesotho 21.6 14.4 20.3 21.5 22.2 100.0 To South Africa 24.3 30.4 7.9 15.6 21.9 100.0 Sources: HBS and authors’ calculations. Quintiles of consumption are calculated at the national level using a consumption aggregate defined by the Bank that excludes external migrants. Migrants for educational reasons (students) are excluded. Table 3.16: Characteristics of households with and without textile wage workers (%) households with textile households without textile wage employees wage employees Rural Urban Rural Urban Household size (all individuals) 6.0 5.0 5.0 4.0 Household size (excluding migrants living abroad) 4.7 4.0 4.3 3.7 % of households declaring: - Remittances are main source of income 11.6 3.1 12.3 7.3 - Farming is main source of income 35.1 1.2 48.7 5.1 YOof 15-24 who are students - Men 35.9 56.9 35.0 57.1 - Women 43.0 35.3 38.4 47.9 % of children 10-14 who are students - BOYS 78.1 97.4 82.7 87.6 - Girls 90.5 92.7 89.8 87.5 Total number of household (unweighted) 66 299 2864 2764 Sources: HBS and authors’ calculations. All figures are weighted when not otherwise specified. 79 Table 3.17: Textile workers across quintiles of consumption, 2002-03 Quintile 1 2 3 4 5 Total Rural 19.4 23.4 22.6 21.4 13.2 100 Households with textile workers Urban 9.3 15.4 20 21.9 33.5 100 Rural 21.5 21.6 24.7 20.8 11.5 100 Textile workers Urban 12.9 17.8 25.3 22.1 22 100 Sources: HBS and authors' calculations. Quintiles of consumption are calculated at national level using a consumption aggregate defined by Bank that excludes external migrants. Table 3.18.: Miners across quintiles of consumption, 2002-03 (%) Quintile 1 2 3 4 , 5 Total Rural 11 13.8 20.2 25.6 29.4 100 Households with miners Urban 6.3 6.8 18 24.6 44.3 100 Rural 11.8 15.7 21.3 25.2 26 100 Miners Urban 6.8 8 18.7 ' 26.8 39.6 100 Sources: HBS and authors calculations. Quintiles of consumption are calculated at national level using a consumption aggregate defined by WB that excludes external migrants. Table 3.19: Characteristics of households with and without miners (%) household household with household without household with miners in South with miners miners miners in Lesotho Africa % of household in rural area 85 73 78 86 Household size (all individuals) 6.1 4.6 5.8 6.1 Household size (excluding migrants 4.8 4.1 5.3 4.7 living abroad) % household declaring: - Remittances are main source of 35 7 31 36 Income - Farming is main 8 42 10 7 source of income % 15-24 who are students - Men 53 37 38 53 - Women 53 38 51 53 % children 10-14 who are students - BOYS 89 83 86 90 - Girls 94 89 95 94 Total number of household 702 5292 101 60 1 (unweighted) Sources: HBS and authors’ calculations. All figures are weighted when not otherwise specified. Annex to Chapter 3: Identification of Labor Market Status using the Household Budget Survey 3.57. The Household Budget Survey (HBS) asks about the employment status of all surveyed individuals aged 10 and above. However, the categories are nonstandard and do not correspond to any classification adopted by other African surveys, let alone the standard International Labour Organization (ILO) classifications of employment, unemployment, and inactivity. The HBS uses 1 single question that records the main status of the respondent during the 12 months before the interview. The respondents can classify themselves as employer, employed with salary, subsistence farmer, other self-employed, unpaid family worker, pupil/student, retired, disabled, unemployed, housemaker or housewife, and other. Apart from the ambiguity of some definitions (subsistence farmer, other self-employed), there are no filter questions that enable a more objective definition of employment, unemployment, or inactivity (and the more refined states within these broader categories). 3.58. In these circumstances, respondents are likely to select what they consider their main status over the previous year, even if they spent time in other states. Specifically, they easily may classify themselves as nonemployed (unemployed or inactive), even when according to the L O definition they should be classified as employed.42The problem arises because the ILO one-hour criterion (aimed at providing exhaustive estimates of the GDP) does not necessarily correspond to the respondent’s perception of what employment is. This discrepancy usually is much more serious for women. Women’s work is less likely than men’s work to be recognized as employment. Women see themselves (andor are seen by others) as “housewives,” even when they occasionally engage in productive activities, either because these activities are often unpaid or irregular, or simply because women’s work is not perceived as productive, and therefore not as “work”, especially when they are contributing family members. 3.59. To obviate the likely under-reporting of employment in the HBS, the authors imputed the employment status based on individual- and household-level variables, as described below and illustrated in Figure A. 1: 1. Nonemployed individuals according to the original question who reported earning money in other w’qs, as well as those reporting a meaningful occupation (although they defined themselves as nonemployed) have been classified as casual workers. Essentially, they declared to have done some work. 2. Nonemployed individuals living in households with a farm business43have been classified unpaid family workers-in the farm sector if they originally identified themselves as housewives, and potential contributors if they originally identified themselves unemployed or in other status.” 3. Nonemployed individuals living in households with nonfarm businesses have been classified potential contributors, if they originally identified themselves as housemaker/housewije, unemployed, or other status. 42 According to the ILO definition, the employed comprise all persons above a specified age who during the reference period were either (a) at work or (b) with a job or enterprise but temporarily absent from work. Persons at work are defmed as persons who during the reference period performed work for a wage or salary, or for profit or family gain, in cash or in kind, for at least one hour. Work includes any activity contributing to the production of goods and services as defined in the System of National Accounts (SNA). 43 Internal and external migrants have not been reclassified on the basis of the (farm or nonfarm) business of their household given that they do not live regularly in their household of origin. 3.60. As for sector of employment, a broad category offariners has been defined, including those originally classified as subsistencefurmers, plus all self-employed, unpaid family workers, and potential contributors who work on the farm sector (based on the occupation variable or on the family business sector, when the employment status as been imputed-as described above). Wage employees working in agriculture have been kept as such, not reclassified.asfarmers. Figure A.l Definition of “economically active” (EA) and inactive person from the HBS - Main Person Employment /Person declares/ Person is activity earned - status economic- active one B B Other having “meaningful” occupation (= + No - -+ 1 internal or J external migrant - Yes I + as determined - I by previous steps] (41 1=1 to (q12=2 to 4 m .( . ( EA (unpaid is housewife family 0 (as in q l 1 ) in househol with farm is unemployed or worker) (potential is other inactive Inactive Person lives in househol with nonfm or other (41 1=9, Source: Authors illustration based on the HBS 2002-03 questionnaire. Notes: The group of the “unemployed” will be defined by subtracting from the original group of the unemployed (41 1=9) the individuals who have been redefined as “potential contributors.” 4. AGRICULTURE, FOOD, AND NUTRITION SECURITY 4.1. Agriculture has been shrinking as a share of Lesotho’s GDP, falling to less than 15% by 2005, and has not been the main source of income for most households for over 15 years. Nonetheless, agriculture has absorbed a major share of the labor force and is an important livelihood .choice for households. We have labeled this phenomenon Lesotho s incomplete structural transformation. We have little direct evidence on labor productivity or household earnings in agriculture. However, indirect evidence suggests that labor productivity is low. Only a minority of households make a good livelihood in agriculture. These better-off households are able to meet many of their food needs through subsistence production. However, poorer households are not. The high inequality in rural incomes has implications for food security, as households that invest the majority of their labor in subsistence agriculture with limited results end up food insecure. Long-term food insecurity is both a cause and consequence of poverty, and it has important implications for nutrition of children and adults, productivity, and overall welfare. 4.2. The next two chapters focus on the rural economy, where the majority of poor households are concentrated. In this chapter, we look at trends in agriculture and food production at the national and household levels, the prospects for improved production among poor households, and the relationship between poverty and food security, including the poor nutritional indicators observed in Lesotho. In chapter V, we examine rural livelihoods: what assets poor and nonpoor households have, how they use them for agriculture and livestock production, what institutions govern these choices, and what food security outcomes are associated with these livelihood choices. We focus on the socioeconomic implications of these livelihood choices, relating them back to Lesotho’s structural transformation to an employment- based economy. 4.3. Our main findings are that for both natural resource and institutional reasons, the agricultural sector is underperforming as a source of income and hence food security. Many rural households that depend on this sector for food often are not able to meet their foods needs from home production, especially in bad years. While food is available for purchase on the local market (milled locally with grain largely imported from South Africa), some households cannot afford this food, including those with limited cash incomes. Food aid often is provided by donors, generally purchased in the local market, with its origins in RSA. While food aid has a role to play in crisis situations, other means of supporting poor households by providing cash-such as public works employment schemes-could be more efficient. The latter would strengthen the local market, reducing the volatility of grain demand. When food aid is used, the Government of Lesotho (GoL) could manage resources better by using grain futures markets in South Africa, securing options contracts, as opposed to buying at higher prevailing spot prices. 4.4. Food insecurity generally exists if one of four key pillars is not fulfilled. These are availability, access, stability, and efficient utilization. The four pillars are defined as follows: a. Availability: Food supplies, dometstically produced or imported, must be adequate to feed the population. b. Access: All people must have physical, social, and economic access to sufficient food. c. Stability: Access and availability must be ensured at all times. 84 d. Effective utilization: The food consumed must be safe and nutritious, and complementary resources available, such as health care, child care, safe water, to ensure good nutritional outcomes. 4.5. This chapter focuses first on food security at the national level, where two pillars dominate the discussion: food availability and national economic access to food. The second part looks at the household level, focusing on whether households can meet their own food availability needs through subsistence production, and/or have both economic and market access to sufficient food. The third section looks at whether households have been able to translate their access to available food into good nutritional outcomes. The final section suggests alternative strategies to improve food security. 4.6. Agriculture is an important component of food availability. Furthermore, in largely rural economies such as Lesotho’s, agriculture usually is a fundamental determinant of economic access to food. Given the sector’s importance for income generation, the topography of Lesotho adds to the challenge of physical market access to food by making transport and supply chains more challenging. The vagaries of climate impact on both the production of food and incomes of the poor, potentially reducing availability and access. Poverty reduces effective demand for food, constraining food markets - put simply people do not have enough money to buy food. Even when adequate nutritious food is available, environmental and health conditions can undermine the achievement of good nutrition outcomes. Poor water and sanitation affect hygiene,practices, resulting in more frequent illness. Under-nutrition both increases the susceptibility to HIV/AIDS infection and reduces the effectiveness of antiretroviral treatment (ARV) 4:7. Cutting across all of these factors are issues related to gender. Substantial evidence indicates that women are the gatekeepers of food security for their households, particularly for children-through agricultural production, food purchases, and meal selection, and through child care. Factors that compromise women’s access to land for food production, their economic opportunities, or time available to care for children (e.g. heavy burdens of work from chores such as collecting fuelwood and water) compromise food and nutrition security outcomes. A. FOOD NATIONAL SECURITY Trends in domestic food production 4.8. Lesotho was self-sufficient in cereal as late as 1978, but is very far from that today. Annual cereal requirements in Lesotho are approximately 180 kg per capita.” Until the 1930s, Lesotho was a net exporter of cereals (Mokitimi 1990). As late as 1978, in a good year, Lesotho produced sufficient cereals for requirements. Since then, Lesotho has never produced enough. By 1986, Lesotho’s total production of maize, wheat, and sorghum amounted to just 79 kg per capita per year, or 44 percent of requirements (Figure 4.1). In 1994-95, a drought year, Lesotho produced just 43 kg per capita per year. In 2002-03 production was 59 kg per capita, or less than one-third of requirements. 4.9. Cereal production is extremely volatile. Recent data show a negative trend, unlike earlier studies, which showed a positive trend. Earlier food security analysis underpinning the Lesotho Food Security Project (LFSP) indicated a positive trend in cereal production between 1980 and 44Minimum requirements are set at 180 kg cereals per person per year (FAO/WFP 2003. This annual total translates to some 1700 kcal/person/day. In typical Basotho diets, approximately 75 percent of the daily calories stem from cereals. Taking into account calorie consumption from cereals (some 500 g per day) and other food commodities results in average total daily calorie supplies of over 2200 kcal per person and day (Marsland and others 2004). 85 2003. However, more recent data do not support this. The trends from both 1968-2006 and 1980- 2006 are negative. Although production volatility makes trend analysis difficult, ultimately, total cereal production has been falling, on average, since 1968. Concurrently, the volatility of maize, wheat, and sorghum production has increased. From 1968 to 1985, the coefficient of variationd5 for maize, wheat, and sorghum was 28, 45, and 37, respectively. However,from 1986 to 2006, the figures were 34, 66, and 5 1. Maize is the dominant crop, rising from approximately 55 percent of production at the beginning of the 1980s to approximately 85 percent today (Figure 4.2). Yields in maize and wheat also have been volatile with no clear trend. Looking at the average since 1980, wheat seems to have shown some yield improvement while maize has been steady with neither decline nor improvement (Figure 4.3).46 Yet maize is increasingly dominant in cereal production at the expense of wheat. Figure 4.1: Cereal Production, 1968-2006 And 1980-2006 I A Sources: F A 0 databases and authors' calculations. Figure 4.2: K e y grains as share of cereal production, 1980 - 2006 (tons) 45 The coefficient of variation is defined as the standard deviation divided by the mean as a %. It is used primarily when comparing two distributions. The coefficient of variation has no units. It measures relative variability, that is, variability relative to the magnitude of the data. 46 It is difficult to find comparable time trends on production in Lesotho. In the improved version of FAOSTAT statistics for Lesotho were no longer published due to quality issues. Very recently (e.g. in 2008) Lesotho has again been included, despite continued quality issues. For example, identical maize import figures are reported for three consecutive years in 1999, 2000, 2001 before falling precipitously to a level less than one fifth the previous average. This trend is difficult to imagine in Lesotho. For this reason no food balance sheet could be constructed for Lesotho, and this table could not be updated. 86 Source: F A 0 data bases and authors' calculations Figure 4.3: Maize and wheat yields, 1980-2006 3000 _... r-- R 7 I I I I I 1 2000 I 1000 500 1C - - Wheat +-Maize I Source: F A 0 data bases and authors calculations 4.10. Cereal production in Lesotho is largely characterized by low-input, rainfed production on poor soils, small land sizes, and high risk due to harsh weather conditions. Lesotho is 70 percent mountainous. Approximately 11 percent of Lesotho's land area is classified as arable, or approximately 0.2 hectares (ha) per person, and less than 1 percent is classified as high-potential arable land (FA0 2007). The climate is highly variable. Climatic risk includes drought, often extensive, and the impacts of hailstorms, snow, and frost damage to crops. Total rainfall generally is not a limiting factor for agricultural production but significant variability across seasons and years with diverse effects across regions is a challenge. Table 4.1 summarizes the different agro-ecological conditions and farming systems across the diverse landscape. Growing seasons for summer crops are limited due to the short period between the melting of the snow and the first winter frosts. This narrow window makes choosing the correct cultivars and timely planting extremely important. 4.1 1. Although more than 50 percent of the labor force works in agriculture, Lesotho has very little commercial agriculture. Most food production in Lesotho is for subsistence, resulting in little or no marketable surpluses. Just 10 percent of cereal production has a commercial orientation. According to preliminary results of a recent World Food Programme (WFP)/Disaster Management Authority (DMA) survey based on the 2006-07 cropping season (a drought year), only 5 percent of households that had access to land sold any maize. Just 0.2 percent of households sold all their production. The average area cultivated per farmer is 1.3 ha; only 11 percent of farmers cultivate more than 3 ha (FA0 2007). Primitive technologies are used, and mechanization is very low. Labor is undertaken mostly by hand, and tasks are assigned by gender, limiting labor flexibility and potentially yields. Fertilizer use is very limited, and generally used only by commercial farms in normal market conditions, making soil management even more important to maximize crop yields. 4.12. The ratio of fallow land is high and increasing. High risk and low profitability at the farm level limit investments and make many potential farmers decide not to cultivate, leaving fallow an increasing amount of crop land. The agricultural census in 1999-2000 indicated a countrywide fallow rate of 31 percent-with an upward trend to more than 40 percent by 2003 87 (Turner 2003). This trend is substantiated by localized surveys undertaken around 1980, indicating fallow rates of 25 percent and 22 percent, respectively (Turner 1978 and Murray 1981 cited in Boehm 2004). The high percentage of land lying fallow is also an indicator of the lack of resources--cash, labor, and implements-to farm the land.” Poverty and being too poor to farm featured prominently as the main reasons for fallow fields in the Poverty and Migration Survey (Sechaba 2005). 4.13. Constraints to improve agricultural production are numerous and difficult. Constraints in agricultural production include depleted soils, poor use of the available water resources, very limited irrigation, no or limited use of fertilizer and hybrid seed, weak extension system, inferior infrastructure, under-developed markets, and little security in land tenure. One estimate indicated that 40 million tons of topsoil are lost each year due to erosion and land degradation, a very serious problem given the thin soils in many parts of the mountainous terrain (de Lange 2008). Erosion and land degradation limit the chances of agricultural mechanization and are issues requiring serious attention. WFPEAO reported that “crop production is decreasing and could stop all together over large tracts of the country if steps are not taken to reverse.. ..the decline in soil The limited use of fertilizers adds to already severe problems of nutrient depletion and soil erosion. The decline in soil organic matter and depletion of soil nutrients are due in part to the increasing burning of manure for energy and feeding crop residues to livestock (KOL 2007). The result is further depletion of soil moisture retention capacity, and increasing vulnerability to climatic variability and dry spells. The combination of a fragile ecosystem, poor and declining soil fertility, and climatic hazards generates high risk and uncertainty. Land tenure insecurity exacerbates these problems and limits other than immediate investment in the land. Large parts of the country-particular in the Mountains-also are sparsely populated with poor ’ access to roads, which makes access to markets for both production inputs and crops sales a challenge. 4.14. Maize production in Lesotho underperforms when compared to production in South Africa. A recent study (de Lange 2008) compared maize yields in Lesotho and areas across the border in South Africa with similar agro-ecological systems. The comparison showed that yields in Lesotho are only 25 percent-30 percent of the yields obtained in neighboring regions of RSA (Figure 4.4).Another study focused on the cost of producing a kg of maize (without husks) on a Basothan smallholding and found the cost to be nearly three times the cost of commercial maize production in SA: 1.96 rand per kilogram in Lesotho compared to 0.7 rand in SA. A simulation to impose the same production input costs (seeds, fertilizer), with the exception of fuel and maintenance costs, and assuming a 20 percent yield increase as a result, reduced production costs in Lesotho to only 1.26 rand per kilogram (Global Development Solutions (GDS 2004). A recent study by F A 0 also found similar results for both yields and cost per kg (Table 4.2). 4.15. On this basis, Lesotho grain production is not v i a b l w v e n as a commercial outgrower relationship of Basothan farmers with RSA. More analysis is required to determine what yield improvement is possible for maize in Lesotho and whether making this improvement would make economic sense compared to increasing imports from RSA. Other areas with similar conditions to Lesotho, such as Kwa Zulu Natal, South Africa, have demonstrated that correct agronomic practices can build up the soils and assist in water retention, thus compensating for the variability of rainfall (de Lange 2008). 47Lesotho thus may be moving towards the South African situation, where more than 80 percent of arable lands in some areas, for example, Matatiete District in the Eastern Cape, are fallow, and the majority of people survive on pensions, child support grants, and what is left of remittance flows (Boehm 2004). 48 News.bbc.co.uWllhi/wor1d/Africa/2029183-561< 88 Figure 4.4 Maize yields in Lesotho and neighboring South Africa, 2007 (t/ha) 35 3 25 2 15 1 05 0 Buthe Buthe/ Founesburg BereaIClowlane M afetenwepener [ 0 Lesotho South Afnca j Source: De Lange 2008. All South African yields have been verified against insurance statistics and physical delivery records. 4.16. Other land usage is likely to offer higher returns than cultivating maize and would increase food security through the economic pathway as opposed to the subsistence production pathway. The dominance of maize as the staple crop for food security leads to a likely suboptimal dominance in farmers’ crop prioritization of maize. Vegetables may well be more suitable for some production areas and yield far higher profits with the benefits of added dietary quality given their higher micronutrient densities. However, vegetable production is riskier, given their susceptibility to drought, floods, and frost. 4.17. Farmers are diversifying out of grain, and ‘subsistence style’ homestead gardening of vegetables and fruits is becoming increasingly common, despite constraints. Many households, especially in the Foothills and Lowlands, have one or more peach trees or other fruit trees (apples, plums, grape vines, mulberries, pears). A 2004 study indicated that, in the late 1990s, at least 70 percent of households in all districts grew some vegetables (Makhotla and Hendriks 2004). The major types of vegetables grown in Lesotho include cabbage, spinach, tomatoes, leafy greens (mustard, rape, collard), carrots, onions, beetroot, potatoes, and pumpkins. These have the advantage nutritionally supplying much needed micronutrients not found in staple crops. Unfortunately, the majority of households grow green leafy vegetables whereas less than 20 percent growing carrots or pumpkins, better sources of vitamin A. 4.18. A major constraint of fruit and vegetable cropping systems is that they are labor intensive, which can be problematic given the AIDS pandemic. Good quality production inputs such as hybrid seeds and fertilizers together with irrigation are critical. if the vagaries of rainfall are not to limit production. However, most poor farmers are too cash constrained to make the required investments. Credit viability, and hence availability, is limited, given the absence of insurance markets to mitigate production risk. 4.19. While more analysis is required, a commercial outgrower relationship with South Africa could be possible for some high-value crops. RSA’s rapidly increasing “supermarket” supply chains for vegetables, particularly given the proximity of their markets, are an opportunity. For example, one study indicated that the production cost of a kg of peas was 4.14 malotgrand in Lesotho but 5.25 rand in SA (GDS LLC 2004). However, a diversification strategy is unlikely to 89 be successful without greater use of quality farm inputs. This is unlikely unless risk management options are made available to farmers-to mitigate the unpredictability of nature and the market. Consequence of Past and Current Agricultural Policies for Agriculture and Food Security 4.20. Since independence in 1966, government intervention and policies have undermined private sector investment in farming and farming-related enterprises. At independence, Lesotho adopted a maize self-sufficiency policy. It led to the establishment of parastatals for the purchase and supply of crops and inputs and the erosion of the existing network of private traders. A series of area development programs also fostered farmers’ dependence on GoL. For example, the Cooperative Crop Production Programme (1976-80) supported sharecropping of winter wheat in the Lowlands. The GoL supplied and paid for all inputs including seeds, ploughing, planting, harrowing, fertilizer, and.combine harvesting. Farmers were responsible for only nonmechanized tasks such as weeding and threshing where land was inaccessible to combine harvesters. This program was christened Ahlama u-je-“manna has fallen from heaven”-by farmers (Phororo 1979 cited in Mokitimi 1990). After harvest, the wheat was divided between the farmers and GoL. 4.2 1. Following the cooperative production program’s demise in 1980, a Food self-sufficiency Program (FSSP) was initiated with a goal of achieving maize and sorghum self-sufficiency within 5 years. It was still operational, a decade later in 1990. Similar to the previous program, the plan was for the GoL to meet production costs in the first year, and farmers would begin cost-sharing .in the second year. Under the FSSP, the Lesotho Agricultural Development Bank and Lesotho Bank supplied agricultural credit for tractor services, seeds, and fertilizer to cover the cost- sharing arrangements. Repayments rates were, however, exceedingly low. In 1992 the Lesotho Bank reported that just 7.8 percent of the money lent in the 1990-91 planting season had been repaid. Again, farmers christened this program Ahlama ~ - j eThe ~ . ~ impact of this policy is still felt today. Since tractor services under FSSP were largely provided by GoL using its own machines, farmers had no incentives to invest in tractors and service provision. Hence, most tractor services today still are provided by GoL, even though these tractors are known to often be unavailable and have untimely services. Furthermore, the agricultural credit schemes in essence amounted to government transfer schemes. By fostering poor financial discipline in borrowers, these two schemes undermined the development of vibrant rural financial services, which would have supported broader rural investment. Provision of public agricultural services undermined the development of private sector services and equipment provision such as seeds, fertilizer, and tractors. 4.22. Government controls (imports and prices) also undermined private small-scale milling. As part of the government’s strategy to attain maize self-sufficiency, all maize and maize product imports were tightly controlled by the Ministry of Agriculture. This fact, in conjunction with policy in South Africa, essentially ensured that virtually all-97 percent in 1992-93-maize imports were authorized to Lesotho’s 3 big industrial mills, one of which was operated by GoL. Government-controlled prices were paid to Basothan farmers for maize sold to these mills. As a consequence of the concentration of milling, the existing network of small indigenous hammer mills shrank. The use of large-scale milling might have been more efficient. However, the milling technology used was such that it was less profitable when maize was purchased in the small quantities sold by most Basothan fanners, and the maize was of highly variable quality. Hence the industrial mills made their most profit when Lesotho had a poor maize harvest and they imported grain from RSA to mill-not exactly the right incentive for enhancing local production. 49 Much ofthis paragraph is based on Mokitimi 1990. 90 4.23. Inflated maize prices (due partly to South African policies) led to unsustainable maize planting, which contributed to soil erosion. Despite its food self-sufficiency drive, by the early 1990s, Lesotho was dependent on South Africa for maize imports. South Africa operated a domestic Maize Marketing Board (MMB) to which all surplus producers had to sell at a pan- territorial price regardless of location. MMB’s operating costs-administrative, storage, financing crop purchases-were recouped by a levy on sales to both the domestic market and Botswana, Namibia, Lesotho, and Swaziland. If South Africa produced a large surplus, then it added a stabilization levy to prevent prices falling too low and to support maize exports. Exacerbating the impact of these pricing policies on Lesotho, the South African marketing board would issue a permit for maize exports to Lesotho only when the buyer could produce a Lesotho import permit. Then the South African marketing board would allocate maize from a more remote production region to Lesotho, thus increasing the transportation costs. The end result for Lesotho was inflated maize prices. The high prices contributed to farmers (subsistence and commercial) growing maize as opposed to other crops and planting on slopes and unsuitable land, leading to soil degradation and erosion. 4.24. Lesotho’s continued use of input and consumer subsidies in response to droughts and other environment crises still undermines market development and sustains a farmer “wait and see’’ attitude. GoL favors a production subsidy response in the agricultural season following a drought. In 2002, following two consecutive droughts, GoL implemented a farming input subsidy scheme, with subsidies as high as 50 percent. The total cost of the subsidy amounted to approximately 7.5 million maloti, reflecting increased spending on agriculture in 2001-2002 and 2002-2003 before the subsidy declined (Table 6.1). The most recent drought crisis, in 2006-07, again produced a “subsidy” response. On the crop production side, all farming inputs were subsidized by 30 percent for the 2007-08 season. Receipts were required to be submitted to the Ministry of Agriculture and Food Security (MoAFS) by the end of March 2008. 4.25. Based on a total cultivated hectarage of 150,000 for maize, with average production input costs per hectare (ha) of US$170, the total budgetary cost of a 30 percent subsidy on production input costs is US$9.18 million or 64.26 million maloti (Table 4.2). Added to this is a system of food subsidies on the consumption side. These cover specified staple foods and operate through retail outlets. It is understood that consumers take receipts for the stipulated products to the district commissioner’s office to receive their subsidies. The budgetary costs of these are unknown. The repeated government interventions in the agriculture sector, essentially a private sector activity, has resulted in farmers’ waiting for government support as opposed to taking timely decisions regarding their fields. The results are late planting and untimely use of inputs. Given that farming is a “timely” occupation geared to crop cycles, government intervention and the resulting passivity of farmers undermine production, local food availability, and the incomes of those dependent on agriculture. 4.26. The agricultural policies also have increased inequality. Subsidy systems traditionally used in Lesotho are inherently regressive, excluding people who need them most. FSSP was targeted to those owning more than 15 ha. Because richer farmers own more land, they got greater subsidies and loans, most of which they did not repay. Even GoL’s recent response in 2007 to the drought exacerbates inequality due to its focus on subsidies that disproportionately end up with the wealthier. Absolute input subsidies are largest for those with the greatest landholdings because they use more inputs, and for those with available cash to invest and later get reimbursed. Consumer subsidies are similarly regressive. Since wealthier households have higher absolute expenditures on food than poorer households, the former will receive a larger share of the subsidy. The poorest most needy people are excluded from receiving subsidies if they do not have money for food in the first instance. Moreover, the poorest are likely to live in more remote areas, so they incur higher transport costs to get to the District Commissioner‘s office to redeem their subsidies, reducing the value of the subsidy still further. 4.27. In sum, misguided government intervention in agriculture has left Lesotho with a legacy of degraded and eroded soils, stripped rural areas of an entrepreneurial spirit, and resulted in rural communities who sit back and wait for the next round of government intervention. As a consequence, malnutrition, as measured by anthropometric outcomes, is high and persistently so. Access to Food 4.28. Lesotho depends heavily on the international market for access to food-something Lesotho is better positioned to do than most. The declining domestic production of cereals in Lesotho has led to more dependence on commercial imports. South Africa is the largest cereal exporter in the region and given Lesotho’s geographic location within South Africa and its status as a domestic buyer, Lesotho is unlikely to lose market access-even in the event of shortfall in production in South Africa. Furthermore, Lesotho’s foreign exchange earnings from its Southern African Customs Union (SACU) membership, and the currency peg with the South African rand ,eliminate domestic exchange rate risk and secure adequate foreign currency. 4.29. Built-in inefficiencies in transportation negatively affect food access for the poor by raising costs. There is a railhead from South Africa to Maseru, but the overall cost of using it seems to be high. As an example, Limiio and Venables (2001) indicated that whereas shipping a 40 ft container from Baltimore to Durban costs $2500, onward transportation 347 km to Maseru results in an additional $7500. It costs approximately R7125 to transport a 40-foot container via rail from Port Elizabeth (South Africa’s main port) to Maseru. The same route and volume by truck would cost between 9,500 and 12,500 rand. However, although trucking costs are up to 75 percent higher than rail, trucking takes only 1 day as opposed to 9-30 days to complete delivery by rail. This huge difference is due to many inefficiencies in the rail, warehouse, and clearance systems. Maize does have priority on 4 rail cars per day for delivery to Maseru, but this creates distortions in the overall freight system. 4.30. T h e reliance on the market exposes Lesotho to local and world price volatility. Lesotho essentially takes its maize price ,from South Africa. However, domestically, the import parity price is inflated by transport costs and logistical issues. Importation takes place largely in the Lowlands, and significant costs are incurred in the long and difficult supply chain to the Highlands. Importation of both wheat and maize is handled largely by 2 main mills: 1 in the north close to the main domestic production areas and 1 in Maseru. These companies use the South African Futures Exchange (SAFEX) to hedge production price risks. However, they manage some of the logistical and transportation risks by procuring directly from the towns close to the Lesotho border at spot prices.5o 4.3 1. South African producers have come to expect Basotho food purchases and allocate a part of their local production to supply the Lesotho market. This is not without risk. Production in this area of South Africa is largely rainfed and therefore subject to weather-related variability. The price volatility was starkly shown during the HBS period. In April 2002, following 2 poor crop 50 The mills buy options contracts on SAFEX to purchase grain at a fixed price at a future point in time, independent of the prevailing market price at that time. By also buying locally, they minimize transport costs because the Randfontein base of the SAFEX is close to Johannesburg, in SA’Smain maize production area. The options contract ensures the mills against crop failure in the SA local markets, which otherwise would force them to buy at spot prices on the SAFEX. 92 years and drought, the SAFEX maize spot price was R2045 per ton. It fell to R782 per ton in April 2003. More recently, in 2006-07 the production decline in Lesotho, combined with expected below average production in South Africa (7.1 million tons forecast against 9.6 million tons on average) led the SAFEX price of maize to rise to R1930 rand in March 2007, compared to R1350 in January 2007, well above the same period in 2006 (Figure 4.6). 4.32. As a net food importing country the current ‘global food price crisis’ has presented Lesotho with an additional challenge, beyond the normal maize production volatility and rand exchange rate risk. Between January 2005 and June 2008 the world price of maize tripled. Fig 4.5 indicates a similar pattern with regard to maize prices on the South African market. Wheat prices increased by 127% and rice prices by 170%. The price of food oils showed similar increases. Much of this is driven by a focus on biofuels, sourced from maize, sugar and oils, in many countries. Maize prices are unlikely to fall significantly in the short run, despite increased plantings, given reduced production prospects in the USA, the largest exporter, due to severe floods in key growing areas. Increased maize prices combined with high oil costs was a major terms of trade shock, significantly affecting Lesotho’s national income. F A 0 classified Lesotho as one of 34 countries most hit by the crisis. Figure 4.5: Maize monthly spot price, SAFEX, 2000-2007 (Wt) 2500.00 2000 00 - h 6 1500.00 . s i ,E 1000 00 - P 500.00 . Source: SAFEX 4.33. Declines in purchasing power, whether due to drought or to the food price rises, result in the poor households unable to buy sufficient food for their needs. This results in lowered demand for food imports and a gap between aggregate food needs and aggregate food supplies. This gap is bridged by food aid. The importance of food aid has varied over time (Table 4 . 9 , but has played a key role in reaching many poor in recent years when production failure has been at its worst. Food aid is financed by GOL, the donors, or both. Most food aid is now purchased locally in Lesotho, but the grain largely originated in RSA. Thus commercial imports, fig. 4.6, do not show the volatility that would be expected due to reduced purchasing power. The estimated figure for commercial imports in 2007-08 is higher than average as it is based on import requirements. It was estimated that of the 256,000 tonnes of imports required, 219,000 would be handled by the private sector leaving an uncovered balance of some 30,000 metric tons - a food aid requirement. 93 Figure 4.6: Commercial imports in 2007-08 I 50000 45000 40000 Figure 4.7: Availability of cereals (maize, sorghum, wheat and rice) - total (‘000 Tonnes) and per capita (Kg/year), (2002/03 to 2007/08) n P er cepita cmsumption &g) --c P rodudion -imports (commacial Flus bod dd) Total amileibilty 350 300 250 200 150 100 50 0 2002103 a103f04 2m4m 2005~06 20~307 2007#08 estimate Source FAO/WFP 2007 B. FOOD HOUSEHOLD SECURITY Availability of Food 4.34. The poorest Basothan rely the most on the market for food security. In many African households, subsistence agriculture constitutes a significant part of the food supply. The proportion increases with deepening poverty. Evidence from 2004 indicates that the pattern in Lesotho is reversed (Marsland and others 2004). The proportion of household food supplied from own production in Lesotho increases with increasing income (Figure 4.7, Table 4.3). The “better- off’ households usually are able to produce more food and-in most cases-a larger share of overall food requirements from their own crop and livestock assets than poorer households. 94 Poorer households generally produce less than wealthier households, in both absolute and relative terms (Lesotho Livelihood Profiles 2006). There are some variations between rural regions. A similar pattern is found in the HBS data in which the share of food that derived from own production is lower among the poor than among the rich.” It follows that the poorest are most dependent on the market for food. The market chain from the Lowlands to the poorest areas in the Highlands is both long and complex, thus increasing food prices. Figure 4.8: Percent food from own production, 2004 (%) 60 ‘50 b0 30 20 10 0 SouOLlern Northern Lowlands Foobliiis Smqu Rver Valley Mountains Pen-urban Lowlands IMPoor MMiddie OWell-offj Source: Lesotho Vulnerability Assessment Committee (LVAC) 2004 cited in Marsland and others 2004. Note: Table shows proportion of calories from own production in a “normal” year. A “normal” year is defined as 1 of the last 5 years that is “typical” for the area, that is, when food access was neither exceptionally good nor exceptionally bad. 4.35. A 2007 national community household surveillance conducted jointly by the Disaster Management Authority and WFP,52 revealed that only in Mokhotlong and Butha Buthe, was food/crop production the primary source of livelihood. In Mafeteng, Mohale’s Hoek and Qacha’s Nek the primary livelihood source was food assistance. This situation is clearly not sustainable. Access to Food 4.36. Access to food in Lesotho is driven by access to income and markets. In the past, the poor in Lesotho may have benefited from South African maize meal demand patterns. During milling operations both high-quality maize meal (super maize meal) and lower-quality maize meal (special maize meal) are produced. In South Africa, consumer demand is for the super maize meal, far outstripping the demand for the special maize meal. As a result, South African companies “sell” the excess special maize meal in Lesotho, primarily in the more accessible Lowlands (de Lange 2008). What in reality is dumping creates two issues for the two local millers. First, it overly distorts their market base to serve the Highland areas, which requires more difficult supply chains and higher costs. Second, the SA special maize mill reduces potential economies of scale in Lesotho’s own milling operations. However, the marked increase in maize Despite these low subsistence levels, there are accounts in the qualitative literature that dependence on crop and livestock production has become an increasingly important source for certain groups, particularly among the “chronically poor” (LRAF’ brief 5, Lethola 2005). 52 WFP/DMA CHS survey 2007 has national coverage but is not nationally representative in that it overs samples the food aid population and does not include vulnerable urban households. 95 prices in the recent past may reduce or even eliminate this ?selling.? Poorer SA consumers may switch to consuming the special maize meal and reduce the mismatch between the demands for super and special maize meal, reducing or eliminating the surplus of the latter available to sell in Lesotho. 4.37. Very little locally produced maize is marketed within Lesotho as there are few commercial-scale producers. This fact leaves idle both aged hammer mills and more modern roller mills (de Lange 2008). Households owning more assets were more likely to sell maize, but even of these households, less than 10 percent engaged in market sales. On average, sellers sold 43 kg of maize annually, less than 1 average sack (60 kg). More than 50 percent of households sold less than 40 kg. Almost two-thirds of sellers sold maize from their homes. When maize was transported for sale, the average distance was 27km. Almost all households milled their own grain. Approximately 75 percent used local mills. Just 16 percent milled their grain by hand. Forty-six percent of households took less than half a day to get their maize milled, but for 35 ? percent, time invested in milling was beyond half a day. The majority milled between 1 and 5 kg. More than half of the households had to pay cash to transport their grain at an average cost of 10 rand. Those with more assets were more likely to pay in cash and to have access to transport, including government transport. Those lacking assets were more likely to hand-carry their maize, or pay for transport with maize..s3Local milling of maize is promoted by GoL as the answer to ensure maize availability in the more remote regions. However, little local maize is marketed, and most of it is of poor quality and cannot be processed into quality meal. As a result, local maize is more likely processed in the aged hammer mills than in the modem roller mills. 4.38. In summary, Basotho are dependent on their own limited production and economic access to markets for their food needs. Where the rural economy fails to generate jobs and poverty constrains effective demand for food, dependence on food assistance and gifts has been the result. C. OUTCOMES NUTRITIONAL 4.39. Anthropometric measures usually are used as the measure of food and nutrition security, despite the fact that they reflect more than food security. Absent living in a good environment with quality health care, even consumption of sufficient diversified foods will not ensure good nutritional status, a measure of nutritional security. 4.40. Anthropometric indicators for children under 5 years of age are disappointing on 2 levels-the magnitude and their resilience to sustained change. Looking at the trend from 1992 to 2006 in stunting, underweight, and wasting, there is no clear progress; all indicators are consistently high (Table 4.4). Even using Lesotho?s ?best? nutrition results from DHS 2004, WHO would classify the levels of stunting and underweight as high and therefore a public health concern. It should also be noted that national-level statistics mask distinct regional variation in anthropometric outcomes (Table 4.5) Clearly, some trends are alarming. The high underweight and wasting rates in 2004 are undoubtedly part of the explanation for the high stunting manifest in 2006, reflecting the longer term results of inadequate nutrition. 4.41. International .benchmarking shows that Lesotho is trailing other Southern African countries in stunting-particularly considering its higher GNI per capita. Relative to its purchasing power parity (PPP) gross national income (GNI) per capita, Lesotho is underperforming with respect to achievement in nutrition (Table 4.6). Using the DHS 2004 best ?Data in this paragraph are drawn from a preliminary report by Wahito Kabaire, WFP/DMA. 96 result for Lesotho, only 4 of the 17 countries with less stunting have higher levels of GNI per capita. 4.42. Recently updated growth standards change Lesotho’s anthropometric indicators as expected. The international growth standards used to benchmark anthropometric standards have ~hanged.’~ Since the new standards represent a better benchmark against which to assess child growth, the DHS data have been used to calculate anthropometric outcomes using both the former and new growth standards. According to WHO, the new standards are likely to reveal increased stunting throughout the age range; underweight will increase significantly for children aged 0-6 months and then fall, and wasting will substantially increase for young infants up to 70 cm before falling. The results for Lesotho are consistent with expectations (Table 4.7). The observed rate of stunting has increased at almost all age ranges. Children under six months show far higher underweight rates, but older children have lower rates. Wasting increases sharply for children under 6 months, compared to the old standard, but falls sharply until the age of 2 years. 4.43. Results calculated using the new WHO growth standards indicate that malnutrition continues unabated, with stunting levels at 41.7 percent. In response to the 2006-07 drought, a national nutrition survey of 8,15 1 children aged 6-59 months was fielded in the last 2 months of 2007. Results indicated chronic long-term food insecurity as opposed to the effects of the recent drought. Stunting rates exceeded 42 percent in Berea, Qacha’s Nek, Mokhotlong, and Tsaba Tseka. Rates of underweight at 13.8 percent and of wasting at 2.3 percent were lower using the new standards than had been shown in the past. However, these rates were consistent with the impact of the new growth standards as opposed to real improvements in outcomes. Underweight rates exceeded 17 percent in Qacha’s Nek, Mokhotlong, and Tsaba Tseka. Similarly, wasting rates were among the highest in these districts. 4.44. Evidence indicates an unhealthy diet transition is also taking place, more akin to many middle income countries. 42% of women in Lesotho have a body mass index in excess of 25, and are therefore classified as overweight or obese. These women are more like to live in urban areas, be older, and wealthier. Unchecked this trend to obesity will result in high health care costs due to the increased prevalence of diabetes and cardiovascular diseases. 4.45. It would appear that there are very few if any core nutrition interventions in Lesotho. This is very costly to undernourished children themselves and to the future economy in direct losses in productivity from poor physical status, indirect losses from poor cognitive function and deficits in schooling, and losses due to increased health care costs. The interplay between HIV/AIDS and undernutrition exacerbates both problems. Malnourished individuals are more susceptible to HIV, and antiretroviral drugs are less effective for HIV-infected individuals who also are malnourished. Nevertheless, during the 2007 Annual Joint Review, very little mention was made of nutrition interventions. The increasing rates of overweight and obesity complicate public health messages related to nutrition. Diet Quality: Micronutrient Status 4.46. Anthropometric measures give little information on diet quality. Rather, they indicate insufficient food consumption, and/or poor health, sanitation, and environmental conditions that 54 It was recognized that the original standards were based on a sample of one developed country’s children who were not necessarily raised according to standards recommended today. Consequently, a multicountry program was initiated including children who were reared in environmental conditions that minimized constraints to growth such as poor diets or infections, and whose mothers did not smoke during pregnancy and practiced exclusive breastfeeding for the first six months. 97 lead to persistent ill health and compromised food intake and dietary absorption. Diet quality outcomes are better captured by micronutrient statistics, but in Lesotho these are scant. The only recent blood-tested, national-level statistics are the anemia statistics in the DHS 2004. ‘ 4.47. Knowledge, rather than income, plays a major role in reducing anemia.55The DHS 2004 shows that 27 percent of women in Lesotho exhibit some level of anemia, with urban levels being far higher at 38 percent than rural levels at 24 percent. Rates are higher for more educated women, which may correlate with an urban residence and lifestyle. Geographically, rates are lowest in Qacha’s Nek (17 percent) and Thaba Tseka (2lpercent), and more generally in the Mountain zone at 12 percent. The highest rates of anemia are found in Maseru, Quthing, Berea, and Leribe. The prevalence of anemia increases not only with wealth-30.2 percent for the wealthiest against 20.8 percent for the least wealthy-but also education: 26 percent for those with no education against 28.3 percent for those with secondary education or more. These findings may be due to underlying correlations between eco zones and wealth and education. 4.48. The Mountain areas tend to have lower wealth levels but are more likely to rear animals as a source of livelihood. Animal-source foods are rich sources of bio-available iron. Berea and ’ Leribe are the “cereal basket” of Lesotho. Grain-based diets often are deficient in micronutrients such as iron. More educated women are likely to have formal employment and urban residences. They adopt different food consumption patterns based on their lifestyles and may consume fewer dairy products, fruits, and vegetables. Pregnant and breastfeeding women show slightly lower anemia levels, despite higher iron requirements, which may indicate some success of the MoH’s promotion of iron supplements to pregnant women. Anemia levels are more than 35 percent among HIV-positive women. Iron supplementation also should be considered as part of the HIV/AIDS care strategy. 4.49. Anemia in children is a major problem. Forty-nine percent of children aged under 5 years are anemic: 50.5 percent of boys and 46.7 percent of girls. However, for severe anemia, the rate for girls is 3 times that of boys. Tsaba Tseka again shows 1 of the lowest recorded rates of childhood anemia at 29 percent. However, Mokhotlong, also in the Mountain zone, has the second highest rate of anemia at 61.9 percent. This dichotomy likely is linked to the distribution of livestock. Tsaba Tseka has almost three times as many goats and twice as many cattle as Mokhotlong, thus increasing its supplies of dairy products. Overall, no relationship of wealth to iron status for children appears, indicating that costs of micronutrient-rich foods are unlikely to be the issue. Combined with the inverse relationship of women’s iron status and wealth, this childhood statistic suggests that knowledge rather than income is the issue for many Basotho. This, in turn, indicates that the key policy question is the role of public health programs in nutrition education. D. POLICY OPTIONS FOR FUTURE FOODSECURITY 4.50. Given the repeated droughts with consequent international food aid responses, food security was rightly identified in the last Poverty Reduction Strategy (PRS) as a key priority for Lesotho. As a result, the Lesotho Food Security Policy (LFSP) incorporating strategic guidelines was drawn up in 2005. It was built on the widely accepted definition of food security agreed at the 1996 World Food Summit .56 The LFSP correctly identified many of the roots of 55 All data used in the anemia analysis are as published in DHS 2004. ’‘ According to the summit, food security is defined thus: Foodsecurity exists when all people, at all times, have physical, social and economic access to suflcient, safe and nutritious food to meet their dietary nee& andfoodpreferences for an active and healthy life. 98 food insecurity in Lesotho. The policy’s core objective is to improve the adequacy and stability of access to food at the household level. Subobjectives are to improve the use of food within the household and to increase the adequacy and stability of food supplies at the national level. These objectives are to be achieved through eight strategies, supported by improved food security and vulnerability information systems and a strengthened institutional framework. The eight strategies are to: e Use employment promotion to ensure sufficient and stable access to food a Effectively monitor the impact of employment polices on food security e Promote agricultural and food production a Promote infrastructure and services to support livelihoods a Promote public transfers and social safety nets e Mainstream HIV/AIDS within the Food Security Policy to maximize the impact of policy measures on households affected by HIV and AIDS a Effectively manage commercial food imports, food aid deliveries, and food stocks Promote improved use of food at household level. 4.51. Agriculture is unlikely to be a major engine of growth for Lesotho, but given its importance to the rural poor, it must not be neglected. Ultimately, Lesotho may have limited potential to vastly increase the production of its staple grains, maize and wheat. These are relatively low-value crops and are produced at lower cost in SA. Lesotho has domestic buyer status in SA so is not at risk of losing market access when the market is tight. Lesotho may be better placed to improve productivity where it is possible to match the SA price plus transport costs, but to look to source the bulk of its needs from SA. 4.52. Lesotho could be more food secure by reducing risk and budget exposure through more extensive use of market instruments. The givens of Lesotho’s situation are the: (a) level of cereal production, (b) level of poverty that constrains effective demand for food, (c) risk exposure to the South African cereal production and the economy, and (d) rising global price of cereals, which is forecast to be a trend rather than a spike. 4.53. Lesotho faces the real potential of increasing future food-availability shortfalls as a share of total demand. These will need to be met by imports (for those who can pay cash for their food) and increased food aid or by other GoL social protection interventions for those who have inadequate income. When the required intervention is food, GoL could use market instruments to safeguard itself against the price and budget volatility. One option would be for GoL to use futures contracts through the SAFEX to purchase grain in the event of crop failure. Taking this precaution would ensure food availability for those who are unable to purchase through the market and to reach those in more remote areas with market limitations. Having this back-up food supply would reduce the unacceptable delays in mobilizing international food assistance in crises, and reduce the irreversible negative impacts of persistent hunger on very young children. Currently, SAFEX instruments are used by the two private mills, but this arrangement responds only to effective demand. Given sharply rising grain prices over the last 2 years, it is likely that the gap between food requirements and effective demand will widen, resulting in more of the population being dependent on assistance in some form. 99 4.54. Options contract were used very effectively by Malawi in 2005 to provide risk insurance. In September 2005, the government of Malawi purchased an over-the-counter option contract for 60,000 metric tones of maize for $17 million with a premium payment of $1.53 million. The contract was specified for onsite delivery, thus insuring the transportation costs as well, important in these days of rising oil prices. Food shortages grew more severe in Malawi during November and December 2005. With prices rising, government exercised its option with on-site delivery. Delivery performance was more effective than other procurement strategies used, and the spot price of maize on SAFEX rose US$50-90 per metric ton above the contract ceiling price. Estimates indicate that a call-option strategy would have resulted in substantial cost savings in 2 of the last 3 years with the option cost being only lost once. 4.55. The key issue in using this instrument in Lesotho is who pays the cost of the option contracts. If the Government does not pay for food aid, instead relying on donors, the government is indifferent to the price, and sees no need to hedge the risk. But food aid provided by donors, while effectively free, comes with a delay which is costly to the food insecure who may sell productive livelihood assets in the meantime. Not only will this delay their recovery from the crisis, but children below 2 years of age in affected households will never achieve their true potential in life. GoL needs to examine these instruments in the context of its own social protection strategy (including the food which is procured for school feeding) and negotiate with ‘ donors for a better solution than the current process of food aid appeals and delayed crisis response. Nutrition Programs 4.56. Implementation of comprehensive nutrition interventions targeted initially at pregnant women and children under 2 years of age by the publicly financed health sector is badly needed to reverse the negative nutritional trends. As yet, GoL seems to give little effective attention to the consistently poor nutritional outcomes in Lesotho. The first two years of life are the key window of opportunity, because malnutrition during this period causes irreversible damage. A key focus needs to be on pregnant women, including micronutrient supplementation and nutrition education on infant and young child feeding. Many countries have delivered this education via community-based nutrition workers. In Bangladesh each of these workers delivers services to 250 households. The workers are responsible for identifying all pregnant women, providing antenatal counseling and advice on safe delivery, exclusive breastfeeding (including colostrum within the first hour), infant weaning, micronutrient supplementation, and diarrhea management. A network of community workers needs to be supported by a well-functioning referral system to local-level health facilities for children who fail to thrive. During field visits in Lesotho, some level of community nutrition intervention appeared to be in place, but how it was supported was unclear, and the referral mechanism was weak to nonexistent. The local hospital did not have the right equipment to measure and weigh small children. 4.57. Good nutrition for infants starts with good maternal nutrition in pregnancy to avoid low birth weight deliveries. The Gambia very successfully implemented a baby-friendly community initiative. To improve pregnancy outcomes, communities were mobilized to better care for pregnant women and encourage women to breastfeed their newborns. Following the training, communities had to identify an initiative they would implement to be certified “baby friendly” with a certificate presented by the Vice President of the Gambia. Many communities came up with systems of “maternity leave” in which a pregnant woman’s fields were cared for up to the month preceding delivery and immediately following delivery. Other mothers built baby shelters near their fields to protect babies from the direct sunlight. Today, as shown in Table 4.6, 100 Gambia ranks 2"d among a number of sub Saharan African countries in terms of stunting despite ranking 30 in terms of GNI per capita. Disaster Risk Management 4.58. Reflecting the significance of natural hazards in rural Lesotho, the government developed a National Action Plan for Capacity Development in Disaster Risk Reduction (2007-2015). In particular, reflecting multiple devastating droughts over the last 3 decades (including 1983-84, 1991-93, 1994-96, and 2002-04), in the plan, drought preparedness is a key concern. Drought is likely to have major negative implications on the rural economy, especially in the Southern Lowlands. This strategy document recognizes that although drought must be perceived as a regular phenomenon, Lesotho's vulnerability to its impacts is aggravated by chronic food insecurity, poverty, low agricultural productivity, and limited watershed development. The action plan represents an effort to reinforce the National Disaster Management Plan toward emphasizing building institutional capacity for disaster risk reduction and more proactive engagement. Social Protection 4.59. Given the time required to generate employment, revitalize agriculture and improve agricultural incomes in Lesotho in order to reduce poverty, social protection mechanisms may be needed for the chronically poor. These need to be part of a comprehensive social protection strategy which addresses chronic poverty, food insecurity, and emergency situations. Social protection mechanisms, including public transfers and safety nets, were envisaged in the LFSP but as yet a comprehensive strategy has not materialized. These mechanisms are even more relevant now with rising grain prices. Social protection programs can increase effective demand when transfers are in cash, or wages for public works in areas with market access, thus supporting markets rather than undermining them. 4.60. There are many potential elements that could meaningfully contribute to such a program. Designing an over-arching strategy but initiating individual elements in phases, with geographic priorities, is the preferred option to ensure fine tuning. Priority geographic areas include Qacha's Nek, Mokhotlong, and Tsaba Tseka, where malnutrition rates are high. These areas have more limited market access and are in the more difficult agro-ecological zones, rendering them more vulnerable to agricultural failure. Evidence suggests particularly poor employment opportunities in these districts. In the next chapter, this issue is developed more fully in the context of accelerating structural transformation and social change in rural areas. 101 Tables Chapter IV Table 4.1 : Agro-ecological conditions and farming systems Zone Mountain Rural Senqu Foothills Lowlands characteristics (and plateau) River Valley YOof total land 59 9 15 17 area YOof total 22 5 13 38 population (estimates from HBS 2002-03) Climate Rain 600-900 Rain 600-700 Rain 800-950 Rain 600-800 mm/year. mm/year. mm/year. mm/year. Drier and parched soils. Frost and cold snow year round. Milder at lower plateau. Main climate Early and late frost, Frost and snow Drought and hail. risks snow. Short growing a t times. Severe . FI ood s Drought, hail, and soil season. Strong winds, drought. erosion/ depletion hot sun, hail in plateau. Main crops Lentils, maize (poor Su mmer wheat, Maize, beans, Maize, winter wheat, quality), wheat, peas, good fruit peas, sorghum, sorghum, peas, beans, potatoes, oats, stone production, good fruit very good fruit fruit; sorghum, sorghum, beans, production, production, and vegetables, fruits in potatoes, peas winter wheat vegetables plateau Rangeland Good alternative Moderately Good range, moderate and grazing, willow denuded Seriously eroded with degradation, some vegetation and popular rangelands, deep dongas and very wetlands intact, conditions trees, limited poplars in poor rangelands, many denuded range in soil erosion watercourses, woodlots and treelots plateau village woodlots Summer Good access for cattle Within and Predominantly Most villages have no grazing a t high area cattle below village village grazing, access to cattle posts posts; above village in access to cattle apart from in Mafeteng plateau posts by some villages Other Rampant livestock Good potential for gravity-fed Rapid settlements with theft, esp. in Quiting; crop fields being livestock kept home in irrigation converted to house sites plateau Sources: Adapted from F A 0 2007 and KOL 2007 with additional data from HBS. 102 Table 4.2: Potential average cost of government subsidy for input costs (2007/08) on 180,000 ha summer crops (US$) Input Total cost Input cost Total cost Subsidy Subsidy Crop cost Ha ($/ha) (L/ha) (US$) (Maloti) (US$) (Maloti) Maize 170 1190 150,000 25,500,000 178,500,000 7,650,000 53,550,000 Sorghum 170 1190 30,000 5,100,000 35,700,000 1,530,000 10,710,000 Total 180,000 30,600,000 2 14,200,000 9,180,000 64,260,000 Sources: Authors’ calculations. Note: Based on an exchange rate of R7 to US$I. Maloti is pegged to the Rand at 1 to 1. Table 4.3: Calories from own production by wealth and location, 2004 Livelihood zone Poor Middle Well-off Weighted average Southern Lowlands 26 36 49 33 Northern Lowlands 22 48 55 37 Foothills 26 42 54 34 Senqu River Valley 18 32 33 24 Mountains 14 42 55 27 Peri-urban 12 32 37 24 Weighted average 21 40 49 Source: LVAC 2004 cited in Marsland and others 2004. Notes: Table shows proportion of calories from own production in a “normal” year. A “normal” year is defined as 1 of the last 5 years that is “typical” for the area, that is, when food access was neither exceptionally good nor exceptionally bad. For more details on this concept see SCF 2000. Note that the LVAC program used the same or similar names for livelihood zones that had been used by BoS, but the boundaries were not identical. In 2006 LVAC changed its boundaries so that its rural areas have same boundaries as those of BoS. BoS has not defined peri-urban areas. Table 4.4: Nutritional outcomes, 1992-2007 1992 2000 2002 2004 2006 2007 Stunting 33 45 47 38 38 4Ia Underweight 16 18 22 20 20 20 Wasting 2 5 12 4 4 6 Sample size 4687 3737 na 1620 847 430 Sources: End-Decade Multiple Indicator Cluster Survey (EMICS) 2000, 2002 Core Welfate Iindicator Questionaire, 2004 DHS, 2006 WFPILVAC CHS draft report, simple average of figures for beneficiary and nonbeneficiary households, 2007 CHS WFPLVACFood and Nutrition Coordination Office (FNCO) nutrition survey. Note: a Calculated using new WHO growth standards. Table 4.5: Regional nutritional outcomes, 2002-06 1 Livelihood zones 1 Stunting Wasting Underweight I Lowlands 28 33 33 4 4 6 14 14 23 Foothills 35 39 34 3 4 5 20 21 Mountains 38 45 42 4 4 2 21 27 Senqu RV 32 45 54 3 10 01 15 27 18 103 Table 4.6: Lesotho Stunting Relative to GNI Compared to other SSA countries Ranking Country GNI per Stunting Country Ran king GNI capita (percent of Stunting (2005) children) 1 Botswana 5530 16 Senegal 1 2 Eq. Guinea 5410 19 Gambia 2 3 Gabon 4390 21 Cote D’Ivoire 3 4 Namibia 2960 21 Gabon 4 5 Swaziland 2210 22 Togo 5 6 Angola 1410 23 Botswana 6 7 Djibouti 1010 23 Djibouti 7 8 Cameroon 1000 24 Namibia 8 9 Congo 950 26 Zimbabwe 9 10 Lesotho 930 30 Ghana 10 11 Cote D’Ivoire 840 30 Guinea Bissau 11 12 Senegal 700 30 Kenya 12 13 Sudan 650 30 Swaziland 13 14 Mauritania 580 31 Benin 14 15 Kenya 540 32 Cameroon 15 16 Nigeria 520 34 Sierra Leone 16 17 Benin 5 10 35 Guinea 17 18 Zambia 500 35 Mauritania 18 19 Ghana 450 38 Congo DR 19 20 Guinea 440 38 Lesottr o 20 21 Burkina Faso 43 0 38 Mali 21 22 Chad 430 38 Niger 22 23 Mali 3 80 38 Nigeria 23 24 CAR 350 38 Tanzania 24 25 Togo 350 39 Burkina Faso 25 26 Tanzania 340 39 Eq. Guinea 26 27 Zimbabwe 340 39 Liberia 27 28 Mozambique 310 39 Uganda 28 29 Madagascar 290 41 CAR 29 30 Gambia 290 41 Chad 30 31 Uganda 280 41 Lesotho 31 32 Niger 240 41 Mozambique 32 33 Sierra Leone 220 43 Sudan 33 34 Guinea Bissau 180 45 Angola 34 35 Malawi 160 47 Ethiopia 35 36 Liberia 120 48 Madgascar 36 37 Congo DR 120 48 Malawi 37 38 Burundi 100 50 Zambia 38 39 57 Burundi 39 40 SSA 743 37 SSA 40 Source: WDI 2007. State of the World’s Children.2008. 104 Age group, 0-5 6-1 1 12-23 24-35 36-47 48-59 months Old New Old New Old New Old New Old New Old New Stunting 15 24 40 25 43 47 43 52 44 48 45 49 U-weight 2" 17 21 11 23 13 23 20 22 16 25 20 Wasting 4 13 4 8 4 5 3 ,3 3 3 4 4 Table 4.8: Household Livelihood Sources, by district (% of households reporting each source) District First O Y Second YO Third YO Butha-Buthe Food crop production 71 Remittances 58 Casual labor 46 Leribe Casual labor 51 Food crop production 34 Begging 34 Berea Casual labor 43 Pension 27 Food crop production 24 Maseru Casual labor 42 Gifts 39 Remittances 34 Mafeting Gifts 42 Food assistance 41 Casual labor 40 Mohale's Hoek Gifts 44 Casual labor 37 Food assistance 33 QWng Casual labor 45 Food assistance 43 Food crop production 31 Qacha's Nek Food assistance 35 Casual labor 33 Gifts 32 Mokhotlong Cash crop production 44 Gifts 44 Casual labor 40 Thaba-Tseka Casual labor 40 Small business 32 Gifts 28 Source: WFP/F 3 2007 105 5. RURAL ECONOMY, EMERGING LIVELIHOODS, AND SOCIAL DYNAMICS 5.1. Lesotho’s recent economic and social transformation has provided opportunities for some rural households, but risks and threats for others. Some key underlying drivers of change have been the retrenchment of (predominately male) workers in South African mines and industries, the impact of HIV and AIDS at household and community levels, and the increased employment and migration of (predominately female) workers in factories in urban Lesotho. These changes have affected household structures, gender and generational relationships and traditional social sharing mechanisms, including systems of sharecropping and access to land for men and women. They have also affected income levels and employment patterns among men and women. 5.2. A distinguishing feature of the rural areas is its heterogeneity in terms of income and asset patterns across social groups and livelihood zones which is a result of a variety of factors, including differential access to employment and remittances, unequal access to land and farm assets, locational factors and diverse impacts of climate risks, as well as differential access to markets, economic and social institutions.” 5.3. In this chapter, we analyze determinants of economic outcomes in rural areas using a livelihoods approach, considering the sources of incomes, farm assets (including land and labor at the household level), and how they affect decisions about household resource allocation. The chapter also highlights the effects of economic and social change on rural subgroups and household structure (age and gender), and the implications for poverty reduction. In the final section, policy implications are suggested including social protection (started in the previous chapter on food security). A. INCOMES AND ASSETS OF RURAL HOUSEHOLDS 5.4. A number of studies have addressed household income sources, activities, and livelihoods in rural areas. Unfortunately, no national representative survey covers this subject in depth and with sufficiently high-quality data to map out detailed livelihood patterns across time and location. In particular, the HBS data do not contain production and income data for agriculture. We have highlighted that close to 50 percent of all rural households cited farming as their main source of income in 2002/3; the next most common source in rural areas was private sector wage employment (20 percent of all households, Table 5.2). Among the poorest consumption quintile; 60 percent reported farming as their main source of income. Farmers are generally older and less educated; wage employees are more likely to be young and better educated. Owing to the factors discussed in the previous chapter, agriculture most often does not provide a sufficient source of income, and it is for most households a subsistence activity. A small commercial sector exists, including in wool and mohair, livestockkattle production, and fruit and vegetable gardening with some but limited potential. Due to limited farm-income, most farm households engage in several kinds of non-farm income-generating activities and depend on several sources for livelihoods. This trend has increased since 1994. (Boehm 2004, Gill-Watson 2005, LRAP Brief 5 2004, Leduka and Gay 2007). A small panel survey found that 40 percent of ’’One reason for the heterogeneity of rural areas is that in the survey data, “rural areas” refers to an administrative classification that does not follow population density. Thus, the University of Lesotho is located in an area that is classified. The HBS does not correct for this by using a population-based classification. 106 rural households in 2002 relied on 2 or more sources of income, compared to just over 30 percent in 1993. The same survey also found that income from informal work such as small enterprises, hawking, and manufacturing produce increased by 58 percent (LRAP Brief 5 2004). They also found that whereas cattle size was unchanged, many households had reduced their herds of smallstock (except ownership of pigs, which had increased). Overall, households had increased incomes from livestock by 65 percent over the period. Illicit income sources such as sale of dugga (marijuana) and livestock theft also are reported to be increasing, but relatively little is known about them (Leduka and Gay 2007). 5.5. Nonfarm activities have developed less in rural Lesotho than in many other African countries (World Bank 2005a). Most rural income activities are farm-related wage work or rural services and petty trade, often spin-offs from agricultural activities (beer brewing, fuelwood sale, water collection, bartering, simple construction This lag in nonfarm rural activities is explained by the country’s proximity to South Africa (SA) and open borders, which have meant that many products that otherwise might have been produced in rural areas are imported from SA. Except for draft power and labor, much of the agricultural inputs and implements come from SA. The development of private input supply services in the agricultural sector is also limited if not non-existing; this is partly an outcome of past agricultural policies (chapter 4). Moreover, most agricultural produce, such as wool and mohair, the two key export commodities, leave the country without being processed. Only a small cottage industry utilizes these products. Tourism has remained basically unexploited despite the potential of Lesotho’s scenic Mountain areas. From the late 1980s to the early 2000s construction of the Highland Water Scheme provided temporary income for many rural laborers. 5.6. Some of the emerging livelihood and coping strategies have specific gender and generational implications. Besides retrenchments of men from the mines in South Africa and the new urban employment opportunities for women, there are more subtle rural livelihood transitions observed at the local level. Some of these reinforce traditional gender relationships, whereas others are changes in past gender roles. Traditionally, rural tasks have been segregated by gender, and there were women’s tasks and men’s tasks. (de Lange, 2007) In some instances, men have chosen to enter traditional women livelihoods domains such as vegetable gardening, small stock tending, small-scale services, and petty trade (Leduka and Gay 2007, Boehm 2004, Gill-Wason, 2005). In others case, women may engage more in traditional male activities, including livestock and other farm related-work. This happens most often where there is no male in the household. 5.7. Land ownership is not necessarily the key factor to secure sustainable rural livelihoods in Lesotho. Most rural households (70 percent) have land. Those who do not own any land are more likely to be better educated than other households and headed by absent males. Looking at the distribution of land across consumption quintiles, most noticeable is the lack of correlation. The size of owned land is distributed relatively equally across consumption quintiles with both rich and poor owning small and larger pieces of land (Figure 5.1). Even if the direct income from agriculture is not high, its overall significance for the rural economy needs to take into account its importance for subsistence consumption as well as other multiplier effects related to the local beer economy, inputs to agro-processing, small-scale trade, and farm employment. There are for example approximately 60,000 herd boys, 12 000 shearers, and 10,000-15000 farm workers. 107 Figure 5.1: Rural land distribution by consumption quintiles, 2002/03 100% 80% 60% o2to5ha 0 1 to 2 ha 40% L less than 1 ha 20% 0% Poorest Second Third Fourth Richest All Rural Source: HBS 2002/03 and authors’ calculations 5.8. Land ownership is significantly skewed in favor of older heads of households, and against de facto female-headed households (where the husband is absent).- (Table 5.2). Close to 50 percent of the household heads in age groups above 65 possess more than 2 hectares (ha). In contrast, among heads aged 25-34 and 3 5 4 4 , only approximately 30 percent own 2 ha or more. The latter cohort also contains a large number of landless (approximately 40 percent). This is the age level at which a person would be expected to have taken up farming if it were to become a main livelihood. In age categories 55 and above, only approximately 20 percent are landless. De facto female-headed households are more prominent among those who own no land, but other wise, there seems to be little correlation between gender of head and access to land (land size held by household). The ratio of male heads and female heads across land sizes is fairly equal, which is surprising in relation to the general discussion in Lesotho about women’s limited access to land. These data indicate that female heads of households do hold land or at least have access to their own land or family land almost on a par with male heads. The figure is likely to hide the fact that women’s access to land was in the past mostly mediated through the men or the men’s family, and providing women with weaker access rights, in particular if the husband is absent, gets a divorce or dies. These issues are being addressed in the new legal reform, but the reform of the formal law is likely to only gradually affect customary and local practice. 5.9. Landlessness has increased steadily over the last decades. In 1950, 7 percent of all Basotho households were landless (KoL 2000). The HBS find that the level of landlessness increased to 30 percent of all households in 1994-95 and 41 percent in 2002-03. Among rural households, 22 percent had no land in 1994-95, which rate rose to 28 percent in 2002-03 (HBS). The poverty surveys of 1990, 1993, and 1999 find a similar trend with reported 23 percent, 33 percent, and 4 1 percent landlessness, respectively (Leduka and Gay 2007). Sharecropping 5.10. Sharecropping in its various forms has increased rapidly in Lesotho since the 1970s. During that decade, less than 20 percent of fields were sharecropped; the figure has risen to nearly 70 percent in the 2000s. Sharecropping represents a central characteristic of the rural economy (Leduka and Gay 2007, referring to the Lesotho Agricultural Census 2000). Sharecropping has always been a key form of cooperation among Basotho farmers. The traditional form of sharecropping made it possible for those such as the (young) landless household heads to have access to land and for household heads such as older females with access to less labor or capital inputs to have their land tilled or other operations carried out. 108 5.1 1. Sharecropping arrangements take many different forms, which are not well described in the literature. Contracts often are negotiated through unequal power relationships, although many of them are entered into among extended family members. Boehm (2004) describes variations that include sibling relationships (contracts between sisters and brothers), partnerships (as between a tractor-owner and a cart-owner), and patronage systems (as between widows and men). Network systems between kin can be more egalitarian.59 5.12. The high rate of sharecropping is likely to be an effect of different types of asymmetry in ownership of land, and in other assets such as cash, labor and farm implements, between different types of households and individuals, as well as custom and tradition. Many households with land do not have adequate labor available, nor do they have animals of implements to substitute. For instance, hlly one-third of households that own land do not have any adult males available to work the land full time and nine percent have no available adults, male or female. (Table 5.3).60 The latter households tend to have heads over 65, who is more likely to be a woman. Households are short of labor not because one adult has migrated, but rather because the adults are missing - perhaps owing to premature mortality. 5.13. Even households with access to land have not invested in equipment; as noted above, agriculture in Lesotho is not mechanized or modernized. Only 50 percent of land-holding households in rural areas own ploughing implements so they need a lot of labor to produce anything. (Table 5.2).6’Redundant or retired miners with a small pension-ar with retirement in the form of a lump sum that has been spent- or widows have limited cash to invest in the farm, but they can use the income earned from renting out their land. 5.14. The arrangements governing Lesotho’s sharecropping are not well known. It is not clear which types of sharecropping systems represent optimal allocation of resources and for whom. Decisions at the household level about allocating scarce labor and assets to nonfarm activities are likely to represent rational choices within given constraints that also reduce risks through diversifying income sources. To design new pro-poor policies, it is key to know more about the types and efficiency of these arrangements. Sharecropping potentially benefits both the land poor (enables them to farm) and older heads (increases their income) and helps involve both men and women-younger and elders-as farmers by addressing asymmetries in asset ownership (Turner 2004 and 2003, Boehm 2004, Lethola 2005). However, the structure and impacts of these various types of arrangements and how they are made should be investigated further. While these arrangements may represent a rational response to structural circumstances, they often may be ineffective from the perspective of sustainable agricultural development. For example, because crop-sharing contracts often are short term, they discourage longer term investment in land improvements and soil conservation. The result may be lower yields, lower income at the farm level, and limited surplus from production (Leduka and Gay 2007). 5.15. As indicated in the previous chapter, there is a declining trend in the area cultivated and production for most major crops, including for peas and beans. There has been a market 59 Typically, sharecropping is an arrangement whereby 2 or more heads of households decide to pool the resources required to farm for 1 season, including land, labor, traction power, and inputs. Traditionally, the contract often would be sanctioned by the chief (Boehm 2004). In sharecropping, the harvested crop is shared according to conditions determined in the contract. In the past, the conditions often would be on a 50-50 basis, but today there are many variants over the way produce is shared-ften to the detriment of the one who rents the land. 6o An available adult male is defined as man between 15 and 65 who is present in the household (not living abroad), does not have wage employment, is not an employer, or disabled. 6’ In the Lesotho Agricultural Census 1999-2000, the Bureau of Statistics found that 40% of households had ploughs; 20% owned planters; and 15% had harrows. 109 increase in the rate of land left fallow (about 30-40 percent of the arable land in any one season) reflecting decreased capability and interest of many rural households in crop cultivation due to a variety of perceived constraints, including drought and lack of remittances to buy farm inputs. While the focus of government support in the past was on grain production, more recent efforts by 3vemment and donors have been on horticulture (e.g. keyhole gardens). For many poor hoi. ,eholds these vegetable gardens have become key sources of household nutrition and food security (LRAP Brief 6 , 2004, PRS Progress Report, 2007). Livestock 5.16. Livestock is an important asset of rural households and contributes significantly to rural income and sustenance in many households. The demand for dairy products, pig, poultry and other livestock products is still mainly met through imports from South Africa. Cattle production is the largest contributor to the national livestock economy. The total herd size of cattle increased slightly over the last decade, indicating that some farmers have been able to overcome stock theft and disease.62But estimates of the national herd size of cattle, goats, and sheep over the last decade (1994/95-2004/05) fluctuate greatly from one year to the next. A trend toward a larger cattle herd size is indicated. The trend for goats and sheep is unclear. The number of pigs and poultry has increased (KoL 2007b, Table 9-1 1). The HBS shows that the percentage of households owning cattle, goat, and sheep decreased nationwide between 1994-95 and 2002- 03, while poultry increased (Table 5.3). These changes were found in all livelihood zones in the country. The most striking change is being observed in the rural mountains (15 percent reduction among households owning goats and 24 percent reduction in sheep). There is a significant increase in households holding poultry (Table 5.3). Wool and mohair are the main cash generating activities for smallholder farmers in Lesotho, however, production and exports have gradually declined.63This relates to decline in number of animals, wool yield and quality and value of the wool and mohair produced. It is also argued that overgrazing and stock thefts have hampered progress. 5.17. Rural households own livestock for purposes of drought power, payment in kind, or sale. Livestock also is held for social value, cultural ceremonies, and status. It serves as an insurance mechanism of last resort and as a store of wealth that potentially gives a dividend (insurance against crop or other income failure). Although livestock has decreased in importance for many households, its overall significance and contribution to the rural economy has been increasing over the last decades. Livestock is accumulated through the lifetime of the household. Households headed by elder heads more frequently hold animals than younger heads. Small sample panel surveys in 1993 and 2003 show that the households often had kept their cattle but reduced their herds of small stock. Ownership of pigs had increased. Income from livestock had increased by as much as 65 percent over the period, with reference to increased income from pig rearing, poultry, egg production, and sale of milk and cattle products. 5.18. The rate of ownership of livestock is distributed fairly equally across consumption quintiles. The type of livestock kept varies with location; sheep and goats being next to cattle in importance. There also are significant numbers of horses and donkeys. Pigs and poultry have increased substantially in importance at the household level. The first quintile might be less likely to own livestock, but there seems to be limited difference among the other quintiles (Table 5.6). ~ ~~ The total number of cattle is reported to be 680,000 head. There are 940,000 sheep and 780,000 goats (KoL 2007a). Pigs and poultry have become increasingly important. There is also are considerable national herds of horses and donkeys, which remain important for transport among poor households. In the 1930s, wool and mohaircontributed almost half of the total national exports (according to PRS Progress Report, August 2007). 110 Table 5.7 shows that cattle are common across all regions but are held most frequently among households in the mountains (54 percent). Sheep also are most common in the mountains (30 percent of households have sheep), whereas pigs are rather rare (12 percent), compared to the rural average of 20 percent (sheep) and 21 (pigs). A significant share of the cattle herd-22 percent-is held in urban areas. B. CHANGE SOCIAL 5.19. The profound changes in the structure of the economy and of livelihood pathways for men and women-young and old-experienced over the last decade affect gender and generational relationships and the formation and structure of households and communities. To some degree, old structures are eroding, including traditional household and family structures, and new ones are being shaped by individual and collective actions (as well as by wider political and economic relationships). Young men are faced with different challenges than their fathers had. Today, labor migration to mines is less of an option, and access to wage labor often depends on higher levels of education. Young women may find new opportunities in urban labor markets, but also confront different and often more challenging social and cultural contexts than did their mothers. Research is not conclusive on the degree to which these changes affect the wider society. However, several of the trends described here are fairly obvious and are noted in a variety of village and smaller sample studies (Boehm 2004, Turner 2004, Gill-Wason 2005, Magrath 2005, Leduka and Gay, 2007).64The high and increasing rate of HIV/AIDS is having major impacts on many rural households and communities. The epidemic is having growing impact on the livelihoods of rural households, loss of productive capacity, marriage breakdown during sickness, and increase in number of widows and widowers. 5.20. Several observers perceive the changing gender roles and relationships as more positive than negative for many women (Magrath 2005). If women have the minimum educational requirements and live in, or can move to, urban areas, the structural changes in the economy offer them opportunities to secure wage jobs. Associated with this trend is delayed marriage. The more varied household formations may reduce women’s economic dependence on men, even if some of these women also may be struggling to survive and support their children. In addition, urban women face higher risk of HIV/AIDS, due to both their limited power to protect themselves and their particular social status and role (migrant workers, single mothers, widows). Some also establish semicommercial sexual liaisons (Gill-Wason 2005). Even though gender law and policies have become more progressive, women continue to face constraints of customary law and traditional gender attitudes and norms. Change in values and praxis takes time and depends on many other factors than simply changing the law. 5.2 1. Shifting gender roles may appear positive for women in generating income and in creating independence from men, but, this shift in economic roles may not always truly empower women due to added burdens in maintaining the household combined with continuing domestic responsibilities and caring for children. Their relative “freedom” also may create feelings of disempowerment among men (LRAP Brief 4). Another observation is that men, after being laid off from the mines, see few opportunities but to enter traditional female occupations such as gardening and street trading, possibly pushing women out of these professions, while not obtaining a salary sufficient to cover household requirements (Boehm 2004, Leduka and Gay, 2007). This reality also may undermine masculine identities and self-esteem. Finally, an ’ 64The nationwide studies of Sechaba (1994,2000) covered household, gender, and intergenerational dynamics to a lesser degree. 111 increasing number of elderly women whose expectation had been to be cared for in old age find themselves caring for orphaned grandchildren or sick married children (“granny parenting”). 5.22. Observers suggest that social, cultural, and gender roles have changed the most in communities and families that are under the most severe economic pressures. The transitions are being pushed by economic pressures or choice of new development paths. Magrath (2005) suggests that gender relations are being renegotiated between women and the men in their households who cannot find employment and who do not succeed under the present system of cultural norms, rather than among the educated liberal elite. The former attempt to transform the division of labor and gender relations, while creating new household structures and confronting traditional ownership and inheritance rules. Examples are “the woman beer brewer who pays men to fetch water and fuel for her, and the women owning property and passing it on to their daughters” (Magrath 2005). 5.23. Increase is seen in “incomplete households” due increase in the number of divorced, widowed, and not yet married men and women. (Table 5.8) The HBS data suggest falling marriage rates, increasing numbers of widows, and rising incidence of female-headed households. These data, as well as various qualitative studies (such as Boehm 2004), suggest that “complete” nuclear and extended households, which include at least one adult male, have greater success in the labor market and in agriculture. “Incomplete” households often are less successful. These are households in which one spouse is missing, or an entire generation is missing (due to migration or deaths). Such households may experience higher dependency ratios and fewer livelihood option^.^' 5.24. There was a significant decline in the married category of the rural population between the two HBS survey periods: from 51 percent in 1994-95 to 46 percent in 2002-03. In the same period the rural areas show a clear increase in the number of widowed-from 10 percent to 13 percent of the population (Table 5.8). Further, the share of rural households headed by women increased from 27 percent in 1994-95 to 32 percent in 2002-03. These trends also find support in several rural (and urban) case studies (Magrath 2005), in which increasing divorce rates also are noticed. These demographic trends should not be surprising since the driving factors-mine retrenchment, male unemployment, HIV/AIDS, and, in some locations, increased female employment-are widespread. Furthermore, these processes are not likely to be reversed in the medium term (Magrath 2005). 5.25. These trends require a rethinking of the household as the main “unit” for understanding rural society and for promoting development. Households have become more fluid and less stable as rural institutions. Although they are still critical for cooperation, production, and reproduction, it should be recognized that economic interests, chosen strategies, and vulnerability vary among individual members of the same household (Boehm 2004, Magrath 2005). 5.26. What are the implications for individual women and men and for the rural economy if these trends keep spreading? Turner (2004) suggests that the trend toward delayed marriage is alarming because this affects resource access for the younger generation and their ability to create a living in rural areas. As seen above land and other assets are skewed towards elder heads 65 Our quantitative evidence does not bear out this conclusion of Magrath’s based on qualitative analysis. Households with male migrants were still much better off, on average in rural areas in 200213, and urban female-headed households are not worse off,controlling for other, exogenous factors such as age and education. This is one reason why multivariate analysis is helpful. However, we did find that female headed households in rural areas are 10 percent poorer. 112 of households. It is important to note that men and women who do not get married will have greater problems in accessing land and other assets. Women in particular tend to gain access to land as well as wage employment through marriage and family networks. If they fail to get married, they may be forced to find alternative development paths, often through urban migration. Forced migration may not result in optimal choice of livelihoods, such as becoming sex workers. The diversity in livelihoods promotes new household structures, including increased female- headed households, and sibling-based households (Magrath 2005). These trends seem to undermine household formation, and to reinforce marriage break-up and greater family fluidity (LRAP Brief 4). Hence, the new livelihood strategies are likely to have longer term consequences for the robustness of the family household and rural institutions. 5.27. If household members often do not reside together, they are less likely to pool and share resources in an “altruistic7’manner within the household. Recent observers of social transitions have questioned the “cyclical” household model in the context of Lesotho today. They are referring to the differing livelihood pathways chosen by men and women, increasing fluidity of domestic arrangements, and fact that key family members often are not co-resident (Magrath 2005, Turner 2004). Today, households are being perceived as fields for negotiations and struggles between individual men and women and between young and old, more than as social “units” for production and reproduction (Boehm 2004). 5.28. The rural livelihood transitions have particularly negative effects on male youth, by reducing their opportunities for permanent employment, economic accumulation, and formation of stable relationships and families, and, ultimately, for moving out of poverty. Chapter 3 pointed tQ the increased insecurity of employment, especially for younger men. An increasing number of youth often find themselves in the midst of disintegrating family networks with limited access to land or jobs, sometimes irrelevant education for work, and being forced to remain within the parental household working on their parents’ fields or to migrate to town to an uncertain employment situation. They were “born too late” (Boehm 2004) to become miners. Their lack of wage-earning opportunities affects their male identity and limits their capacity to accumulate the necessary resources to become married and establish independent homes and agricultural bases. Quite a few young men remain within or attached to the family household- with only minor employment provided by the parents-but invariably under some degree of social contestation and negotiation. They are less able to develop masculinities along male parent role models as farmers or miners. Many migrate or “drift” into periurban income activities with little chance of creating a surplus for accumulation and prone to enter the illicit economy. Many of these fail to marry or delay marriage (Boehm 2004). This alienation of male and female youth may promote intergenerational conflict over land, family property, and assets. The alienation also becomes a source of frustration and anger among the younger generation, leads to abuse of alcohol, resentment of government, and fuels urban riots in times of political tension. c. POLICIES AND INSTITUTIONAL ARRANGEMENTS 5.29. The Poverty Reduction Strategy (PRS) identified six strategies to be achieved to improve agriculture and food security during 2004/05-2006/07. They were: Adoption of appropriate farming practices and timely access to inputs Development of appropriate irrigation systems Strengthening and decentralizing extension services at the area level within all districts 0 Ensuring an efficient and standardized land tenure system 113 0 Improving livestock and fodder production 0 Improving marketing systems. The government has through these policies encouraged farmers to adopt more appropriate land use practices and cropping patterns adapted to the local ecology, explore block farming (to overcome labor constraints), prioritize the use of land for horticulture and vegetables around homes, construct water tanks and introduce water harvesting, and design farming packages for orphandchild-headed households/youth. 5.30. The PRS Progress Report (2007) assesses the initiatives undertaken to reach these strategic goals, but largely concludes that limited progress has been achieved in creating more effective services and policies. 5.3 1. In line w.ith the PRS, the agricultural policy (KoL 2005) suggests reinforced efforts in the four key areas of the agricultural and livestock sectors that have shown potential for growth and poverty reduction with emphasis on decentralized agricultural service delivery and greater involvement of the private sector. The four areas are: 0 Home gardening (vegetables and fruits), for example, through homestead gardening (keyhole gardens and improved trenches; machobane farming system for horticulture crops with appropriate adaptation to land, water, and cash/labor constraints); improved processing and marketing. A key goal is to increase the labor efficiency in gardening, which is a main constraint, while improving productivity, 0 Rangeland management and livestock production (meat and dairy), for example, through scaling up initiatives to enable common property regimes that regulate access and use of rangelands, including clarification of the role of local government and chiefs in law and practice; establishing user associations to manage pasture and Mountain rangeland; improving veterinary services (as by introducing young private veterinarians and/or village veterinary workers); reducing stock thefts through improved policing and local governance. 0 Wool and mohair production, through reinforcement of breeding, veterinary services, improved woolsheds, and 0 Irrigation and watershed development, for both small-scale watershed development (such as water conservation, check dams, small-scale water development, high-value crops) and more medium-scale irrigation development. These include also water harvesting techniques such as roof tanks, drip irrigation, household dams and the use of grey water. 5.32. Government recognizes the past failures in agricultural and rural policies, which were based primarily on government service delivery and over-emphasis on subsidized maize production (see previous chapter). Consequently, the Government has adopted a policy of bringing in the private sector and civil society to provide many agricultural and rural services. So far the progress has been relatively slow as there are many capacity constraints in developing an enabling environment that fosters efficient private sector involvement (World Bank 2007). Challenges to Rural and Agricultural Pro-Poor Policies 5.33. There are significant constraints to increased productivity and sustainable farm- based livelihood systems. These constraints are both those of a technicallresource scarcity nature 114 and those related to institutional and infrastructural limitations. These were discussed in the previous chapter and include the following. weather: including drought and changing rainfall patterns (especially late rains delaying planting). soiZ fertility: the reduction in use of improved seeds and chemical fertilizers represses - yields and further soil fertility decline and drop in yields.66 Marketing: is another key constraint that limits the number of farmers that can enter commercial high-value markets. Marketing constraints can to some degree be overcome by facilitating commodity-based producer associations and cooperatives, however, price, quality and regular supply would be additional constraints given the competition from South Africa. 0 organization: a more commercialized livestock production would need strengthened farmer/producer organizations, improved market information, and addressing private sector constraints in the sub-sector. 0 negative trends in production: especially wool and mohair, which need to be addressed through a combination of public and private investments. assets: most farmers are constrained at the household level in terms of cash and labor, and therefore often “locked” into ineffective sharecropping arrangements geared mainly toward subsistence farming. 5.34. Under the current circumstances, household interest and motivation for engaging in market production seem limited, and this is reflected in the high and increasing fallow rates. Many farmers-particularly many potential new and young (male) recruits to the agricultural sectors-are poorly motivated, geared more toward finding non-farm employment. Older farmers, including many single female heads of households, face other types of constraints, such as labor shortages, while they retain their key roles in the rural economy. Their roles include both production and reproduction-the former even more so since HJY/AIDS is taking its increasing toll among the younger heads of households and members of the families. 5.35. Few farmers have adequate land and farm assets to engage in commercial agriculture in a major way. Sheep and goat production does not generate much income, although it is clearly a commercial enterprise. Less than 5 percent of cattle are reared for meat or milk products for sale (KoL 2001). Finally, the relevant markets for key products would be local and limited, except for wool and mohair, which can be exported. Most other produce must compete with local imports from SA in quality, packaging and branding, and price. Basotho farmers face constraints in supplying vegetables ‘to urban markets. The price of Basotho milk is much higher than milk from SA. It is estimated that the cost of producing maize in Lesotho was some 30 percent higher than the cost of importing maize from SA (at expected yields of 1.35 mt/ha) (World Bank 2007). Different reports have highlighted the lack of profitability of dryland farming at the household level-except in the Mountain areas in which the use of input is much lower (Leduka and Gay 2007, Sechaba 2000, Turner 2003, PRS 2005). Moreover, farmers from SA are better organized and connected to agricultural services and input markets, as well as to many urban markets in Lesotho. The growth of Lesotho’s local markets depends on local 66The decline in seeds/fertilizer use followed the withdrawal of subsidy by the government on farm inputs - combined with limited success in bringing in the private sector in inpuvoutput marketing. 115 consumer preferences and increase in local demand, again dependent on the general economic outlook. 5.36. Nevertheless, given the large share of people making a livelihood in agriculture, and despite the constraints, efficiency improvements in agriculture are of key importance. Such a strategy should focus on diversification, productivity enhancements, and institutional reforms, while also recognizing that there are different types of farmers across age and gender who are located in different areas with different constraints and opportunities. Diversification, Productivity Improvements, and Institutional Reforms 5.3 7 . The current agricultural policy emphasizes more appropriate and diversified farming and livestockhangeland systems adapted to farmers? needs and eco-zones, more effective provision of inputs, improved marketing and infrastructure, and institutional and land tenure reforms. These approaches could reach a variety of farmers, both small-scale and large-scale, across the diverse agricultural landscape of Lesotho. In this context considerations should be given to the impact and interdependence between urban and rural sector policies. New and more efficient approaches than in the past are needed. 5.38. The policy of replacing food grain with higher value crops such as vegetables and fruits has met with mixed success despite attempts to target the strategy to particular areas and groups. Given the relatively short supply chain to RSA, Lesotho is well positioned to target the high value supermarket driven fruits and vegetable sub sector. But this will require considerably more progress on infrastructure development, including electrification and development of the cold chain, and standards development. Improvements of agricultural services through capacity strengthening, stronger accountability mechanisms, and systems of co-production at decentralized levels are critical to success as well. To address past weaknesses of government services, the private sector should to be encouraged to take more active part in providing services and inputs to the farm sector-both the ?for-profit? and the ?not-for-profit? elements. NGOs will have a greater role, and various types of user groups and cooperatives, such as grazing associations and vegetable marketing cooperatives, will be encouraged through co-production systems (PRS 2005, Food Security Policy and Strategic Guidelines 2005). 5.39. Public investments and programs need to be targeted to the diverse needs of farmers. The large subsistence agricultural sector would benefit from improved resilience of systems through climate risk management and improved land and water management - combined with safety nets. But the small commercial sector would benefit from various efficiency measures related to service delivery, improved organization of farmers, and access to fertilizerdseeds and output markets. The agricultural sector is not likely to drive economic growth, but will remain important for food security and poverty reduction. Further declines in productivity and natural resources management will have an adverse effect on food security for many poor rural households. The recent increase in fertilizer and fuel/transport prices have added constraints to rural households in Lesotho, while the increase in food prices has increased risk and vulnerability for those households that cannot produce for own consumption. The structural shift in prices may create an opportunity to refocus investments in agriculture (e.g. horticulture production) and build better safety nets to help the poor cope with their endemic high levels of risk.67 ?? In this regard, there are possibilities of linking support for small farm development to various forms of safety nets , (see F A 0 2008: Linking Social Protection and Support to Small Farmer Development, Report of a workshop held at FAO, Rome, 17-18 January, 2008) 116 5.40. Among possible new approaches, the Government could experiment with development of a micro insurance market which could facilitate farmer access to credit lines for working capital and protect borrower and lenders in the event of adverse outcomes. This enables investment in new higher value crops, which require investment. Examples of these with regard to rainfall are found in India and livestock in Mongolia. The latter may provide a good example for Lesotho, given the dominance of livestock in the highlands. The Mongolia program utilizes a ‘dual’ insurance to cover livestock mortality at both ‘base’ levels, provided by private insurers, and a disaster risk component for extreme events provided by government. Sharecropping and Land Tenure Reform 5.41. Considering that approximately 70 percent of farmers engage in various types of sharecropping arrangements, efforts should be made to better understand these systems. A study that sheds light on how present sharecropping arrangements work, which groups engage in them and why, and how efficient or inefficient they are from different perspectives could present ways of reforming these arrangements, and guide and advise farmers toward more efficient systems. Such a study could be combined with an empirical analysis of the customary tenure systems and the role of the chiefs in land management and land allocation, particularly since the role of local government in land administration is being redefined in law and practice. Land distribution and access remain priority issues in Lesotho, especially given the high fallow rate, but also in relation to understanding the logic of sharecropping. Tenure security is another issue (but perhaps more related to development in periurban and urban areas). Despite mixed performances in various policing and judicial functions in land and resource management, many chiefs continue to enjoy popular support in most village communities. For instance, their role in livestock and rangeland management concerning adjudication of conflicts over access to winter pasture is still critical. Traditionally, chiefs also have held a role in overseeing sharecropping contracts. It is unlikely that local government will be able to significantly supplant the role of the chiefs in the near future. Furthermore, if the substitution of local government were to be accepted, the key question would be how best to provide a role for chiefs (Leduka 2007, Quinlan and Wallis 2003). 5.42. Such an assessment also would give a better understanding of land distribution issues, including how to address the constraints to land access faced by young women and men who would like to invest in agricultural production. Moreover, the assessment could help improve understanding of the dynamics behind the increasing fallow rates in different zones of the country. Improving Infrastructure and Rural Services to Support Structural Transformation 5.43. A key issue is to ensure more effective government leadership and coordination of actors at different levels. Hence, the focus should be governance at the central and local levels. The pover reduction strategies also will need to address social exclusion issues; and, reflecting the gradual erosion of traditional safety net systems, a strategy is needed for a more coherent yet sustainable national social protection system-focusing on risk and vulnerability at the local level. 5.44. Any long-term rural development strategy must continue to supply and upgrade infrastructure and rural services through decentralized approaches, including provision of rural access roads, electrification, rural water access, education, vocational training, and other social services. Appropriate HIV/AIDS strategies will need to be carefully integrated across sectors. On the demand side, an important element of improving young people’s capacity 117 for social mobility and moving out of poverty will be to improve access to secondary and tertiary education, and prepare young people for nonagricultural wage employment-possibly in urban areas, For many, a future in rural agricultural production may represent a future in poverty for themselves and their families. 5.45, Hence, the development of alternative livelihoods to engaging in the agricultural sector is critical--both to create rural employment and stimulate the rural economy, and to encourage migration out of rural areas to potentially more profitable enterprises in urban and periurban areas. First, rural policies should officially recognize that many individuals are part- time farmers only and encourage combinations of farm- and nonfarm-income activities. A second and related point is that new opportunities in formal and informal off-farm and nonfarm activities need to be enabled through a combination of rural strategies with both medium- and long-term perspectives improvement in services and infrastructure). The nonfarm economy is usually dominated by resource-based activities (rural services, small-scale trading, agroprocessing). Such efforts may be critical to support young men and women who want to remain in rural communities and to stimulate economic growth and sustained income from the land. 5.46. One possible mechanism to stimulate individual economic activities would be to use NGOs and other small financial intermediaries provide credit for small-scale enterprises and self- employment among young men and women who possess some entrepreneurial skills or capacity. Such credit schemes should be made available for both farm and nonfarm enterprise development. However, past experiences with credit schemes have been mixed, so innovative ways of providing credit, outside of the public sector, for example, through user groups seems more appropriate. In the past, large infrastructure projects in rural areas, such as building dams or roaddbridges, have provided important employment opportunities. Efforts to continue this source of income, including through public works schemes, is relevant both from a safety net and a rural development perspective. 5.47. Another option would be to investigate and promote new forms of rural-urban migration as an integrated element of job creation aimed specifically at young men and women. Actions suggested are to: 1. Facilitate rather than hinder seasonal farm labor or construction workers going to SA as a way of employing low skilled labor; 2. Encourage more permanent rural-urban migration by those with at least primary education for wage and non-wage work. This could include a targeted public works program to provide “safety net” employment. 3. Support more highly educated labor to move to both urban areas and SA, possibly through a demand-driven skills program. 5.48. Recent studies show that rapidly urbanizing countries grow faster. If urban areas are where the new jobs are being created, households need to be supported to move there. While it is true that external migration is a “brain drain”, it also provides remittance income. In many societies, younger members support for older generations through remittance income is an important part of the informal safety net. 5.49. Underemployment of rural youth-particularly males-is a critical issue. The efforts mentioned above need to consider broad-based rural youth policies that look at education, family, and community issues. Approaches will need to cover public services (mother and childhood support, youth and adolescent education, sports, and healthy lifestyle encouragement), 118 community support systems (community support groups, family counseling), and individual approaches. Social Protection 5.50. Although HBS data do not suggest an alarming increase in rural poverty over the last decade, there are many factors that indicate erosion of community and family social networks and support system. Several categories of vulnerable groups - such as orphans, youth at risk, widows and widowers - are increasing in numbers, and a minority of households is locked in “chronic poverty”. The government - through for example the decision to provide pension to all people above 70 years - as well as donors through emergency and other programs - have gradually strengthened the social protection system in order to address the needs of poor and vulnerable rural groups. Support to the elders should imply that they would require less support from own children, and strengthen their contributions to local safety nets (e.g. as “granny parents” for orphans) (Gill-Wason 2004). However, it is uncertain how effective and efficient the overall safety nets systems are. It is therefore recommended that the present system is assessed with the view of developing a more efficient and better coordinated national social protection program. Such a program would involve a shift from the re-active and emergency oriented support systems of today, to a more pro-active risk management approach that both protects and promotes livelihoods through a combination of approaches. A first step in this direction would be to prepare a social protection “status” report in order to assess the interaction of the present programs with the profile of poverty, risk and vulnerability (looking at efficiency, effectiveness, and coverage across social groups and districtshegions). This assessment could also involve an assessment of the traditional social networks and the degree to which these are able to support the increasing number of vulnerable groups, and if present level of support is likely to be sustained in the next 5- 10 years. 5.51. Lesotho bears similarities to Ethiopia with a fragile eco-system, repeated droughts and evidence that, at least nutritionally, households are not able to bounce back from the repeated shocks, and might be able to learn from their experience with rural safety nets. Ethiopia launched a Productive Safety Nets program in 2005 to address the chronically food insecure. It operates in around one third of the districts of Ethiopia targeting 7.3 million people. In 2006 it generated more than 172 million person days of labor per year. This type of program requires a well developed public works program, a catalogue of activities than can be implemented as and when employment is required, with the program integrated into district development plans to allow for ongoing maintenance of constructed assets. It can be used for a variety of tasks, many of which would also be suitable for Lesotho. In Ethiopia it has established fruit tree nurseries, built soil and stone bunds to reduce erosion, constructed and maintained roads, school classroom construction, and other activities. Activities can be designed to rehabilitate the fragile ecosystem, increase market access, and establish the foundations of new livelihoods. By having a ready catalogue of activities ready to implement the program can be scaled up and down as necessary to provide labor for those who require it. An initial assessment of the results indicated that 3 out of 5 beneficiaries in Ethiopia avoided selling assets to buy food, more than half used health facilities more, and more than one third enrolled more children in school. Use of this type of productive safety net, particularly in Qacha’s Nek, Mokhotlong, and Tsaba Tseka offers the opportunity to rehabilitate the environment, improve soils and market connectivity and move towards the dissemination of improved farming technologies for higher value crops using smart subsidies and transfers. 5.52. It is important that social protection programs not bring in negative incentives for productive activities. Food programs in rural areas can have this effect, which is why 119 employment schemes are preferred. These schemes do not work as well for women with young children, who are not available for work. A system of conditional cash transfers around enrolling you children and mothers in nutrition programs that enable poor people to be able to purchase food supports private markets while improving other food security related behavior would be another key element of social protection. Conditional cash transfers (CCTs) can also be used for sending some of the most vulnerable children to school (AIDS orphans, herd boys), offer the opportunity to promote nutrition security. These have been successfully implemented in many countries. However, CCTs are most effective if accessible, functional health schools and clinics are available, and if the target population in poverty low (Le. targeting only those below the food poverty line). 5.53. These alternatives to food aid for poor and vulnerable households would have knock-on effects on food prices and markets. By increasing household access to cash, it will enable and maintain an effective level of demand for food, which drives the imports from RSA. This more stable demand will improve the operation of the 2 mills, given the scale of food imports will be more predictable, foster economies of scale in milling, and improve their ability to hedge price risks. 5.54. While this chapter and the last have focused on rural livelihoods and insecurity, urban areas also have poor and vulnerable populations and food security issues. As structural transformation proceeds, the issues in the two areas will have a tendency to converge. In any case, social protection is a national not a regional problem in Lesotho even if the situation today is more compelling in rural areas. 120 Tables Chapter V Table 5.1: Rural household main source of income and consumption quintiles, 2002/03 (YO) Poorest Second Third Fourth Richest All Rural From public sect 2 3 5 6 9 5 From private sec 13 14 20 20 26 19 Farming 60 55 47 43 37 48 Other HH business 2 6 5 5 4 4 Pensions 1 2 2 2 1 2 Remittances 12 10 13 15 15 13 Other 10 11 8 9 8 9 100 100 100 100 100 100 Source: HBS 2002/03 and authors calculations. Table 5.2: Distribution of land by characteristics of head of household, 2002/03 (%) no land less than 1 ha 1 to 2 ha 2 to 5 ha 5 ha and above Total Gender Male 28 17 17 31 7 100 Male -Absent 38 11 16 28 7 100 Female 29 14 18 32 8 100 Age of household head 15-24 53 20 12 14 1 100 25-34 45 13 14 24 3 100 35-44 37 15 18 26 5 100 45-54 29 17 16 30 7 100 55-64 22 15 19 34 9 100 65-74 18 14 19 37 12 100 75+ 26 12 15 36 11 100 Education of head Incomplete primary education 42 73 69 70 73 Primary 33 21 23 23 22 Secondary 25 6 8 7 6 100 100 100 100 100 Source: HBS 2002/03 and authors calculations. Table 5.3: Available labor in land-owning rural households, 2002-03 (%) Number of adults available for work in agriculture Gender 0 1 2 3 or more Total Males 33 40 17 10 100 Male and female 9 25 29 37 100 Sources: HBS 2002-03 and authors’ calculations. An adult available for work in agriculture is an adult who is present in the household (not living abroad) and who does not have wage employment, is not an employer, or is not disabled and is between 15 and 65. ,121 Table 5.4: Characteristics of households by amount of available labor for working the land, 2002-03 ~~ ~ ~~ Number of adults (male or female) available for work in agriculture Characteristic 0 1 2 3 or more Rural average Average age of head 70 52 50 54 54 Percentage male heads 46 61 72 73 67 Percentage with migrants 23 38 38 51 42 (external or internal) Sources: HBS 2002-03 and authors’ calculations. An adult available for work in agriculture is an adult who is present in the household (not living abroad) and who does not have wage employment, is not an employer, or is not disabled and between 15 and 65. Table 5.5: Households with farm implements, 2002-03 (%) Ploughing implements Wheelbarrow Cattle Scotch cart 50 33 53 14 - Sources: HBS 2002-03 and authors’ calculations Table 5.6: Ownership of livestock across consumption quintiles, 2002103 (%) Pigs Goat Sheep Poultry Cattle Poorest 18 22 15 48 48 2 21 22 16 55 44 3 22 25 23 60 51 4 23 21 21 55 48 Richest 20 25 24 59 49 All rural 21 23 20 56 48 Sources: HBS 2002-03 and authors’ calculations. Table shows YOhouseholds that own each type of livestock. Table 5.7: Households with livestock, 1994/95 and 2002/03 (%) Pigs Goat Sheep Po u It ry Cattle - 2003 1995 2003 1995 2003 1995 2003 1995 2003 Rural Lowland 24 22 16 26 16 29 48 53 46 Rural Foothill 28 29 21 26 16 33 61 54 46 Rural Mountain 12 54 35 53 30 65 65 68 54 RSV 21 37 29 37 17 53 69 55 39 Rural 21 31 23 32 20 39 56 55 48 Sources: HBS 2002/03 and authors’ calculations. Table 5.8: Increase in widowed and never married 1994/95 to 2002103 National Rural 1994195 2002103 1994195 2002/03 Never married 35 37 34 36 Married 51 45 51 46 DivorcedSeparated 0 5 0 5 Widowed 10 12 10 13 Living togethedinformation missing 0 0 0 0 Source HBS Surveys and authors calculations. The table is for adults aged 15 and above. 122 6. PUBLIC SERVICE DELIVERY, GOVERNMENT EXPENDITURES AND POVERTY .REDUCTION A. EXPENDITURES GOVERNMENT 6.1. Government expenditures are a large part of the Lesotho economy (45 percent-50 percent of GDP), and improved public service delivery is a key government policy goal. Four of 8 goals in the Poverty Reduction Strategy (PRS)--develop infrastructure, improve access to health care and social welfare, improve quality and access to education, and improve public service delivery-explicitly relate to improved public service. Many subcomponents of the PRS, such as improved safety, agricultural extension, and combating HIV/AIDS, also are key public service delivery areas. 6.2. Government spending on service delivery areas such as health and education has been above regional averages, but a relatively low share of the overall government budget. The share of the budget allocated to the main ministries responsible for public service delivery- education, health, agriculture, infrastructure and water - has been 35 percent to 40 percent of overall government budget (Table 6.1). This is a smaller share compared with other African countries. For example, the five-year average share of budget committed to these services in Uganda and Mozambique is 47 percent and 50 percent, respectively (World Bank 2006c and 2007b).68 6.3. The PRS activities were not funded, and a low share of poverty-related public expenditures was found. A line-by-line review of the 2004/05 recurrent budget revealed that nearly half of all subprograms, corresponding to approximately 34 percent of the total value of the budget, were directly poverty related. With regard to the 2004/05 development budget, nearly two-thirds of projects, accounting for three-quarters of the total value of development budget allocations, were deemed to be directly poverty related (World Bank 2007). The public expenditure review (PER) also revealed that approximately only half of the activities identified in the PRS were included in the 2004/05 budget. In other words, less than half of the estimated incremental costs of the first year of PRS activities were included in the budget. 6.4. In general, the PRS did not work as intended as the key joint vehicle between the government and stakeholders to coordinate, finance, and implement poverty-reducing efforts. Rather, the PRS was a mix of pre-existing policies and wishes, and it never took a central role in policy programming and development. 6.5. Overall public financial management is weak. Reliable information to monitor budget execution generally is not available. There is a general failure to observe financial rules and regulations, leading to over-expenditures on individual budget items, some unauthorized expenditures, and unreliable financial reporting. The lack of adequate trained accounting staff has been another major contributing factor. Since 2005/06, government has been phasing in a new budget/accounting system (MTEF), which should strengthen the link between budget allocations and activities and programs, including the activities identified in the PRS (World .Bank 2006). 68Thenumbers for Uganda and Mozambique are based on total government expenditures less projectized, off-budget donor funding, whereas Lesotho is based on current expenditures. The 5-year average was 1999/2000-2004/04 for Uganda and 200246 for Mozambique. 123 6.6. Nevertheless, among a range of policy issues, Basotho rated public service delivery highest (Figure 6.1 and Table 6.2). However, even though service delivery in 2005 was rated higher than other areas, only a few of the public services were rated positively by a large share of the population. GoL’s handling of crime and basic health services were rated positively by a little more than 50 percent of the people, while fewer than half rated provision of household water and corruption positively. The policy rated highest was the educational needs-more than 90 percent approved of GoL’s handling of this policy area. Combating HN/AIDS also got a decent approval rating of 69 percent. Chapter 7 evaluates this area in detail. Figure 6.1: Popular approval of GoL handling of policy issues, 2005 (%) /OPooresl .Riches11 Sources: Afrobarometer 2005 and authors’ calculations. The variable used is a dummy (goodhad) transformed from the original question with five options: “very badly, fairly badly, fairly well, very well, don’t know.” The option “don’tknow” has been left out. Note that HIV/AIDS has 12% ”don’t know.” The poverty ranking is quintiles based on the first component of a factor analysis, based on how frequently households have been without food, water, medical, and cooking fuel, and ownership of six assets. B. CARE HEALTH 6.7. In the health sector, the PRS focused on three objectives: a. Improve access to quality health care b. Improve nutritional status of vulnerable groups c. Provide social welfare services for vulnerable groups. 6.8. These objectives were reflected in 5 strategies and 5 outcome indicators. The 5 outcome indicators were: a. Percentage of births attended by skilled health personnel. b. Proportion of under- 1-year-old children immunized against measles. c. Infant mortality rate per 1000 live births. d. Under-5 mortality rate per 1000 live births. e. Maternal mortality rate per 100,000 live births. 124 6.9. None of these targets was met. In fact, 3 indictors worsened over the PRS period; 1 improved margin all^^^; and 1 (under-5 mortality rate per 1000 live births) remained unchanged (GoL 2007). Health Care System 6.10. A network of hospitals, clinics, and health centers provides health services in Lesotho. The services are offered by the Ministry of Health and Social Welfare (MoHSW), in conjunction with various nongovernmental and private agencies, as well as donors. The largest nongovernmental health provider is the Christian Health Association of Lesotho (CHAL). CHAL is a church-based organization that runs approximately 40 percent of Lesotho’s health facilities, including hospitals, and serves approximately 30 percent of the population, mostly in remote areas. These facilities are partly financed by MoHSW. 6.1 1. The health care typology was recently redefined based on levels of care sophistication into a hierarchy from Tertiary Care Hospitals to Health Centers”. In addition, a village network of community health workers provides local services. Finally, the Lesotho Flying Doctor Service provides emergency medical service to the remote mountainous areas and delivers essential supplies, including fuel and vaccines, to areas in distress (MoHSW 2000 and 2008). 6.12. The health care system also is being decentralized. This process has been slow but is finally taking shape. Most District Health Management Teams (DHMTs) have approximately 60 percent of the core staff positions filled; 30 percent work in dedicated offices. All of them prepare annual operational plans to guide their activities (MoHSW 2008). Furthermore, public-private partnerships (PPPs) are being undertaken to improve health care provision. 6.13. The most critical challenge facing the health sector has been, and remains, the lack of human resources. Lesotho continues to face a chronic shortage of human resources for health in terms of both skills and actual numbers. The shortage has been occasioned by limited production capacity and high attrition, especially among medical officers and nurses. An Emergency Human Resources Plan is being implemented and seems to produce improvement. Among other initiatives, nurses from Kenya and Zimbabwe have been brought in. However, by the end of 2007, only 3 CHAL facilities and no GoL facilities were fully staffed to standards. 6.14. The critical need to attract staff also is reflected in the fact that, in a small 2007 survey, more then 70 percent of health workers expressed dissatisfaction with their working conditions . Salary and work load were the main complaints (MoHSW 2008). Health Expenditures 6.15. Per capita, Lesotho spends more on health than do poorer neighboring countries but less than Swaziland and Botswana (Figure 6.2), and more could be spent and achieved if funds were managed more effectively. The nominal health budget has increased steadily since 2002 and was set at M83lmilion (US$102 million) in 2007-08.71 This figure corresponds to 11.4 percent of the national budget and M441 per capita at current prices (US$63 per capita). International benchmarking of government health expenditure per capita in 2004 shows that 69 The proportion of under-1-year-old children immunized against measles increased from 77% in 2000 to 78% in 2006. 70 The complete hierarchy include Tertiary Care Hospitals, Regional Hospitals, District Hospitals, Local Hospitals, Filter Clinics, and Health Centers 71 Based on exchange rate of 8.15 Maloti to US$l. 125 Lesotho had higher health expenditure then poorer proximate countries including Mozambique, Zambia, and Zimbabwe. However, Lesotho spent below other high-HN-prevalence countries such as Swaziland, Botswana, Namibia, and South Africa (Figure 6.2). Moreover, more could be achieved if all the money were actually spent every year. During many years, a significant share of the allocated budget was not spent in full. For instance, average expenditure of allocated recurrent budget in 2005-06 was 77 percent, in 2006-07 88 percent in, but in 2007-08 a low 47 percent.” Figure 6.2: High government health expenditures per capita, 2004 300 250 200 150 100 50 0 w 0 Source: WHO 2007. Note: Government per capita health expenditures in US$. Health Care Utilization 6.16. There is a lack of reliable data on health care utilization. Administrative data indicate low utilization of health facilities whereas household data show that patients do seek health care when sick. Administrative data from MoHSW indicate that the average number of patient encounters per capita in 2007 was 0.5, among the lowest in the world; and that bed occupancy rates also were very low (MoHSW 2007). The low utilization could be due to constraints in Basotho’s access to health facilities. Constraints could’be distance, price, or low demand due to poor quality of service. However, comparing regionally the percentage that went for treatment when children had fever or when adults had genital discharges, ulcers, or sores show that Basotho are at least as, if not more likely, than neighboring countries and Sub-Saharan Africa in general to seek treatment (Figure 6.3). Low utilization could also be due to low need, that is, good general health. Considering Lesotho’s development level and its high H N prevalence, lack of need seems unlikely. Comparing regionally the percentage of children having symptoms of fever, diarrhea, and acute respiratory infections shows that Lesotho’s children are neither better nor worse then other countries in the region (Table 6.3). Thus, it is puzzling how utilization can be low if patients do go when in need. Administrative data is often of quality . As part of a general M&E system, statistics are under improvement. However, the system is not yet fully implemented. Many facilities do not report timely if at all, and the quality is still not reliable (MoHSW 2008). Another challenge is that the data sources (patient encounters per capita, bed occupancy rates, and percentage who seek health care when having ’’Measured until December 2007. 126 symptoms) are not directly comparable. However, the inconsistent data raise the question if health facilities have access capacity. This, in turn, raises the question of whether health facilities' capacities are matching the population's needs. This question has been addressed by MoHSW in the past (MoH, 2003) and is a question beyond the scope of this report. Figure 6.3: Basotho seek treatment when having symptoms 90 0 80 0 70 0 60 0 50 0 40 0 30 0 20 0 10 0 00 .Children with fever Elwornen with genital discharge, ulcer or sore 1 6.17. Access to health care-seen as those who seek health care when having symptoms- depends on wealth but also on location. In 2004 being poor made one less likely to seek treatment when having symptoms (Table 6.4). However, a large variation existed between rural and urban areas. When looking at the percentage of children who were taken for treatment after having had symptoms including fever, diarrhea, and cough within the last two weeks, as well as adults who had had symptoms of STI within the last year, these differences are clear. Fifty-seven percent of men and women in the least-wealth quintile went for treatment for STI symptoms, compared to 77 percent in the wealthiest quintile. In urban areas, 75 percent of men and women went for treatment when having these symptoms, whereas 67 percent of the rural population went. For children with fever, diarrhea, or cough, 62 percent went for treatment in rural areas, compared to 66 percent in urban areas. With respect to seeking treatment for their children, the pattern across wealth quintiles is mixed, with the poor and the rich being the least likely to seek treatment for their children. 6.18. Regional comparison of access equality is favorable to Lesotho--especially considering Lesotho's high consumption inequality. Lesotho ranks as having a high level of consumption inequality in Africa. Despite this, the ratio between the first quintile and the fifth quintile of those who seek treatment is better than the ratios of Sub-Saharan countries on average, and at par if not better than other countries in southern Africa (Figure 6.4). If access to health care, seen as those who seek treatment when sick, were perfectly equal, the ratio between the first quintile and the fifth quintile would be 1. Lesotho has a respectable 0.82 for children with fever and 0.70 for women with genital discharge, ulcer, or sore. 127 Figure 6.4: Regional comparison of ratio of poorest to richest quintile of YOseeking treatment when having symptoms ImChildren with fever OWOmen with genital discharge, ulcer or sore] 6.19. When health care services were accessed, the access varied by location and wealth. When a poor Basotho with a sick child visited a health facility .in 2004, in comparison to richer Basotho, she was more likely to visit health center or a community worker. The richer Basotho were more likely to visit a hospital or a CHAL facility for the same problem. The poor also were three times more likely to seek the advice of a traditional healer then were the better off. Among the 2 poorest quintiles, 14 percent took their children to a traditional healer for treatment. Hospitals and private providers are more commonly used in urban areas, whereas' CHAL relatively provide more services in rural areas. (Table 6.5). Barriers to Health Care 6.20. Cost of treatment has been a large obstacle to seeking treatment. This obstacle was mitigated recently. In 2004 when people were asked why they did not seek health care despite having had symptoms for more than 2 weeks and feeling that the symptoms were severe enough for treatment, 82 percent of women and 68 percent of men cited Approximately 5 percent of men and women cited distance (Table 6.6). Since, in 2004, facilities under MoHSW did not charge user fees for most basic services, the barrier was likely related to the nongovernmental providers-to CHAL, the largest nongovernmental provider, in particular. After 5 years of negotiation, in late 2007 CHAL and MoHSW finally reached a new agreement. It includes a new financing formula, a system for accreditation or certification of facilities, and standardization of user fees between of GoL and CHAL facilities. The new agreement, which began implementation in January 2008, also includes removal of user fees from all GoL and CHAL health centers and hospitals. The abolition of fees includes health services for TB, HIV and AIDS, and others with significant externalities or "public goods" dimensions, and the exemption of user fees for the poor and destitute patients (MoHSW 2008). 73 Unfortunately, the question did allow for the break-down of kind of cost. Therefore, the cost could have been for transportation, user fees for visit, or any other cost encountered by seeking health care. Looking at means of transportation usually used by those who cited cost as the main reason for not accessing the facilities revealed that 73% of them usually walk to the facilities. Since walking is inexpensive, if not free, this%age indicates that, for the majority, user fees were the main barrier to access in 2004. 128 6.21. Distance and transport also were seen as access barriers in 2004. When asked more broadly what people considered major obstacles in accessing health care, distance and transport also were mentioned frequently (Table 6.6). The long distances are embodied by the 25 percent of the poorest and 25 percent of the people who live in the Mountains and have more than three hours to travel to nearest health facility (Table 6.7). Distances to a health facility in the Rural Foothills and Sengu River Valley also are large. Approximately only half of the households in these areas live within 1 hour of the nearest facility; many are more than 4 hours away. In addition, only 30 percent of households have access to vehicles--either private or public-when they need to seek health care. The vast majority (69 percent) walks to access health care, and only 1 percent use horse, donkey, or mule (Table 6.8). 6.22. On a positive note, the knowledge of where to seek health care and getting permission to go do not seem to be problems in accessing care. When women were asked about barriers to access health care, less then 2 percent thought that getting permission to go was a big problem. Only 3 percent of women thought that knowing where to go was a big problem (Table 6.6). Among those who did not seek treatment and had symptoms severe enough for treatment, close to 0 percent did not do so because of lack of permission (Table 6.9). 6.23. Poor service at facilities was another major concern. In the 2005 Afrobarometer survey, people were asked which problems they had experienced with public health care. Provider problems including lack of medicines and supplies, absent and disrespectful doctors, and long waits at facilities were the most frequently mentioned (Figure 6.5).74 The same three problems were found to be major problems in a small client satisfaction survey in 2007. This survey found that the average wait was from 1 to 7 hours, with most hospitals averaging 2-3 hours (MoHSW 2008). Figure 6.5: People experiencing service problems in public health clinics, 2005 (76) 60% 50% 40% 30% 20% 10% 0% Source: Afrobarometer 2005. Notes: % = share of people who experienced problems “a few times” or “often,” as opposed to “never” or “once or twice” among those who answered. 74 It is noteworthy that a visit’s being too expensive was not ranked highly as a problem in comparison to provider issues. 129 Implications for the poor 6.24. The abolition of user fees should improve access to health care for the poor, as many perceive fees to be a barrier to access. However, distance, access to transport, and quality of care remain obstacles to access to good health care. These obstacles are greater for the poor since most of the poor live in remote areas. On the positive side, Basotho generally do seek health care when in need, and access is more equal in Lesotho than in most other Sub-Saharan countries. C. EDUCATION 6.25. The PRS highlights seven educational These objectives are captured by eight indicators and targets for progress. The PRS progress report shows that only 3 of 8 indicators had been reached by 2006: primary and secondary student-teacher ratios and primary completion rate. Three targets had not been reached: primary, secondary, and technical and vocational education and training (TVET) enrolment. There is no available data on the two remaining targets: early childhood care and development (ECCD) coverage and literacy rate (PRS progress report 2007). Education System 6.26. The education system in Lesotho works in five tiers; pre-primary, primary, secondary, post-secondary, and tertiary. Pre-primary schools are operated informally by private individuals, local communities, and nongovernmental organizations (NGOs). Primary and secondary schools are owned and operated predominately by churches; some by the government; and the remaining schools by other providers, including communities, NGOs, and private institutions. The government provides direction and financial support for primary and secondary education, mainly through the payment of teachers’ salaries (MoET 2007). 6.27. The analysis of the education sector builds on relative rich data (compared to other sectors) available from Ministry of Education and in particular the 2004 DHS survey. The DHS survey included a number of questions on education and also has measures of wealth that allow an analysis of education from a poverty perspective. 6.28. The educational system has been and is being transformed in key aspects. From 2000 to 2006 a free primary program was implemented, abolishing user fees in primary. From 2007 . new policies to expand infrastructure and reduce fees in secondary are also being i n t r ~ d u c e d ~ ~ . Education Expenditures 6.29. Spending on education in Lesotho skewed toward tertiary education. If it were more balanced the budget could finance needed expansion and improvements in the primary and secondary systems. In line with GoL’s commitment to education, educational expenditures both nominal and as a percentage of the overall government budget increased from 1999/00-2005/06. Educational expenditures were 13.2 percent of GDP in 2004-05-much higher then most other 75 The seven educational objectives are to (1) Expand and promote early childhood care and development (ECCD), (2) Ensure that all children have access to and complete basic and secondary education, (3) Improve quality in basic and secondary education (4)Develop and expand technical and vocational education and training (TVET) to converge with the economic needs of the country, (5) Strengthen nonformal education programs, (6) Increase access to, and relevance of, tertiary education, (7) Promote culture to unify the nation, develop tourism, and generate income. 76 Note that the 2004 DHS survey was conducted during the implementation of the free primary education policy. As such the full impact of the policy is not captured in the following analysis. 130 countrie~?.~? However, the expenditures were heavily biased toward tertiary education. In 2006- 07 more than 35 percent of total educational expenditure went to tertiary education, while primary and secondary received 40 percent and 23 percent, respectively (MoET 2008). Nearly half of tertiary education expenditure went to the National Manpower Development Secretariat (NMDS). NMDS provides bursaries and student loans to tertiary students, making tertiary education virtually free for most students. In the region, Lesotho?s public expenditures per tertiary pupil as a percentage of GDP per capita is almost 3 times higher then any other country, and 5 times higher then most countries (Figure 6.6). Public expenditures on primary and secondary pupils also are higher than other countries?, but not to the same extreme degree as tertiary.78 Recent calculated cost per student show that the cost of tertiary students in Lesotho is 20 times the cost of a primary student. On average, a tertiary student costs approximately 20,000 maloti (US$2964) a year, whereas a primary student costs approximately 1000 (US$148) (MoET 2008). If resources spent on free education at the tertiary level for one student were redirected to primary and secondary, they could provide for 42 more primary students, or 18 more secondary students. It also should be noted that having virtually free tertiary education is extremely rare. Most countries provide partial funding and loans for students that are later repaid when students begin to reap the benefits of their educations. In Lesotho student bursaries are free, and loans are rarely repaid. Figure 6.6: International comparison of public expenditure per pupil as % of per capita GDP, 2005 1400 400 200 0 IUPrimary MSecondary MTertiary] Source: UNESCO database. Primary 6.30. The Free Primary Education (FPE) policy introduced by GoL in 2000 initially boosted primary enrolment significantly. However, since then, progress in net enrolment has stalled, and the PRS target of 90 percent net enrolment by 2006 was not met. The FPE policy was intended to achieve universal education. The policy started free education with a new entrant cohort (to grade 1 in 2000) and has continued the FPE policy with each new entrant every year since. GoL completed the FPE coverage for the 7-year cycle of the primary schools in 2006. The 77 In 2001 educational expenditures in Lesotho were over 11% of GDP, compared to the world average of 4% (World Bank 2005). 78 Note that 25 percent of Lesotho?s expenditures on primary education go for school feeding. This expenditure may not be included in other countries. 131 first year (1999-2000) of the FPE saw an impressive 75 percent increase of new entrants in the first grade. As a consequence, the net enrolment rate jumped from 66 percent in 1999 to 82 percent in 2000. However, the following years saw a decline in students in grade 1. Some of this was clearly a catch-up effect in which children above grade 1 age saw their chance and enrolled despite being over age (Hua 2007). In general since the introduction of the FPE policy, the primary net enrolment rate has been steady at approximately 84 percent. No significant progress has been recorded since 2000 (Table 6.10). This rate shows that challenges remain in accomplishing 100 percent primary enrolments with children enrolling at the right age and graduating at the nominated time. 6.3 1. The FPE policy has not successfully addressed issues of efficiency. Late entrance, and high student repetition rates and dropout rates are major problems, problems that hit poor students the hardest. The net enrolment rate is consistently low due partly to a high number of students entering primary at an older age then they are supposed to. In 2006, 55 percent of students starting grade 1 were above 6 years old; 23 percent were older then 7; and 4 percent older than 10 years (MoET 2007). Repetition rates also contribute: each year, nearly 20 percent of students repeat classes. Information on repetition is only available for 2004-06 and shows a trend of higher repetition rates each year (MoET 2007). Dropout rates also are high-approximately 6 percent annually (MoET 2007). Overall, not more than half of the students who started in 2000 survived to grade 7 in 2006 (Table 6.11). 6.32. Boys and girls do not fare equally well. Repetition rates, dropout rates and the survival rate all are worse for boys. These rates are reflected in the share of female students over time. For instance, in 2001 female students were 45.5 percent of all students in grade 1. When the same cohort reached grade 6 in 2006, 58.8 of students were female. Had boys and girls been doing equally well, the proportion of female students would have stayed more or less constant through out the years (Haiyan 2007). 6.33. The lack of efficiency has the greatest toll on the poor. For instance, the average dropout rate among students from the lowest quintile is much higher then those from the highest quintile (Table 6.12).” Boys in the lowest quintile had a dropout rate of 5.1 in 2004, compared to 0.8 for the richest quintile. The rate for poor girls was 3.4 compared to O.6.for girls in the richest quintile. Similar patterns are seen in repetition rates: 17.5 percent and 12.2 percent of the poor boys and girls, respectively, were repeaters compared to 10.7 percent and 10.6 percent, respectively, in the richest quintile (table 6.12). 6.34. The poor, boys, and children in the Mountains and in Maseru have particularly low net attending rates. Poor boys, especially, have low net attending rates. Among the poorest quintile, 34 percent of boys and 18 percent of girls are not attending primary school within the right age range (Table 6.9). Many of these boys live in the Mountains. A recent government program designed to reach herd boys might address this problem. However, low enrolment is not only a rural problem. Among boys and girls in Maseru, 18 percent and 20 percent, respectively, are not attending within their right age group. The high number of children not in school or out of age range in Maseru is unlikely to be due to long distances, as can be the reason in more rural settings. A better understanding of why children in Maseru are not enrolled is key to improve Lesotho’s low primary enrolment. 79 Calculated as (average dropout rate over grades for highest wealth fare quintile)/(average dropout rate over grades for lowest wealth quintile). 132 6.35. GoL supported the FPE by significantly increasing the number of schools and teachers and expanded school feeding. The number of primary schools increased by 13 percent between 2000 and 2006, due almost exclusively to GoL expansion. This expansion rate is remarkable since GoL generally owns very few schools in Lesotho. Most primary schools are owned by church-based organization. However, during the FPE period, 83 percent of new schools were built by GoL. As a result, even though GoL’s share of total enrolment nation-wide is only 9 percent, the enrolment under government-run schools grew by 1 1-fold. The number of teachers increased by 21 percent over the same 7 years (Table 6.10). The increase in number of schools and teachers over the years of the policy led to a decrease in average school size and a decrease in average student-teacher ratio from 48 students per teacher in 2001 to 41 students per teacher in 2006 (Table 6.10).School feeding programs also were expanded with the FPE policy to increase parents’ incentives to send their children to school and to improve child concentration and learning. For instance, the number of cooks and caterers hired to provide the school feeding increased from 1230 in 2000 to 5052 by 2007 (Table 6.1 1). 6.36. The school feeding program is costly while little is know about it’s effectiveness. The school feeding program made up 25 percent of the primary educational budget in 2002-03 (World Bank, 2005). The allocated has increased over time as school feeding was phased-in with the free primary education program - one grade per year. The program had many start-up difficulties, but administrative and logistical problems associated with school feeding in Lesotho (late payment of cooks, food delivery issues, short catering contracts), have been resolved, often through a more decentralized process. Nevertheless, it remains an open question whether school feeding is achieving its objectives with regard to child learning or having any impact on the nutritional status of children. An evaluation is recommended to determine whether the program is reaching its goals and whether some arrangements are more efficient than others, particularly considering the program’s significant cost. 6.37. Despite the official ban, 94 percent of all primary schools of charge fees. This practice discriminates against the poor. According to the FPE policy, school fees were to be abolished by all schools. However, according to the Ministry of Education, 94 percent of all primary schools still charge some kind of fee (Hua 2007). Observers believe that many schools at first removed the fees, then when they had problems making ends meet, reinstated them. Secondary 6.38. .Secondary enrolment rates have improved but are still below the target and African standards, and are a privilege of the rich. The net secondary enrolment rate increased from 21 percent in 2001 to 26 percent in 2006 (Table 6.10). At the same time, the transition rate from primary to secondary increased from 61 percent in 2000 to 69 percent in 2006. However, in comparison to other African countries and controlling for GDP per capita levels, Lesotho still falls below average (Figure 6.7). Enrolment also is lower then the PRS target of 28. Furthermore, enrolment in secondary is still only a privilege of the rich. Only 1 of 100 boys in the poorest quintile attends secondary; 5 girls from the same quintile attend. These figures contrast strongly with the 38 and 54 of 100 girls and boys, respectively, who enroll within the right age from the richest quintile (Table 6.9). 6.39. Fees seemed to be a main obstacle to further expansion of secondary enrolment. More than 60 percent of males under 24 years old who are not attending school cite too-high fees as the reason for not being in school in 2004 (Table 6.13). In 2007 new efforts to increase secondary enrollment begun, including allowing all students passing the PLSE exam entering secondary. This has led to overcrowding in many schools and a need for further expansion of 133 capacity. A new policy on fees that will likely include a standardization of the fees is currently on the drawing boar$ but the current funding level is inadequate to fund current infrastructure, much less the new infrastructure needed. Figure 6.7: Net secondary enrollment r a t e across Africa, 2005 90 27 29 31 33 35 37 38 4 1 4 3 45 Log GDP per Capita (PPP) Source: UNESCO education database. Data is for 2005 or nearest available year Implications for the poor 6.40. Benefits incidence analysis reveals extremely regressive educational expenditures due to the high expenditures in tertiary and the very low access of poor students to secondary and tertiary. Forty-two percent of all educational expenditures go to the richest 20 percent whereas the poorest 20 percent only receive 9 percent. To estimate how much the poor and rich benefit from education expenditures, a simple benefits incidence analysis has been undertaken. The share of total enrolment in primary, secondary, and tertiary by each asset quintile was estimated from the DHS 2004 survey. Combining these enrolment shares with the share of expenditures spent on each educational level in 2004-05 yielded an estimate of the share of total recurrent educational expenditure going to each asset quintile.80 6.4 1. The regressive expenditures are driven by very high expenditures, particularly for tertiary, which in 2004-05 received 42 percent of the educational budget. The cost of an average tertiary student was 42 times higher" than the cost of an average primary student. Moreover, although primary students came from all wealth quintiles in 2004, 85 percent of students in tertiary education and 65 of secondary students came from the 2 wealthiest quintiles. The substantial amount of resources flowing to tertiary education covers both cost of running the facilities and a substantial amount for student grants and loans Student loans have very low repayment rates. The cost of running the student loan program has been included in the unit cost of tertiary education (Table 6.14). 6.42. Primary expenditures are slightly pro-poor. In contrast, secondary expenditures went predominately to the upper quintiles. Sixty-five percent of secondary expenditures went to the top 40 percent of the population in Lesotho. Regarding gender, at the primary level, boys and girls This method implicitly assumes that the unit cost of students is the same within each educational level. These calculations are from 2004, whereas newer estimates from 2008 indicate that tertiary education is 20 times higher. 134 are equally enrolled. At the secondary and tertiary levels, more girls enroll than boys. Boys were 43 percent and 44 percent in those levels, respectively. Their lower enrolment rates mean that a smaller share of public education expenditures goes to boys than to girls (Table 6.14). 6.43. The lower enrolment and higher dropout rates among the poor reinforce current inequalities and hinder future inclusive growth. Without access to good quality primary education for alchildren, current patterns of inequality will prevail and there will be no social mobility. The poor will remain trapped in their poverty, because they are disadvantaged from first day at school. In chapters 2 and 3, we showed how education is a prerequisite for good employment, and how education and employment both are absolute key elements in avoiding poverty. For instance in urban areas a household with a head that had not completed primary, but is able to read and write had 14 percent higher consumption than those that could not readwrite. Households headed by heads with primary or secondary education had 36 and 70 percent higher consumption respectively (also in urban areas). Over time (1994195-2002/03) those with little education have become more concentrated in the bottom quintiles. An educated work force also is key for long-term growth. Chapter 2 further showed that pro-poor growth will translate average growth rates into both decreasing inequality and more rapid poverty reduction. Equal access to primary and secondary education is a key element in achieving this inclusive growth. 6.44. The fact that secondary has a significant cost to the student, whereas tertiary is virtually free seems perverse from a poverty perspective. A poor student cannot access secondary as the 'costs are too high and there are not enough scholarships, whereas his fellow students with more wealth can pass secondary and continue into tertiary, which is virtually free because the loans for tuition are not paid back. This is hardly a pro-poor policy that provides opportunity for those in need. 6.45. Education policies could become more pro-poor and inclusive by: (a) introducing user fees and/or requiring student loans pay back at tertiary level; (b) reducing fees and increasing the' number of bursaries for secondary school to ensure that these are not a barrier to progression for the poorest; (c) Improve quantity, in the case of secondary, and quality and efficiency of all schools in poor neighborhoods to ensure equal opportunity and enrollment; and (d) early childhood care for development (ECCD) programs have proven to have a significant impact on learning. D. INFRASTRUCTURE 6.46. The PRS goals for improvement in infrastructure include increasing access to roads and transport, access to improved water, access to improved sanitation, access to telecommunications and electricity. Finally, the PRS highlighted the fast and unplanned growth of peri-urban areas in the Lowlands, and described a list of reforms that would ease problems created by these migratory trends. The PRS also list 5 indicators; kilometers of road, percentage households with improved water, percentage households with improved sanitation, percentage households with access to electricity and number of people with access to telephone services. Some of these targets have problems of time consistent indicators, and only the target on number of people with access to phone service can be said for sure to have been archived within the PRS period. Access to Water 6.47. Private access to water comes from two main providers: the Department of Rural Water Supply in rural areas and the Water and Sewage Authority (WASA) in .urban areas. Both operators fall under the Ministry of Natural Resources. They differ in that the Department of 135 Rural Water Supply is an integrated part of the Ministry, whereas WASA is a parastatal governed by a service agreement between WASA and the Ministry of Finance. 6.48. Despite the abundant availability of water in Lesotho, many people still lack adequate access to clean water. This paradoxical outcome is related partly to Lesotho's geography and partly to lack of adequate investment and maintenance of existing infrastructure. The unreliability of urban potable water supply puts a disproportionate burden on the poor, who are more likely to pay higher rates for water through informal markets. The poor who have a house connection often pay more than 10 times the lowest WASA tariff. This phenomenon also is noted in numerous studies that have revealed that even low-income groups evidence a high willingness to pay for improved water services. However, there also are some destitute households dispersed throughout urban areas, including central urban neighborhoods, that may not be able to afford water, no matter how low the price. In response to this burden on the poor, there are plans to introduce a cross-subsidy tariff mechanism that will reflect water for basic human needs only (30 liters per capita per day) for customers who are unable to afford the cost of even the lowest service. 6.49. Access to improved drinking water likely did not improve during the PRS period. However, inconsistent data complicate exact monitoring. The baseline of 79.8 percent having access to improved drinking water is based on the 2002 Core Welfare Indicator Questionnaire. The best available data source for an updated estimate is the 2004 Demographic Health Survey. The two surveys have a different set of options for access to drinking water that largely overlap but differ in a few key areas. Furthermore, neither of the two surveys allows a definition of access to improved water that follows the standard definition established by WHO and used to track the MDGsg2 6.50. The main problem is the inclusion of water collected from springs and whether these are protected or not. The 2002 data did not include springs as an access option. The 2004 data did include springs as an option but did not specify whether the spring was protected. According to the WHO definition, protected springs are an improved water source; unprotected springs are not, To track access to improved water over time, the key question is whether water accessed through springs in 2004 should be considered an improved water source or not. For 2004, if springs are assumed to be protected and therefore included as an improved water source, the percentage of the population with improved access is estimated at 80.4. If springs are not included, the percentage is estimated at 75 .283(DHS 2004). Neither case gives any indication that substantial improvement was achieved or that the PRS target of 85 percent was reached. 6.51. The poor, particularly those located in the Foothills and the Mountains, had the least access to improved water sources. In 2004, 57 percent of the poorest and 64 percent and 65 percent of the population in the Foothills and Mountains, respectively, had access to improved water. In comparison, 95 percent of urban households and 95 percent of people in the highest wealth quintile had access to improved water (Table 6.17). 6.52. The poor and those living in the Lowlands and Foothills spent the most time collecting water (Table 6.15). In 2004, there was a clear trend of the poorer one was, the more time needed to fetch water. However, the location also was important. People in Maseru and the Sengu River Valley spent the least time collecting water, but even in these areas, some people 82 According to the MDGs, improved drinking water sources include household connection, public standpipe, , borehole, protected dug well, protected spring, and rainwater collection. 83 Note that, using the same source, the PRS progress report cites a level of 76.2%. Using this source and the same calculations, the authors arrived at 75.2%. 136 needed more than 1 hour while approximately 60 percent needed less than 15 minutes. People in the Lowlands have the worst situation with more than 20 percent having to spent more than 1 hour to collect drinking water. Only 36 percent had water close enough to their homes to get drinking in less than 15 minutes (Table 6.15). 6.53. Lack of financing for and maintenance of existing infrastructure limit progress in access to water. The Water Commission has stated that coverage has increased only marginally over the PRS period, due mainly to lack of finance. The commissioners estimate that one-quarter of rural infrastructure is inoperable due to lack of maintenance. Given current available funding, the Water Commission estimates that it can increase coverage by 1 percent annuallya4 by repairing current infrastructure. Resources are focused on repairing current infrastructure, and no new pipes are being constructed (PRS progress report). Access to Piped Water 6.54. Access to piped water on the premises is a service provided almost exclusively in urban areas. Normally in a lower middle income country, rural households get water through a standpipe as connections are too expensive. In 2004, 52 percent of the urban population utilized piped water, while only 2 percent in the rural areas had such a privilege. For the 40 percent of the population living in the Mountains, Foothills, and Sengu River Valley, only 1 percent of the population had piped water on their premises (Table 6.17). 6.55. Broadly speaking, there are two possible explanations why people remain unconnected to piped water. The first is that the service is simply not available in the communities in which they live. This unavailability is essentially a supply-side problem that requires increased investment in infrastructure expansion. The second explanation is that the households fail to take up the service even when it is available in the community. This is essentially a demand-side problem that may be less costly to overcome in investment terms. It is possible to capture this difference by comparing the availability rate and the take-up rate. The availability rate is the percentage of 8 households who live in communities in which the service is available. The take-up rate is the percentage of households who live in communities in which the service is available who actually connect to the service.85 6.56. The very low coverage of piped water on the premises in rural areas is both a matter of general service not being available and the difficulty of spreading the service throughout the village. Fully 78 percent of rural areas do not have 1 household with piped water. Among the 22 percent of the population living in communities in which at least 1 household has piped water, only 9 percent of households in that community access the water even though it is present. In urban areas, the dominant reason for not having piped water is more on the supply side. Once piped water is available in an urban area, 78 percent of households get access to it. Unfortunately, only 66 percent of urban areas have the option of connecting as there is no piped water available to them (Table 6.18). 84Defined as those who have access to 30 liters per capita per day within 150m of their homes. Same calculations were used. The authors arrived at 75.2%. The estimates were made following Foster and Araujo 2004 However, neither HBS 2002-03 nor DHS 2004 has a question on the availability of service in the respondents’ local areas. As an alternative method, availability was based on the enumeration area (EA) of the survey. If at least 1 household within an EA used the service, the authors assumed that the service was available to the entire community in that EA. 137 Electricity 6.57. Electricity in Lesotho is provided by the Lesotho Electricity Company (LEC). In comparison to other countries, Lesotho's progress in electrification over the long term has been slow and access low (World Bank 2006b). However, over the last five years, the privatization of LEC and stronger focus on markets and income-generating connections have expedited expansion in connections. The PRS target for electrification was 17 percent of households by 2006, up from 10 percent in 2002. However, the baseline seems to have been calculated incorrectly; therefore, the absolute numbers are less useful (PRS Progress reports6). 6.58. Using the expansion rate implied by the PRS targets, the target for electricity connections seems to have been reached. The PRS target of 17 percent of households being connected by 2006, up from 10 percent in 2002, implied a target expansion rate of 70 percent more connections over 4 years. The number of connections reported by LEC indicated that this expansion target was reached, as the number of connections expanded by 82.5 percent from 2002 to 2006. According to LEC, it had 29,921 consumers connected in 2002 and 54,612 in 2006 (Figure 6.8). 6WOO - 5OWO - - I 40000 il 3w00 2-0- loo00 - o t . . , . . . . . . . . . . . . . . ' , , . . . . . . . . . . . . . , , 86 The calculations in the PRS Progress Report also seem wrong as they assume a population of 2.2 million in both 2002 and 2006-significantly larger than found in the 2006 census. 138 3 - 2 3 3 2 3 3 3 3 3 3 3 2 3 m m c 2 3 3 3 m 4- o Table 6.2: Approval rate of governments handling of political issues, 2005 (%) Poorest Poor Rich Richer Urban Rural Economy 31 38 39 43 39 37 Creating jobs 11 12 14 19 12 15 Keeping prices stable 11 12 15 22 14 15 Narrowing income gaps 15 13 19 22 19 17 Reducing crime 53 51 51 55 55 52 Improving basic health services 51 61 58 64 60 58 Addressing educational needs 94 95 92 90 91 94 Delivery household water 35 45 45 56 52 41 Ensuring enough t o eat 45 48 46 45 45 46 Fighting corruption . 50 46 46 47 46 47 Combating HIVIAIDS 66 66 67 75 69 69 Sources: Afrobarometer survey 2005 and authors’ calculations. Notes: The variable used is a dummy (good-bad), transformed from the original question, which had four options: “very badly, fairly badly, fairly well, very well, don’t know.” The “don’tknow’‘ option has not been included. The HIV/AIDS question has 12% who answered “don’t know.” The poverty ranking is quintiles based on the first component of a factor analysis, based on how frequently households have been without food, water, medicine, and cooking fuel, and ownership of six assets. Table 6.3: Children with symptom in the last two weeks, 2004 (%) Acute respiratory Diarrhea Fever infection Kenya 16.4 40.7 18.4 Lesotho 14.5 25.6 18.8 Madagascar 27.1 32.2 23.9 Malawi 17.6 41.6 26.7 Mozambique 14.5 26.7 9.8 Namibia 12 19.4 17.7 South Africa 13.2 na 19.3 Sub-Saharan Africa Average 19.7 34.4 16.2 Tanzania 12.6 24.5 8.1 Uganda 19.6, 43.9 22.5 Zambia 21.2 43.1 14.6 Zimbabwe 13.9 25.8 15.8 Sources: Lesotho DHS 2004, authors’ calculations, and “Socio-Economic Difference in Health, Nutrition and Population within Developing Countries (World Bank 2007). Notes: Data from all countries are the most recent DHS estimates. 140 Table 6.4: Persons seeking t r e a t m e n t f o r illness, 2004 (%) C h i l d r e n f o r fever, Women-men f o r .$TI** d i a r r h e a , or cough* National 62 69 Residence Rural 62 67 Urban 66 75 Wenlth quintile Lowest 56 57 Second 67 65 Middle 59 67 Fourth 68 75 Highest 58 77 Sources: DHS 2004 and authors’ calculations. Notes: All % are weighted with the sample weights provided by the DHS. * = Children suffering from at least 1 of the symptoms within the last 2 weeks who sought care for,these illnesses as % of all children with symptoms. ** = Women and men suffering from STI within the last 12 months who sought treatment for these diseases as % of all sick women and men. STIs include sexually transmitted infections andlor symptoms such as genital discharge, sores, or ulcers. T a b l e 6.5: Facilities visited for t r e a t m e n t of c h i l d r e n , 2004 (%) National 13 32 5 8 19 9 13 100 Region Urban Maseru 17 25 25 8 25 100 Urban other districts 38 11 1 11 14 5 20 100 Rural Lowlands 12 30 5 10 21 11 11 100 Rural Mountains 4 46 9 3 18 12 9 100 Rural Senqu River Valley 24 33 5 1 21 1 16 100 Wealth quintiles Lowest 5 37 11 4 14 14 15 100 Second 9 39 6 4 17 14 12 100 Middle 19 34 7 6 18 7 8 100 Fourth 17 26 12 26 4 14 100 Highest 18 20 1 21 21 4 15 100 Sources: DHS 2004 and authors’ calculations. Notes: All % are weighted with the sample weights provided by the data providers. Table shows where children suffering from at least 1 of the symptoms (diarrhea, fever, cough) within the last 2 weeks sought care for these illnesses. CHAL facilities cover both CHAL health centers and hospitals. Private covers the options of private doctor, private hospital or health clinic, and other private. Other includes the options pharmacy, family planning clinic, community-based distribution (CBD), other public, health post, and other. 141 I- m I- wm03 I- w m m w d b t C c‘1 C! - E a 0 E .- n eu 0 .- v1 m - V m a N U B 0 bl E 9 E .I 03 Z 00-0 m N N d V m f s .. a L (u m d - V E, m a c 0 m . L m v) M V V m 0 - L . M V m w 8 .. N 2 a ij F m cd 0 0 .e L . E Table 6.7: Distance to nearest health facility, 2004 (%) Less than 15 min to Between Between Between More 15 min 1 hour 1 to 2 hr 2 to 3 hr 3 to 4 hr than 4 hr National 17 49 18 8 4 5 100 Region Urban Maseru 41 56 3 0 0 0 100 Other Urban 35 59 4 0 0 1 100 Rural Lowlands 13 58 20 6 1 1 100 Rural Foothills 10 39 22 13 8 8 100 Rural Mountains 5 28 24 18 9 16 100 Rural Senqu River 8 45 26 10 6 5 100 Valley Wealth Quintiles Lowest 4 30 25 17 10 15 100 Second 6 42 26 13 6 7 100 Middle 11 53 22 9 2 3 100 Fourth 18 63 14 3 1 1 100 Highest 40 54 4 1 0 0 100 Sources: DHS 2004 and authors’ calculations. Table 6.8: Means of transportation usually used to access health care, 2004 (%) Vehicle Ho rse/d on key/rnu le Walking Other National 30 1 69 0 100 Region Urban Maseru 28 0 72 0 100 Urban other districts 26 0 73 0 100 Rural Lowlands 38 0 62 0 100 Rural Foothills 36 1 63 0 100 Rural Mountains 14 3 83 0 100 Rural Senqu River Valley 37 0 63 0 100 Wealth quintiles Lowest 15 2 82 0 100 Second 27 1 71 0 100 Middle 34 1 65 0 100 Fourth 35 0 65 0 100 Highest 37 0 62 0 100 Source: DHS 2004 and authors’ calculations. Notes: Vehicle covers the options car, truck, bus, taxi, or motorcycle. 143 Table 6.10: Primary and secondary attendance ratios, 2004 (%) Primary net attendance rate Secondary net attendance rate Boys Girls Boys Girls Wealth quintiles Poorest 66 82 2 8 Second 78 86 6 16 Third 87 86 15 28 Fourth 87 92 18 33 Richest 88 87 38 55 Urbadrural Urban 81 87 37 47 Rural 81 87 13 25 Region Maseru Urban 80 82 34 48 Other Urban 90 93 35 46 Rural Lowlands 87 88 16 30 Rural Foothills 84 87 12 20 Rural Mountains 77 84 5 16 Rural Senqu River Valley 87 88 12 27 Household head Male 82 86 17 31 Female 79 87.5 14 25 Sources: DHS 2004 and authors’ calculations. Notes: Net primary attendance rate = the proportion of children aged 6 to 12 attending primary. Secondary net attendance rate = children aged 13 to 17 attending secondary. Table 6.1 1: Trends in school characteristics, 1999-2006 (%) Primarv I Secondarv Net No. of Pupill Teachers Students Students No. of Net . Pupil/ enrolment teachers Teachers qualified with desks with seats caterers enrollment Teachers Year rate ratio and cooks rate ratio (%) (%) (%) (%) 1999 60 44 2000 82 8578 48 51.4 25.6 1.230 2001 83 8762 47 54.1 24.4 1,722 21 24 2002 84 8908 47 51.4 25.1 2,215 22 24 2003 85 9294 46 50 25.4 2,640 22 24 2004 83 9993 44 50.8 24.4 3,072 24 25 2006 83 10154 42 48.2 29.2 5,028 25 27 2006 84 10418 41 59 48.7 30.1 5,052 26 26 145 Table 6.12: Survival rates from grade 1 to 5 and 7 for cohorts starting in 1999 and 2000 (%) Cohort 1999 2000 Excl. repeaters Incl. repeaters Excl. reDeaters Incl. repeaters To grade 5 63 58 44 42 Male To grade 7 53 44 32 30 To grade 5 76 70 61 57 Female To grade 7 70 63 51 49 Source: MOET 2007. Table 6.13: Repetition and dropout rates in primary by gender and wealth quintile (%) Dropout rate Repetition rate Boys and Boys Girls Boysand Boys Girls Wealth quintiles girls girls Lowest 5.1 3.4 4.2 17.5 12.2 14.6 Second 4.6 4.6 4.6 18.5 15.1 16.8 Middle 4.8 5.0 4.9 17.2 13.2 15.3 Fourth 2.8 2.6 2.7 15.7 10.4 13.3 Highest 0.8 0.6 0.7 10.7 10.6 10.6 Sources: DHS 2004 and authors’ calculations. Notes: All % are weighted with the sample weights. Dropout rate = % of students, usual residence in Lesotho, in primary school in the previous school year, with an age between 6-24, who are not attending school in the current school year. Repetition rate = % of students, usual residence Lesotho, in primary school in the previous school year, aged between 6-24, who are still in the same grade in the current school year. Table 6.14: Main reason for men below age 24 for not attending school, 2004 (%) National Urban Rural Aged 13-17 Got married 4 5 4 0 Care for younger children 0 0 0 0 Family needed help on farmhusiness 13 4 14 22 Could not pay school fees 61 61 61 60 Needed to work for money 5 8 5 2 Graduated . 1 3 0 0 Did not pass entrance exams 3 5 3 0 Did not like school 8 8 8 13 School not accessible/too far 0 0 0 0 Other 4 5 4 2 Do not know 0 0 1 1 100 100 100 100 Sources: DHS 2004 and authors’ calculations. Note: The statistics are for men aged 24 or below who are not attending school. 146 Table 6.15: Benefit incidence of educational expenditures, 2004-05 (%) Overall share of Share of enrolment education budget Primary Secondary Tertiary Wealth quintile Poorest 22 5 1 9 2 21 10 2 11 3 20 20 12 16 4 19 25 19 20 Richest 18 40 66 42 100 100 100 98* Gender of students Male 50 44 43 45 Female 50 56 57 53 100 100 100 98* Average unit cost 884 2088 37429 % educational budget 37 19 42 Sources: Enrolment by quintiles is from DHS 2004. Unit cost and education budget data are from World Bank 2005. Notes: Budget data are for recurrent expenditures. Tertiary education includes Lesotho Collcge of Education, Lerotholi Polytechnique. Thaba-Tseka Technical Institute, and National University of Lesotho. The budget data for tertiary include the NMDS recurrent expenditures. NMDS provides bursaries and student loans and makes up nearly half of the expenditures in tertiary education. All data are from 2004-05. * = remaining 2% are administrative and other costs that cannot be directly related to enrolment in any program. The male-to-female enrolment ratio is based on 2006 educational bulletins; therefore, it dates from 2006 and not ideally from 2004-05. Additional tertiary enrolment by gender is based on NUL only. Table 6.16: Time to fetch water, 2004 (%) Less than 15 min 15 min-1 hr More than 1 hr Total Wealth Quintiles Lowest 42 41 17 100 Second 46 38 17 100 Middle 43 40 17 100 Fourth 45 39 16 100 Highest 54 33 13 100 Region Urban Maseru 58 36 6 100 Other Urban 59 31 10 100 Rural Lowlands 36 42 21 100 Rural Foothills 47 37 17 100 Rural Mountains 47 39 14 100 Rural Senqu River 64 29 7 100 Sources: DHS 2004 and authors’ calculations. 147 ' 0 ° 0 - 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 0 0 0 0 0 0 0 NO o o m m - o clclmclo 7. THE CHALLENGE OF HIV/AIDS FOR POVERTY REDUCTION A. INTRODUCTION 7.1. The HIV/AIDS epidemic is Lesotho’s greatest development challenge. At 23.2 percent, Lesotho has the third highest HIV prevalence rate in the world, after Swaziland and Botswana (UNAIDS, 2007). This implies that 1 in 4 adults are now living with HIV/Aids. As elsewhere in sub-Sahara Africa (SSA) more women than men are infected, at 26.4 percent to men’s 18.8 percent. The effects of this epidemic are striking. Premature mortality has been on the increase in Lesotho in all age groups. In 2005, a reported 23,000 people (adults and children) died of HIV/Aids. Whether still in the infection-only stage or as an AIDS patient, the impact at the household level has been severe, as the cost of caring for the sick, diminished labor productivity, social and economic cost of death, all result in challenges to food security and poverty reduction. The number of orphans has been rising in proportion to the increase in AIDS related deaths. 7.2. Among the factors driving HIV infection in Lesotho are culture and tradition. Many Basotho have multiple partners concurrently. This practice has been reinforced by men’s long history of labor migration to South Africa where they had greater exposure to possibility of infection. They carry this to their wives when they go back home. At the same time, wives who are left behind are known to themselves have extramarital relationships in the absence of their husbands. Both men and women practice unsafe sex. Condom use does not seem to have made impressive inroads among the Basotho men and women. 7.3. Concerted institutional response by government has been rather slow. The initial response was fragmented, weak, and undertaken mostly by donors and external NGOs. Although the number of new infections is reported to have peaked in about 1996, HIV/Aids prevalence has been roughly constant since 2000 -the number of new infections stable or rising slightly, and the number of deaths steadily climbing. In 2005, the National AIDS Commission (NAC) was created, which developed the Lesotho National HIV/AIDS Strategic Plan to provide a framework for and guide HIV and AIDS activities in the country and to coordinate the many actors in this field. The government has been working with other organizations on prevention action, mainly of mother- to-child transmission (PMTCT); Behavior Change Communication (BCC); promotion of condom use, and facilitation of voluntary counseling and testing (VCT). Support to orphans and other vulnerable children (OVC) was intensified and laws encouraged people living with HIV/AIDS (PLWHA) to come out and mobilize themselves for treatment and other psychological and physical support. Access to treatment has expanded rapidly. While there is no doubt that the more aggressive and coordinated response which the public sector has mounted since 2005 has been a success, key stakeholders still report problems of fragmentation and inadequate funding and outreach outside of urban areas. 7.4. This chapter reviews the evidence on prevalence and who is at risk, including the feminization of the HIV/Aids epidemic; factors driving the spread, on the consequences for households and families, and on the institutional structure for fighting HIV/Aids: prevention, care and treatment. . 149 B. W H O IS INFECTED WITH HIV/AIDS? 7.5. As in many other Sub-Saharan African (SSA) countries, more women are infected than men: 26.4 percent of females in Lesotho are HIV positive compared to 18.8 percent of males. (Table 7.1) Divorced women have the highest infection rate, 49.7 percent, followed by widows at 46.5 percent, married women at 28.8, and single women at 16.6. Urban women have a much higher rate. Prevalence varies by region, from 29.7% in the worst affected district (Leribe) to 17.7 percent in least affected (M~khotlong).~’ Female factory workers (women constitute the majority of factory workers) have a very high infection rate, at 43.2 percenP. The lowest prevalence rate for women is found in the lowest wealth quintile, while for men it is found in the highest one. Having been in more than one marriage’is a significant risk factor for women in rural areas. 7.6. The age profile of the epidemic shows an alarming gender pattern, as the impact on young women is dire. As illustrated in Figure 7.1, the percentage of HIV-positive women is greater compared to infected men in almost all age groups except men 40 years and older. The prevalence of HIV is considerably higher among young women than among young men. Take the 15-24 age group that has a 10 percent infection rate. A breakdown shows young women’s rate at 14 percent to young men’s 6 percent, a pattern consistent with other countries in SSA. The infection rate is less than 10 percent between 18-19 years. By age 22,30 percent of young women will have been infected, to reach almost 40 percent by age 24 years. Gender relations provide fertile ground for what recent studies have suggested to explain the gender infection imbalance: unprotected sex between teenage girls and adult men, mainly unprotected because women are not able to insist on safe sex8’; older men exercising greater power because they have more maturity and a higher income. Older men involved in sexual relationships with younger women are more likely to be HIV infected. For both men and women, prevalence rates peak between ages 30 and 40, when women’s,rate is at 40 percent by age 30 but men’s peaks nine years later. 7.7. Another factor triggering the feminization of HIV/Aids at a young age is the fact that women are biologically more likely to become infected if they have unprotected sex with an infected male rather than vice versa. 7.8. Risk factors for infection differ by gender and residence. Multivariate analysis was used to find out which risk factors had the strongest effects. (Tables 7.2-7.4). Not surprisingly, one of the most important risk factors was being a miner or being the partner of a miner, Once other characteristics are controlled for, being in an urban area is a risk factor only for women, Other key differences developed from the regression analysis (Table 7.4) are: For rural males being currently married or divorced and middle income increased the chances of being HIV positive; for rural females, being divorced or in the second marriage increased the chances of being HIV positive, as did being in the second or 31d wealth quintile (relative to the first quintile). Education and religion were not significant. For urban males, being a widow, and living in a household in the second up to the top quintile decreased the probability of having HIV/Aids; education and religion were not significant. But for urban female (the highest prevalence rates in Lesotho) being in any quintile except the poorest increased the chance of being HIV positive, as did being divorced, while completing primary education or above reduced the probability. ’’ 88 GoLAJNAIDS estimates, 2005 (GoLKJNAIDS estimates, 2005). 89 Dupas P., “Relative Risks and the market for sex: Teenagers, Sugar Daddies and HIV in Kenya,” 2006. 150 Figure 7.1: HIV prevalence by age and gender 45 40 Q 35 0 c 30 - Q 25 P 20 e! n 15 2 10 I 5 I / U o / I I , I , I I 15-19 20-24 25-29 30-34 35-39 4044 4549 Age group 1 +Women -u- Men 1 Source: DHS 2004 and authors calculations. 7.9. While the weak effect of education is somewhat surprising, education level is correlated with wealth and urbanization, so once these factors have been controlled for, the education itself effect is not as strong. In the bivariate analysis, prevalence does decline with education for both genders. Corn0 and de Walque, (2007) found that over all, education has a protective effect, as it is by itself associated with more knowledge of HN/Aids causes and with protective behaviors such as increased condom use. c. FACTORS DRIVING THE SPREAD OF HIV/AIDS 7.10. Three main factors seem to drive prevalence rates of HIV/AtDS in Lesotho: information gaps, socio-cultural norms (including tolerance of sexual violence), and multiple partnerships. The National AIDS Commission has identified insufficient knowledge about HN/AIDS as a continuing challenge in HIV/AIDS prevention: “The level of knowledge and understanding of the way the HIV virus is acquired and spread among the general population is low especially among vulnerable groups due to inadequate access to information and services. Access to information and the ability to act on the received information was found to be hampered by a number of factors including illiteracy, poverty, stigma and state of denial of the existence of H N and AIDS.yy90 7.11. Within the 15-24 years age group only 18.4 percent men and 25.8 percent women had comprehensive knowledge about AIDS in 2004. In one rural district with the lowest infection rate only 11 percent women and 0 percent men had comprehensive knowledge. In Maseru (which has the highest rate, and also the more educated population overall) only 33 90 National HIV and AIDS Strategic Plan (2006-201 I), National AIDS Commission, Lesotho, Nov. 2006 151 percent women and 26 percent men report this knowledge. In light of the country’s high infection rate, these figures pose a great challenge to the fight against AIDS. Education plays a key role. Nationally 70.2 percent of those with secondary education had a comprehensive knowledge of the disease in 2004. Information gaps are also causing 25% of Basotho children in 2005 to be born with HIV/Aids, arising child mortality rates unnecessarily. 7.12. UNAIDS (2006) observes that the principal mode of transmission of HIV in Lesotho is heterosexual contact, specifically multiple concurrent partnerships. Internal and external migration, often necessary to gain access to wage and salary employment, also drives the HIV/Aids epidemic in Lesotho. As noted above, the epidemic started with migrant male miners, and is also an issue for migrant female textile workers. Traditionally, migrants (external or internal) do not move with their families, and multiple partnerships result - both among those who stay and those who migrate. In a 2004 survey, two thirds of men and one third of women reported having sex with someone other than their long term partner in the past year. These patterns have the created a dangerous “network effect” in the transmission of HIV through multiple partners. 91 7.13. Lesotho’s National HIV/AIDS Strategic Plan, 2000/2001-2003/2004 identifies, inter alia, culture and tradition as some of the drivers of HIV/AIDS. The Plan states that the traditional norms reinforce male dominance, while they instill a level of subordination in women, denying them rights over their own sexuality, despite their higher levels of education and knowledge. Sexual violence towards women appears to be at high levels, at least in urban areas. In a recent study, “25 percent of women surveyed reported ever being physically forced to have sex; 13 percent reported that forced sex was attempted; 31 percent said that they were touched against their will; and 11 percent reported being forced to touch a man’s genitals.” Often, perpetrators are boyfriends and known community members. Contributing factors to women .~~ having unwanted sex are fear, pressure, monetary incentives, and economic d e p e n d e n ~ eForced sex is rarely accompanied by protective measures. 7.14. Culture and tradition have also led to an inability to confront the threat of HIV/Aids within a marriage or union, (despite widespread multiple partnerships). Less than 60 percent of couples reported ever discussing HIV/AIDs with their spouse. Many of these couples live in urban areas. Surprisingly, men who have had more than one marriage are 16% less likely to discuss HIV/Aids with their partner (Corn0 and de Walque, 2007). 7.15. As a result, both men and women practice unsafe sex. In Lesotho, condom use has been promoted by both government and NGOs. The impact of this policy, though, remains unimpressive. Only 11 % of women reported using a condom with their husbands in 2004. Of those who reported having had sex with somebody else than their spouse or co-habiting partner within one year, in 2005, less than half of them used a condom. Married women who have extramarital relationships are less likely to use a condom than unmarried women. Men, and some women, believe that use of condom is not negotiable, i.e. it is up to the man. 7.16. One positive finding from the 2004 data is that there were a high number of partnerships where only one partner was infected (discordant partnerships). There were twice as many partnerships where the male was infected but not the female rather than vice versa, even 91The ratio of women to men here is not entirely plausible, and may reflect the quite common reluctance of women to reveal these relationships, even in a confidential survey. 92 Sexual Violence in Lesotho, Lisanne Brown, Tonya Thurman, Jeanette Bloem, and Carl Kendall, Studies in Family Planning, Volume 37 Number 4 December 2006, pp. 269-280. 152 though women are biologically more likely to get infected. If protective measures were adopted by these couples in 2004, the spread could have been slowed. 7.17. A final factor driving the spread of HIV/Aids might be the practice of a special form of “male circumcision” in Lesotho. What is called male circumcision in Lesotho is not the same practice as in Moslem countries or Western countries. It is a rite of manhood, and involves only a symbolic cut or nick in the foreskin, not removal. It is more common in rural areas; in 2004 60% of 20 year old males reported that they had experienced this event. Corn0 and de Walque (2006) found that controlling for all other factors, young circumcised men were more likely to have had sexual relations, and more likely to be HIV positive. It is possible that Lesotho males mistakenly believe that having had this event, they are less likely to get HIV/Aids, (since this has been found with the full circumcision) and therefore do not wear a condom. Unfortunately, they may be more vulnerable after this event. Again, this is a case where cultural factors and poor information encourage unsafe sex and the spread of HIVJAids. I There exists compelling Box 7.1: Male circumcision as a part of the HIV prevention strategy evidence that male circumcision reduces the risk of transfer of HIV from women to circumcised men when engaging in penile-vaginal sex. Randomized experiments indicate that the risk is up to 60 percent lower for circumcised men in comparison to non-circumcised men. There is currently no evidence on the impact of circumcision on male to female transmission of HIV and its impact on non-penile vaginal sex. It should also be noted that this only applies to male circumcision. There is no indication whatsoever that female circumcision has any impact (WHO 2007b). For this reason, WHO and UNAIDS strongly recommend that male circumcision should be a part of a HIV prevention strategy (WHO 2007b), particularly in countries such as Lesotho where heterosexual transmission is the main form of transmission. So far, this prevention strategy has not been adopted because of the possible confusion with ritual “circumcision” performed traditionally on men as a rite of passage to manhood. The traditional practice involves partial removal of the foreskin in a non-clinical environment. Not only has this type of circumcision not been proven to have a positive impact on the transmission of HIV from women to men, but also analysis using the 2004 data suggest a possible negative impact. Should Lesotho include full circumcision (Le. done medically) in its prevention strategy, there would need to be a substantial information campaign to avoid confusion. Of course, male circumcision would not be an alternative but an addition to the current prevention strategy. Male circumcision (done correctly) reduces the risk of infection - it does not prevent it. D. CONSEQUENCES 7.18. The most obvious consequence of HIV/Aids is a massive increase in the probability of death (Figure 7.2). Adult mortality rose from less than 2 percent per 5-year period in the 1970s to above 8 percent for males and above 12 percent for females for the period 2000-2004. While the data does not imply that all the excess deaths are due to AIDS, this increase in adult mortality corresponds to the rise of the HIVIAIDS epidemic in the country. It also appears that adult mortality increased first among men, as they were the first to be infected. However, by 2004 the female mortality rate became, and remains, substantially higher than among men. In general, the older the age group, the higher the mortality rate, although for the 2000-2004 period mortality is higher among the 35-44 age group than among the 45-54 group (Figure 7.3). This is logical given that without treatment, there is an 8-10-year median interval between HIV infection and death. 153 Figure 7.2: Adult mortality by gender, 1975-2004 0 14 0 12 P g 010 a 0 n % 008 5 0 s 006 - - n r n C L P 004 0 02 0 00 1975-79 1980-84 1985-89 1990-94 1995-99 2000-04 Period I+ 15-54yr old males +15-54yr old females I Source. DHS 2004 and authors calculations. Figure 7.3: Adult mortality by age, 1975-2004 0.18 0.16 B 0.14 / U C F012 f 5 010 m 5 008 0 p - f 0.08 2 0 04 0 02 0004 1975-79 1980-84 1985-89 1990-94 1995-99 2000-04 Period 1-C-15-24 yr olds -0-25-34 yr olds +35-44 yr olds 4-45-54 yr olds I Source: DHS 2004 and authors calculations. 154 7.19. When adult mortality is decomposed by gender and by urban o r rural background, it seems that it is higher for males in rural areas (Fig. 1.6), despite that infection there is slightly less than in urban areas. This could be due to a more difficult access to health care, and to the fact that many men living in rural Lesotho are or have been working in the mining sector in South Africa, where they might have been at risk of HIV infection. In terms of education, while mortality increases substantially for all groups, it appears that the largest increase is for the less educated group, especially after 2000 (Fig. 7.4). Figure 7.4: Adult mortality by educational background, 1975-2004 0.14 , B 0.10 'h d s j : 0.08 0.05 p 5 0.04 0 02 0.w J 1975-79 1980-84 1985-89 1990.94 1995-99 2000-04 Period I t n o n e +-primary +secondary1 Source: DHS 2004 and authors calculations. 7.20. It is important to stress that the estimates in Figures 1.6 and 7.1-7.4, which go up to 2004, predate the introduction of antiretroviral treatment in Lesotho. It is estimated that, by 2005, only 1000 individuals had access to antiretroviral treatment. Scaling up antiretroviral treatment since 2005 should have substantially reduced premature mortality. 7.21. Other consequences are a set of threats to families and communities: an increase in children who live without one or both parents (30.9 percent of all children in 2004), increases illness in the family which takes time away from economic and domestic work, increased female headship, and loss of working age adults to stabilize communities.. In terms of actual numbers, estimates of the number of orphans in Lesotho range from 157,600 (MOET, 2006 based on number of orphans attending schools) to 180,000 (USAID). 93 Orphans may be living with a surviving parent, grandparents or some other relative. 94 46 percent of orphans do not live with either parent. Among these, 27 percent live with grandparents; 11 percent live with other 93ActionPlan for the development of a joint concept note and implementation plan, European Community support to Lesotho HIV and AIDS response: Empowerment of orphans and vulnerable children. 94 Orphans are defined as children under 18 years of age who have lost one parent (single orphan) or both (double orphan) 155 relatives; and 2 percent are not living with relatives (Table 7.5). Data on orphaned children show that 13 percent have lost their mother; 69 percent have lost their father; and 18 percent have lost both biological parents. It might appear surprising that the percentage of paternal orphans is substantially higher than the percentage of maternal orphans when HIV prevalence is currently higher among women. Figure 1 suggests that adult mortality started to increase earlier among men than women, indicating that, initially, HIV prevalence might have been higher among men. In addition, the men died first. 7.22. Although almost all Basotho children, regardless of orphan status, are able to attend primary school, secondary school is enrolment for orphans is lower. At ages 14-17, orphans , have a 3 percent lower chance of being enrolled than non-orphans, controlling for wealth, etc. Secondary enrolment is much higher for wealthier quintiles, and the wealthiest quintile has the lowest percentage of orphans, so the real effect comes from the combination of wealth and . ~ ~ decline is worse among rural boys, probably because they are needed to herd ~ r p h a n h o o d The animals. Losing both parents (being a double orphan) has a major negative impact on years of education; losing a mother seems to have greater impacts on girls than losing a father. (Table 7.8). On a positive note, it seems that orphans who lived with a surviving parent or close relative (e.g. grandparents) were able to complete more years of school in 2004 than otherwise. (Tables 7.9 and 7.10). 7.23. HIV is both cause and effect of rising TB in Lesotho, which has the fifth highest incidence in the world. More than 80% of TB patients in the country are HIV positive. Clearly, HIV has created conditions for the come back of TB, while TB weakens those infected in the fight against the quick transformation from HIV to AIDS. 7.24. There has been little analysis of the economic impact of HIV/Aids on the economy. In other countries, the loss of trained labor force is estimated to have a significant effect, for 1% of economic growth to over a 10 year period to 25% over a generation. A study by the IMF in 2002 estimated the medium-term growth impact to be 7 percent. A second study by the IMF in 2004 projected a fiscal deficit of 5 percent of GDP by 2010, with even worse prospects in the future. Most of these studies were done before the more widespread availability of ART. Clearly the private sector has been significantly impacted by HIV/AIDS. The apparel industry em loys approximately 45,ObO Lesotho citizens and is the country's largest private sector employer?'HIV prevalence rate in the apparel factories was estimated at approximately 33 percent in 2007. Without an HIV treatment program for workers, it could be that approximately 1500 employees in the apparel industry will die from HIV-related causes each year. (Colvin, et al, 2006). E. RESPONDING TO THE CHALLENGE Institutional Response 7.25. The institutional response to the HIV/AIDS crisis in Lesotho today is multifaceted, coordinated by the National Aids Commission. The GoL had developed a National AIDS Strategic Plan in 2000/2001. But there does not seem to have been much public action on the AIDS front until 2005 when the National AIDS Commission WAC) was established to 95 The regression analysis in Table 7.7 shows that wealth is a much more important factor in determining enrollment of children at secondary school age, and for the top 60 percent, wealth made up for the orphan penalty in terms of, expected years of education for secondary school children in 2004. But for poorer children, they would be expected to be on average, nearly % of a year behind (or half are at least one year behind). 96 Apparel Lesotho Alliance to Fight AIDS (ALMA), Private Sector Response to HIV/AIDS in Lesotho, May 2006, by Dr. Mark Colvin, Genieve Lemmon, and Dr. Julian Naidoo. 156 coordinate overall response to the epidemic. Soon after, the National AIDS Strategic Plan and Policy were prepared. The country’s delayed official coordinated response, especially given the gravity of the crisis, makes Lesotho’s current problems that much more difficult to solve effectively. 7.26. Since 2005, GoL has significantly improved the coordinating mechanisms, adding more structures to enable itself and its partners to exchange information and work together. The first action NAC undertook was to prepare the National H N and AIDS Strategic Plan, 2006-201 1. The plan drew on the existing National HIV and AIDS Policy, on the joint review of the national H N and AIDS response up to that point, and stakeholder consultation to provide a framework for the many until then uncoordinated responses. The joint review stated that the national response had been based mainly on goals of individual implementing agencies rather than on a national strategic direction. 7.27. The plan identifies key strategic objectives in thematic areas with specific targets, some with concrete timelines. It also includes ways of measuring these targets, sources of hnding for accomplishing these targets, and the organization responsible for reaching the target. The four strategic foci are (1) Management and Coordination Mechanisms, (2) Prevention Challenges, (3) Treatment Care and Support, and (4) Impact Mitigation. The discussion of this section of the chapter is organized along these themes: 7.28. Funding for the Strategic Plan comes from a variety of sources. Lesotho’s line ministries have an obligation to set aside 2 percent of their budgets for HIV/AIDS related activities. However, the greater part of funding for HIV/AIDS programs throughout the country is from a series of grants from the Global Fund to Fight AIDS, Tuberculosis and Malaria (GFATMJ. Lesotho has been approved for grants from 3 rounds of Global Funds grants, with a total of $39 million from Rounds 2 and 5. The grants have targeted prevention, care and support of the infected, VCT and scaling up of TB and STI interventions. The third round grant focuses on OVC. In 2007, an additional grant was provided to focus on TB with partners from the civil society as implementers: CHAL, National University of Lesotho (NUL), Traditional Health Practitioners Council (THPC). It also assists the National TB Program. With the GFATM grants, increased accountability structures have been put in place.97With the infusion of large amounts of funds from GFATM, efforts in the different components of the HIV and AIDS program have improved appreciably. Information and Prevention 7.29. The Plan identifies prevention strategies and action, and has plans for increasing access to information on HIV/AIDS generally and on behavior change communication (BCC) specifically. Other prevention strategies include encouraging abstinence, faithfulness and condom use, improving PMTCT, encouraging peer and life skills education for the youth in and out of school, outreach programs to high risk groups such as commercial workers, and scaling up information campaigns on the risks of HN/AIDS that include how to avoid HIV infection, but if ~~ infected how to live p ~ s i t i v e l y . Prevention efforts have also increased activities in voluntary counseling and testing, with an extensive national plan and action on the Know Your Status (KY S) campaign. Much attention has also given to prevention of mother-to-child transmission (PMTCT). The Plan encourages the involvement of community and government leaders. 97 Global Fund HIV and AIDS Bi-Annual Progress Report, Round 2 Phase 11, Jan.-Jun. 2007, Global Fund TB Bi- Annual Progress Report Round 2 Phase 11, Jan-Jun. 2007, Bi-Annual Progress Report for Round 5 Global Fund Support in Lesotho, Reporting Period Nov. 2006-Apr. 2007. 98 National HIV and AIDS Policy, National AIDS Commission, Lesotho, Nov. 2006. 157 Box 7.2: A plethora of assistance needs Government coordination Many organizations - multi-lateral and bilateral donors, all the key parts of the UN system, non- governmental (NGOs), including major foundations, faith based organizations (FBO) and others from civil society and the private sector - have been assisting the government in the fight against and management of AIDS. The full list is presented in Annex 1 ; key players include: - Irish AID is the designated coordinator of bilateral donors, while WHO is for multilateral donors. These two act as liaisons between GoL and the donors, and assist GoL with overall management and financial aspects of the re~ponse.~' - The World Bank has been a technical advisor to GoL. The Bank's HIV/AIDS Capacity Building and Technical Assistance project aims to assist the GoL to adequately use the large grants from the Global Fund (GFATM). - The US government concerns itself with M&E and distribute's funding from the President's Emergency Plan for AIDS Relief (PEPFAR) to a variety of organizations working on HIV/AIDS and monitors their activities. - The EU dedicates its hnds to OVC. - WHO provides training for health workers and supports national partners in implementation. - The United Nations family --the World Food Program (WFP), United Nations Children's Fund (UNICEF), and UNAIDS-- is involved in a variety of projects, as both technical advisors and implementing agents. Recently, the UN agencies have significantly increased their internal coordination by establishing a Joint Committee on HIV/AIDS. - The Christian Health Association of Lesotho (CHAL) provides approximately one-third of the health care in Lesotho through its hospitals and health centers. - The William J. Clinton Foundation is focused on providing TA and training for the treatment aspect of the response.]'' Many others are addressing HIV/AIDS on the national, district, and community levels. While most welcome the coordination being provided by the NAC, some are not guided by the national framework, and do not appreciate the coordination of NAC. At the same time, as of 2007, stakeholders were still finding NAC's lack of capacity to effectively direct funding, especially to the district level, frustrating. Io' 7.30. In its Strategic Plan for 2005-2015, MOET commits itself to acting in accordance with the National AIDS Strategic Plan, to address HIV/AIDS at all educational levels, including awareness campaigns about HIV prevention.'02 This ministry has been providing training for its staff on HIV/AlDS prevention. In 2005, in coordination with UNICEF and UNFPA, it trained 8000 primary school teachers and 1,000 Early Childhood Care and Development teachers. 2000 student teachers from the Lesotho College of Education received training on HIV/AIDS.'03 7.3 1. Several organizations run after-school and extracurricular programs on HN/AIDS for youth. In 2005, with G o L support, UNICEF started operating the Life Skills Road Shows throughout Lesotho to increase awareness of the virus and the epidemic. Included were some performances on HIV/AIDS and life skills. HIV testing and counselors accompany these shows. The Lesotho Youth Federation received a grant from the Sentebale Organization, founded by princes of both Lesotho and Britain, to train 1500 young people in HIV/AIDS prevention through youth camps and sports activities and 100 peer educators to teach other youth about 99 Dr. Mark Colvin, Genieve Lemmon, and Dr. Julian Naidoo, Apparel Lesotho Alliance to Fight AIDS, "Private Sector Response to HIV and AIDS in Lesotho" (May 2006). loo World Health Organization, www.who.int/ lo' Key stakeholder from UN interviewed for this assessment. lo* Kingdom of Lesotho Education Sector Strategic Plan, 2005-201 1,113. IO3 Lesotho 2005 UNGASS Country Report, Status of the National Response to the 2001 Declaration of Commitment on HIV/AIDS, 1/03-12/05, UNAIDS. 158 HNIAIDS.Io4 Peace Corps, with funding from PEPFAR, operates sports camps that integrate HIVIAIDS education throughout the c~untry.’’~ In 2004 GoL received a grant from GFATM to expand HIV/AIDS prevention services to adolescents with a specific focus on girls’ education.lo6 7.32. Faith-based organizations were slow to act on the HIV/Aids front, discouraging an active program of education in schools. Lesotho’s religious and faith-based organizations have powerful influence in Lesotho: they own 90 percent of the schools and 50 percent of the In 2007 religious leaders of the major churches operating in Lesotho publicly signed .ho~pitals.’’~ a commitment to help fight HN/AIDS in a variety of ways, including educating their congregations about HIV/AIDS and speaking out against stigmatization of those infected. “HIV rates in Lesotho are threatening the future o f the country. Knowledge o f HIV status combined with on-going counseling can impact positively on behavior, for those who test positive ’’ (Minister, MoHSW, 2007) 7.33. .The National H N and AIDS Strategic Plan promotes universal voluntary counseling and testing (VCT) and significant involvement of PLWHA and other role models to reduce stigma and discrimination attached to the disease.Io8 The first VCT site in the country was opened in 2000 by CARE Lesotho. In 2004, the site had a waiting list of 2000 people. In 2004, Population Services International (PSI) launched VCT services in partnership with MoHSW. This organization manages five testing sites throughout Lesotho. Each site operates a mobile VCT service that is used to provide testing access to more remote population^.'^^ However, in 2004, Lesotho had limited capacity to process HIV tests. There were only 3 laboratories and 2 ELISA machines to do H IV antibody tests. There was also a shortage of laboratory technicians.”’ This notwithstanding, WHO has noted a significant improvement in the availability of VCT. Voluntary Counseling and Testing facilities have increased from 18 in 2003 to 102 in December 2005.”’ Moreover, there has been an increase in the number of people who have been tested HIV test. 229,092, or 12 percent of Lesotho’s population, had been tested by 2007-a nearly three-fold increase since 2005.”2 7.34. In March 2004, GoL launched the Know Your Status Campaign to encourage Basotho to come forward for HIV testing and counseling as an HIV prevention measure, but also as a way of enabling the infected to receive care and treatment. During the launch, the Right Honorable Prime Minister of Lesotho, Pakalitha Mosisili , got tested for HIV in his home state of Qacha’s Nek, making a strong statement against the stigma attached to HIV testing in Lesotho.’I3 The campaign’s aim was that “all people over 12 years old living in Lesotho would know their H N status by the end of 2007”, a goal which proved too ambitious. The campaign has received 104 Sentebale: http:/’/www.sentebale.org/home/PreviousGrants.html PEFPAR-Funded Partners in Lesotho, US Government, January 2008. The Global Fund Website: http://www.theglobalfund.org/search/docs/2LSOH~246~197~gsc.pdf 107 Statement of commitment by Lesotho’s Church Leaders on AIDS http://data.unaids.org/pub/ExternalDocuinent/2007/2007071 8 lesothodeclaration en.pdf and Lesotho Churches Bolster HIV/AIDS Fight: http://www.afrol.corn/articles/26086 l), National AIDS Commission. National HIV and AIDS Strategic Plan (2006-201 log PSI: http://www.psi.ordwhere we work/lesotho.html ‘lo Sara Bowsky, Family Health International, with Support from USAID Regional HIV/AIDS Program, Matla ea Lesotho ke Sechaba: Hlahlobo ea Kokelo ea Bakuli ma Haeng/”Lesotho’sStrength Is Its People: A Rapid Appraisal of Home- and Community-Based Care,” CARE Lesotho (April 2004). ‘ I 1 World Health Organization: httD://www.who.int/hiv/HlVCP LSO.pdf Lesotho 2008 UNGASS Draft Report. 113 WHO website: http://www.who.int/3by5/newsitem6/en/ 159 technical support from the WHO and financial backing from the GFATM, the United Nations Development Program (UNDP) and the Global Business Coalition on HIV/AIDS. More of both is needed for the campaign and actual testing to meet its goals. 7.35. An average of 7,000 babies become infected with HIV every year."' In addition, a sizable proportion of HIV-infected women (38.7 percent) intend to have a child. Thus, there is World Health Organization reports that considerable potential for mother-to-child transmissio~i."~ there has been "a steady upward trend in the proportion of pregnant women, 20-24 years old, testing HIV The Government has aimed for universal free access to PMTCT services for all seropositive women. Information on PMTCT services would be provided to all pregnant women and their partners. Prevention of mother-to-child transmission services would be decentralized to increase access in more rural areas'l7, while minimum quality standards for PMTCT care would be established. The National HIV and AIDS Strategic Plan goes further and calls for establishing "universal access to PMTCT services in 100 percent of health care service delivery facilities in the country.'y118 7.36. The Family Health Division of MoHSW initiated a PMTCT program in'2003. If a pregnant woman tests positive, she goes through follow up tests, is assisted with medication. She and her partner receive counseling, and with the help of WFP, she receives nutritional food that is distributed at the centers of testing. The US government funds partners that provide technical assistance for PMTCT. It also assists with personnel training at some PMTCT sites. Prevention of mother-to-child services are offered at all the hospitals in the country and at 20 health centers. But many women go to local clinics for their antenatal check ups. These do not have the capacity for PMTCT services, starting with testing. Thus by December 2006, PMTCT coverage stood at 15.6 percent.Ilg Since then, the pace of improvement has been impressive. In 2007, the number of HIV positive pregnant women on ART to reduce the risk of mother to child transmission had risen from 5 percent in 2005 to 26.2 percent. In that year, the government announced a new plan whose goal is universal access to PMTCT services and reduce the risk of mother-to-child transmission by 80 percent by 201 1, through a partnership between the MoHSW, WHO, and UNICEF.'~~ 7.37. Organizations from the civil society have focused on reaching high-risk groups. CARE Lesotho-South Africa, with support from the Bristol-Myers Squibb Foundation, FHI/USAID, and CARE Africa Fund, ran the Sexual Health and Rights Promotion (SHARP!) program until 2006.12' SHARP! operated in 4 border towns between South Africa and Lesotho. The program included peer education training for youth, high-risk women, low-income women, migrant laborers, and truck and taxi drivers. It supported the Maputsoe and Maseru resource centers, where these target groups could access information, peer education support for HIV/AIDS learning, and other community resources.122In coordination with USAID, PSI is 'I4 Government of Lesotho, "Guidelines to prevent Mother to Child Transmission of HIV," June 2004, \v~wv.heaI t h .gov.lsidoc iinisiitsiRevised"6220 PMI'C:'E"20(; I.! I D H.,IN I S p d f Adair 2007, 7. 'I6 WHO HIV Overview: hap://www.who,int/liiv/HIVCP LSO.pdf 'I7 National AIDS Commission, National AIDS Policy, 2006. 'IB National AIDS Commission, National HIV and AIDS Strategic Plan (2006-201 1). 'I9 "Sustaining Treatment for AIDS in Lesotho." Presentation by Dr. Mphu K Ramatlapeng, Minister of Health and Social Welfare, Lesotho, 2007. 120rr Lesotho: New Plan to Reduce HIV Infections in Children" (April 2007): hm://w ww.irinnews.org/Report.aspx?ReportId=7 1633 Lesotho 2008 UNGASS Country Report, Draft. '22 Matla ea Lesotho ke Sechaba: Hlahlobo ea Kokelo ea Bakuli ma Haeng / Lesotho's Strength is its 160 involved in the Corridors of Hope HIV/AIDS Education Project, which targets high-risk groups similar to those of the SHARP! Program: sex workers, migrant laborers, factory workers, youth, and long-distance taxi and truck drivers crossing Lesotho?s main border posts.?3 Population Services International also has launched a call-in radio show and has a public service announcement encouraging women to delay initiating sexual a ~ t i v i t y . ? ~ 7.38. The National HIV and AIDS Strategic Plan includes specific objectives for access to condoms including increasing their use in concurrent multiple sexual relationships to 80 percent by 2009 and increasing access to condoms to 80 percent of the population at risk by 2008.?? Population Services InternationaVLesotho began to sell a low-priced male condom in 2001 that helped to triple the number of shops in Lesotho that sold condoms. The organization also sells female condoms that are generally distributed to female sex workers, who are taught how to use them.?26In an USAID project with implementing assistance from PSI, condoms are also distributed to high-risk populations through the Corridors of Hope initiative at border areas between Lesotho and South A f r i ~ a . ? ~ ? 7.39. But between April 2003 and February 2004, only 94,500 male and 6000 female condoms were distributed. 7 million condoms were in storage for almost a year due to distribution issues. Is the issue one of constrained availability or the unwillingness/hesitation of people to use condoms? Hospitals and health clinics were the only places where government condoms were widely available, although there have been efforts to expand access through village health care workers.?28 Care and Treatment 7.40. The main goal of the strategic focus for Treatment Care and Support is a rollout of antiretroviral therapy (ART) and related social infrastructure, and improving technical capacity. To counter excess mortality, access to antiretroviral treatment (ART) is progressing rapidly. In 2007, of an estimated 56,000 people who were in need of ART, 24,700 had received ART treatment, compared to 8000 in September 2005.?29 The Strategic Plan aims for 80% coverage, including opportunistic infection in particular TB. Since March 2007, ART has been available at People: A Rapid Appraisal of Home and Community Based Care Summary Report, Prepared for CARE Lesotho by Sara Bowsky, Family Health International with Support from USAID Regional HIV/AIDS Program, April 2004. USAID Health Profile: Lesotho HIV/AIDS: http://www.synergyaids.com/Profiles_Web/Profiles~PDFs/LesothoProfileJune2OO5 .pdf and Assessment and Review: HIV and AIDS Behaviour Change Communication Botswana, Namibia, Lesotho, Swaziland March 2003,Southern African Development Community: http://www,aidsportal.org/repos/SADC%2Obehaviour%20communication.pdf ?24 Population Services International website: http , / \ & \ k n p v o r g h hzre_w_\+orks?\-regionA htinl 12?National HIV and AIDS Strategic Plan (2006 -201 1) November 2006. Populations Services International website: http://www.psi.org/where we work/lesotho.html USAID Health Profile: Lesotho HIV/AIDS: http://www.synergyaids.com/Profiles_Web/Profiles~PDFs/LesothoProfileJune2OO5 .pdf and Assessment and Review: HIV and AIDS Behaviour Change Communication Botswana, Namibia, Lesotho, Swaziland March 2003,Southern African Development Community: http://www.aidsportal.org/repos/SADC%2Obehaviour%20communication.pdf US Government Rapid Appraisal for HIV/AIDS Program Expansion Lesotho September, 2004,United State Agency for International Development (USAID) and Centers for Disease Control and Prevention (CDC). Ministry of Health and Social Welfare, Government of Lesotho, ?Sustaining Treatment for AIDS in Lesotho,? Maseru (2007).This figure has not been verified by an independent survey. 161 over 70 sites, and is available at all the 19 district hospitals, 4 filter clinics, and 14 private practitioner clinics. It is being expanded to 34 local health centers, health posts, and community health centers. In 2007, the Government started using the first class WHO drug regimen.?30 7.4 1. A number of organizations are involved in the increased delivery of ART and work in partnership with the government. The GoL has entered into partnership with donors whereby the government pays 60 percent of ART and donors 40 percent. A public-private partnership has also been forged. Pharmaceutical companies such as Bristol-Myers Squibb have partnered with the country in care provision for PLWHA. Through its Secure the Future initiative, Bristol Myer Squibb opened Senkantana Center in 2004 as a joint venture with the GoL. The center makes ARVs available to the general public and was designed to be a learning center for ARV treatment throughout the ~ountry.?~? Foundations such as the Clinton Foundation HIV/AIDS Initiative (CHAI) have stepped in to scale up treatment. This foundation started working with MoHSW in 2004 to develop a 5-year operational plan to scale up ARV treatment at reduced cost. CHAI also provides technical assistance. It maintains foreign volunteer doctors and nurses in Lesotho to train and support health care staff.132Non Governmental Organizations, for example MCdecins Sans Frontibres (MSF) support in the provision of care and treatment in Lesotho. Medecins Sans Frontikres started its support at a district hospital and 14 health care clinics. It provided ART to 1393 people within 1 year. 133 In partnership with the GoL, CHAI, WFP, Lesotho Flying Doctors, and Mission Aviation Fellowship, Partners in Health (PIH) began a rural health initiative in 2006, which focuses on providing HIV and TB care to rural communities accessible only by horse or single-engine propeller plane. PIH now provides ART and TB medication, treatment for PMTCT, training for health care workers, nutritional support for patients and malnourished community members, home visits, and educational community meetings in 4 rural mountain communities. PIH plans to expand to 10 communities by the end of 2008.?34GFATM grants have assisted in care for PLWHA. As a complement to drug treatment, WFP provides food aid to patients on ART through hospitals and health centers. 7.42. Nearly 18,000 children aged 0-14 are HIV p~sitive,?~? yet in 2006 the estimated ARV coverage rate among children was 14 percent.?36By the end of July 2007, 1,743 children were on ARVs.13? The only place equipped to treat HIV-positive children with ARVs is the Baylor/Bristol-Myers Squibb Children?s Clinical Center of Excellence. It opened in 2005 and is operated in coordination with the Ministry of Health and Social Welfare as a public-private partnership. To address Lesotho?s shortage of pediatricians and qualified health care professionals, Baylor School of Medicine, with funding from Bristol-Myers Squibb, supplies volunteer pediatricians to work at the clinic treating HIV-positive children and to train Basotho 130 Sustaining Treatment for AIDS in Lesotho, Ministry of Health and Social Welfare, Presentation by Dr Mphu K Ramatlapeng, Minister of Health and Social Welfare, Lesotho, 2007. 1 3 ? Five Years of Securing the Future, Continuing to Invest in the Fight Against HIV/AIDS, Bristol-Myers Squibb Company, httu://securethefuture.coml. Clinton Foundation: http://www.clintonfoundation.org/search?q=cache:KGWzDqw3uqY :http://www.clintonfoundation.org/cf-pgm-hs-ai- countries 1 1.htm+lesothoandsite=clintonfoundationandoutpu~ml~no~dtdandclien~clintonfoundationandaccess=pand roxystylesheet-clintonfoundation p33 MCdecins Sans Frontikres, ?Help Wanted, Confionting the Healthcare Worker Crisis to Expand Access to HIV/AIDS Treatment. The MSF Experience in Southern Africa,? Johannesburg (May 2007). 1 3 4 Partnersin Health: http://www.pih.org/where/Lesotho/Lesotho-org.html 13? UNAIDS website: http://www.unaids.org/en/CountryResponses/Countries/lesotho.~p 1 3 6 HIV Insite website: http://hivinsite.ucsf.edu/global?page=cr09-lt- 00andpost=19andcid=LT#Access%20to%2OCare%2Oand~02OTreatment 13? Sustaining Treatment for AIDS in Lesotho, Presentation by Dr Mphu K Ramatlapeng, Minister of Health and Social Welfare, Lesotho, 2007. 162 health care providers.138The Government has a new plan to reach all HIV-positive children who need treatment by 2011. It plans to train many more health care workers and create an accreditation process. The latter will enable trained nurses to administer pediatric, medications, relieving some of the burden from the few doctors.'39 7.43. The National HIV and AIDS Strategic Plan aims to place 90 percent of AIDS patients under home care instead of their being cared for at hospitals. Home-based caregivers receive no compensation for their work. To increase their capacity, they would be trained in basic HBCS standards and their application. Efforts would also be made to involve men in home based care. Funding will come. from GoL, Irish AID, UN, and GFATM,I4' while the MoHSW is the governmental coordinating body. Some NGOs assist with the program. CARE- Lesotho supports an HBCS program with funding from the CORE initiative. Red Cross Lesotho, World Vision, Skills-Share Lesotho, and the District AIDS Task Forces for the districts of Maseru, Quthing, and Berea are also involved in providing HBCS. Many less formal groups organized by community members or self organized provide support to the sick and their care givers. Churches also play an important role in providing psychological and some material support for sick member^.'^' 7.44. A useful tool for home-based care givers is the home-based care kit. It contains many basic items such as aspirin, protective gloves, and antiseptic. However, a significant challenge for these workers is that these kits are not always readily available or they are not re-supplied once the contents are used up. GFATM reported that kits supplied early in the Round 2 grant have not been replenished because there was inadequate reporting on the status of the kits. This points to a lack of coordination and data collection about the use of these kits, resulting in supply not meeting the demand.. GFATM actually reported on the high demand for the kits.14* Thus sometimes care givers use their own money to provide basic material needs for their patients. The lack of remuneration for home-based care workers may become a significant issue if community-based health care workers begin to be paid for their work, as is planned, especially as their work may overlap that of home based care givers.'43 The Private Sector and HIV/AIDS 7.45. 'The National Strategic Plan includes two targets for addressing H[V among workers: (i) workers should have access to HIV/AIDS services and be free from discrimination and (ii) employers should, have employment policies that reflect these norms. 144 In 2006 ALAFA produced a critical detailed report on private response to HIV, claiming that the sector has not developed a strategic response to HIV/AIDS despite the disease's presumed negative impact on business, and governmental calls for employers to address HIV and 138 Lesotho's Crisis: 6 Doctors -- 800,000 Kids: Parents, U.S. Doctors Pitch in to Help Overwhelmed Pediatricians, 9/3/07, http://abcnews.go.com/WN/GlobalHealth/Sto~?id=3553436andpage=l 139 LESOTHO: New plan to reduce HIV infections in children, April 16,2007, http://www.irinnews.org/Report.aspx?ReportId=7 1633 I4O National HIV and AIDS Strategic Plan 2006-201 1, National AIDS Commission. 14' Matla ea Lesotho ke Sechaba: Hlahlobo ea Kokelo ea Bakuli ma HaenglLesotho's Strength is its People: A Rapid Appraisal of Home and Community Based Care, CARE Lesotho, By Sara Bowsky, Family Health International With Support from USAID Regional HIV/AIDS Program April 2004. '42 Matla ea Lesotho ke Sechaba: Hlahlobo ea Kokelo ea Bakuli ma Haeng/Lesotho's Strength is its People: A Rapid Appraisal of Home and Community Based Care, CARE Lesotho, By Sara Bowsky, Family Health International With Support from USAID Regional HIV/AIDS Program April 2004 143 Global Fund HIV and AIDS Bi-Annual Progress Report, Round 2, Phase 11, Jan-June 2007, prepared by Global Fund Coordinating Unit and Ministry of Finance and Development Planning, Lesotho '44 National HIV and AIDS Strategic Plan (2006-201l), National AIDS Commission. 163 ’ AIDS at the work ~ 1 a c e . I ~ Although there have been planning sessions and workshops at different levels, they have not been translated into action. Reasons include limited funding for HIV/AIDS within the industry; absence of an obvious choice for an organization that could spearhead a sector-wide intervention; and lack of M&E to judge the impact of existing HIV/AIDS initiatives. As a result, sector level HIV initiatives have been sporadic, unstructured, and uncoordinated; and most initiatives have been at the company level. 7.46. Activities at the sector level are picking up, starting with the Lesotho Textile Exporters Association, who, in partnership with the Ministry of Labor and MoHSW and with assistance from the US Embassy, in a VCT drive, advised all members to adopt an HIV/AIDS policy. It coordinated HIV-related training for member factories. The Association of Lesotho Employers has run programs for its members that include condom distribution, with help from PSI; VCT training in coordination with MoHSW; educational workplace programs led by the International Labor Organization (ILO); and sensitization workshops for Chief Executive Officers (CEO) and managers.’46Thirty six labor inspectors have qualified to promote HIV issues in the ~ o r k p l a c e . ’ ~ ~ F. IMPACT MITIGATION 7.47. The strategic focus for Impact Mitigation has been to improve the situation for OVC and PLWHA by strengthening community level services and support and improving access to medical and psychological care. Other priorities include setting up a national registration system for orphans, developing policies to protect the rights of OVC and PLWHA, and training PLWHA in developing means to survive. Developing and overseeing programs dealing with OVC and PLWHA has been one of the NAC’s coordinating responsibilities, as has been distributing funds to deserving activities. The MoHSW is the main implementing body. GFATM, through its grants, is the source of most of the funds for the activities in this area, but other bi- and multilateral donors are also supporting OVC and PLWHA: Irish AID, EU, World Health WHO, UNICEF, and the World Bank. Orphans and Other Vulnerable Children (OVC) 7.48. The National Strategic Plan contains a number of objectives aimed at improving the situation for OVC. These include legislation to protect the rights of orphans; increasing the amount of support for orphans at the community level; and creating a system for identifying and tracking orphans.I4*In 2005, GoL, through the MoHSW’s Department of Child Welfare, with UNICEF and other stakeholders produced the National Policy on OVC, the Costed National Action Plan for OVC, and the M & E Plan for the National Action Plan. These same partners also established the National OVC Coordination Committee with representation from all key stakeholders. A Children’s Protection and Welfare Bill, which ensures the rights of children and requires the government to provide basic services such as health care and education to OVC, is waiting to be passed in the Senate.I4’ In 2006, national guidelines were prepared that mandate the Department of Social Welfare to assess, register, and monitor all residential care providers for children, including licensing homes, identifying the children, and providing appropriate 14’ ALAFA, Private Sector Response to HIVIAIDS in Lesotho, May 2006, by Dr. Mark Colvin, Genieve Lemmon, and Dr. Julian Naidoo, p 28. 146 ALAFA, Private Sector Response to HIVIAIDS in Lesotho, May 2006, by Dr. Mark Colvin, Genieve Lemmon, and Dr. Julian Naidoo, p 37. 147 Lesotho 2008 UNGASS Country Report, Draft. 14* National HIV and AIDS Strategic Plan (2006-201 l), National AIDS Commission. 149 Lesotho 2008 UNGASS Country Report, Draft. 164 services.15'. The challenge, however, is the institutional capacity to implement and monitor these plans. 7.49. Other governmental organs and partners are running programs for OVC. The WFP also provides monthly food aid to families who care for orphans. These food packages are often distributed at the schools. Primary education is free by law, (but see Chapter VI re primary school fees) and through MOET, GoL provides bursaries that cover school fees, books, examination fees, stationery, boarding, and food to enable OVC to attend secondary school (Forms A, B, and C). GFATM grants have been deployed to fund OVC to attend high school (Forms D and E), with priority given to double orphans. In 2007 MOET provided 20,817 bursaries to OVC in Forms A through E. Non governmental organizations, FBOs, and other organizations also provide some bursaries for students who may not be covered by the MOET bursaries. Some pay for uniforms and shoes. 7.50. The GFATM Round 2 grant funds a number of organizations from the civil society: Skills Share-Lesotho, CHAL, and World Vision, to address OVC needs. These provide basic packages of care and support, pay school fees, and repair dilapidated homes where OVC live. By June 2007, almost 55,000 orphans had benefited from the GFATM grant. Lesotho has just received an additional GFATM grant (Round 7) that will focus on assisting OVC.'" GoL also is developing a national registration system to identify and track orphans, an essential component to efficiently provide services.152The MOET provided bursaries to 21,000 OVC. 7.51. In 2007, MoHSW, MOET, UNICEF, and the EU began a pilot project in 3 communities to carry out many of the objectives stated in the National Action Plan for OVC. In this project to be monitored by the National OVC Coordinating Committee, the EU has committed more than 11.3 million Euros over 4 years to benefit over 60,000 OVC. The project meets the children's material, physical, and psychological needs. At the central level, MoHSW receives technical support and assistance with infrastructure and staff to help implement the program. At the district and community levels, community care-giver groups are trained to help orphans access essential services such as school and health care, sufficient food, and HIV- The prevention-related life ~ k i 1 l s . l ~ ~ groups receive funds and information on how to connect to service providers. Monitoring and Evaluation would be undertaken, in accordance with the National OVC Action Plan.'j4 Despite delays in setting up the requisite structures, the project is moving forward effectively.I5' People Living with HIVIAIDS (F'LWHA) 7.52. Stigma was a serious problem in Lesotho, as a 2004 study showed that almost 50% of adults would not buy products from an HIV-infected vendor. To fight stigma and prejudice, PLWHA have been open and active. People Living with HIV/AIDS have an active network: Lesotho Network of People Living with HIV and Aids (LENEPLWHA). The network has made significant progress in many areas. It created networks of PLWHA for meaningful I5O UNICEF, Lesotho: New guidelines set for care of orphans and vulnerable children, March 2006. Is' Global Fund HIV/AIDS Bi-Annual Progress Report, Round 2 Phase 11, Jan-June 2007, Global Fund Coordination Unit, Ministry of Finance and Development Planning, Lesotho. Is* Lesotho 2008 UNGASS Country Report, Draft. 153 More aid for Lesotho's numerous orphans,April 2008, www.a~ol.com/articles/24954. Action Plan for the Development of a Joint Concept Note and Implementation Plan, European Community Support to Lesotho HIV and AIDS Response: Empowerment of Orphans and Vulnerable Children, Draft, Prepared by PIIT, 4/07. Interview with key stakeholder at European Union. 165 participation in HIV/AIDS policy formulation and development of plans. LENEPLWHA also has improved access to treatment, care, psychosocial and economic support for PLWHA, nati~nally.’~ ~ Round 2 GFATM grant aims to increase awareness of the human rights of PLWHA. During 2007, members of the Child and Gender Protection Unit (CGPU) within the police units from all 10 districts were trained on the basic laws of Lesotho relating to inheritance, property rights, and sexual abuse, all of which affect PLWHA significantly. Written material on the laws has been distributed to the CGPUs and the general community. Radio programs, too, broadcast awareness campaigns.’57 7.53. I n recent years, there have been significant legal gains for PLWHA in the workplace. In 2006 the Labor Code Act was changed to prohibit pre-employment HIV testing or testing during employment. It prohibits discrimination against HIV-positive employees. And calls for respect for confidentiality. However, although GoL is one of the largest employers in the country and this law does not apply to civil servants. Lesotho also has developed national guidelines for the implementation of workplace H N programs.”* G. ASSESSMENT 7.54. The effectiveness of Lesotho’s HIV/AIDS plan hinges on three things: (a) sufficient and effective coordination and harmonization of HN/AIDS activities in the country, with NAC and MoHSW playing lead roles; (b) availability of information and data to guide sector strategies, inform interventions and facilitate monitoring and evaluation; and (c) capacity of both the planning/coordinating bodies and implementers and facilitators. Lesotho faces challenges in all these areas. All programs seem to be hampered by three major things: capacity, funding, and information gaps. Institutional Arrangements and Coordination 7.55. The GoL was rather slow in responding to the challenge of HIV/AIDS, leaving room for many actors to initiate activities in an uncoordinated manner. The national commission provides the framework for HIV and AIDS related activities in the country through the Strategic Plan it developed. The commission’s mandate includes coordinating the activities and administering distributing the requisite funds. Since the NAC’s establishment in 2005, the many actors, including government ministries, have gradually come to appreciate their roles and have, albeit hesitatingly, come to work with this agency. However, in terms of being guided by the national plan, some stakeholders are still finding their way to acknowledging the plan as the framework for their work. Hence, instead of the plan being the starting point, it is used for reference in developing initiatives. As earlier stated, this may be because of the late start when stakeholders had started working in a different way, or it may be because some have their own agendas and mandates to be guided by. Examples: the recently accepted proposal for Round 7 of the GFATM addresses OVC. Many of the activities appear as goals in the National Strategic Plan and it was informed by that plan. However, it had not used the Plan as the initial point for planning the grant pr~posal.’~’ Another example is the development of the European Union- funded pilot project addressing OVC. This was developed in consultation with MoHSW and UNICEF. Again, OVC support is a priority in the National Strategic Plan, but it was not the initial departure point for developing this OVC program.’60 To decide on programming, the Joint 156 Lesotho 2008 UNGASS Country Report, Draft. 15’ Global Fund HIV and AIDS Bi-Annual Progress Report, Round 2, Phase 11, Jan-June 2007. Lesotho 2008 UNGASS Country Report, Draft. Interview with key stakeholder from GFATM. Interview with key stakeholder from EU. 166 UN Committee on HIV consults with GoL about the latter’s priorities and then considers how these fit in with the UN committee’s mandate and funding constraints. Keeping both of these agendas in mind, the UN analyzes how these potential programs fit in with the National Strategic Plan after the program is already planned, to some degree.16’ 7.56. The limited use of the National Strategic Plan also comes from its being a fairly general plan in terms of actual action. Most programs can fit somewhere or other into the Plan’s stated objectives, and therefore initiators can pick and choose. A more specific operational plan could have proved a more useful tool for GoL to use in coordinating programs.I6* 7.57. Progress has been made in the NAC’s coordination responsibilities. But it is a process. Even at this stage, NAC is still not fully staffed at the district level, and it is not yet carrying out all of its prescribed roles, such as collecting and distributing all AIDS-related funds.163 The different stakeholders are themselves still growing into their roles. Therefore, comprehensive planning and coordination with GoL’s taking full ownership of the national response is still lacking. Insufficient coordination affects how NAC works with other organs of government. Take NAC and the MoHSW as an example. The two are the leaders in HIV/AIDS related issues and work, but they are not working together as well as they could be. While communication about existing programs has improved, challenges at the planning stage prevent ideal program harmonization. Often, planning is done between individual donors and ministries instead of in a more formal, inclusive forum. With this approach, donors do not always know which one of them is taking responsibility for which need until after programs are in place, creating the risk of duplicating efforts. This may be encouraged by donor imperatives to fit projects to be supported into their own national mandates. 7.58. Inadequate coordination has adversely affected the effectiveness of service delivery systems. The health and HIV/AIDS care system has been decentralized/deconcentrated. But there is confusion about the respective roles of the national government bodies at the district level, and actual local government staff about who should be implementing which programs. Communication problems exist between district local governments, district representatives of national ministries and the Ministry of Local Government at the national level, which defines the roles and responsibilities of the sub-national government^.'^^ The lack of clarity of roles and I~~ responsibilities at all levels is a major challenge to implementing effective ~ r 0 g r a m s . To date, one UN agency does not feel comfortable funneling its funding through NAC, because of NAC’s lack of capacity to effectively direct the funding to the district 7.59. Coordination and harmonization is yet to take root among the government’s development partners. One goal of the CCM is effective communication with all partners. However, it appears that some of the members of the CCM do not report the contents of the meetings to their constituents as effectively as they should, creating gaps in coordination and ~~~ c o r n m ~ n i c a t i o n .Moreover, NAC and MoHS W do not always get information from some organizations about their activities.16’ Interview with key stakeholder from UNICEF. 162 Interview with key stakeholder from US Government. 163 Interview with key stakeholder from UNICEF. 164 Interview with key stakeholder from MoHSW. Lesotho 2008 UNGASS Country Report, Draft. 166 Key stakeholder from UN interviewed for this assessment. ‘671nterview with key stakeholder from Global Fund. Interview with key stakeholder from MoHSW. 167 7.60. As can be seen from the data in this chapter (much of it from 2004), Lesotho suffers from insufficient data and usable information on critical HIV/AIDS-related issues. Presently, available data is not only inadequate but remains uncoordinated and inarticulate, to some degree. One of NAC’s responsibilities is to be the repository of HTVIAIDS related information. Building strong information bases could support monitoring the effectiveness of the current programs and activities. For example, there is an increasingly coordinated and widespread effort to assist OVC. But without a national orphan database, it is difficult to know whether a significant number of the OVC in need are actually receiving assistance or to track those already being assisted. Monitoring and evaluation of the implementation of the plan, of promoted projects such as home-based care and PLWHA is frustrated by lack of or uncoordinated and inarticulate nature of the available data. 7.61. Capacity remains a major challenge for the national structures and for the national plans to provide the necessary frameworks, insisting that development partners and others be guided by them. Thus: donors continue to operate in a seemingly unilateral manner; NAC budget is underutilized despite the AIDS challenge; programs such as PMTCT and VCT cannot scale up activities nor sustain them at desirable levels because of insufficiency of qualified staff. The Minister of Health sums up the challenge he terms ‘crisis’in these almost palpable words: .the government has established an Emergency Human Resources Task Force, ‘I.. proposed raising pay o f health care workers, and launched a plan to recruit 200 nurses +om within Lesotho and outside the country (particularly Kenya). But the Task Force has no terms of reference and has yet to be convened, the salary raises are stalled, the nurse initiative has so far identified only a handful of nurses in the country, and the external recruitment plan is yet to befinali~ed.”‘~~ Interventions 7.62. Lacking information (even basic monitoring in a integrated data base), it is difficult to assess the impact of the Government’s response. With regard to the planned interventions, some of which are currently being acted upon, prevention is a key area that does not seem to have been taken as seriously given the decisive role in the fight against HIV/AIDS. 7.63. Prevention has received less attention than treatment, and activities have not targeted high risk groups such as young women and married, heterosexual couples. Data on infection rates speak to their gendered nature. The government has acknowledged the role of culture and tradition in the spread of the disease. But possible action has not been forcefully encouraged, and gender is not targeted. Behavior change by both men and women had not sufficiently encouraged condom use in 2004, and beyond adoption of the condom, behavior change needs to confront the social and cultural norms. Prevention messages are general instead of being aimed at addressing concerns of specific groups such as miners and teenagers, or specific issues such as gender and male circumcision. But inadequate focus on preventive tools could be the result of funding that has focused more on treatment and care than on prevention. 7.64. Effective campaigns targeted to women to increase HIV/AIDS information, using community groups, religious groups, NGO participation are needed and should be funded. As for the content of the education campaign, specific issues are especially relevant for women, for example, the provision of technical skills, including learning how to use female condoms, ’69 Help Wanted, Confronting the Healthcare worker crisis to expand access to HIV/AIDS treatment, the MSF experience in Southern Africa, Mtdecins Sans Frontikres, Johannesburg, May 2007 168 communication skills (negotiation, assertiveness for safer sex), identification of triggers for risky sex, interventions to encourage social or group support. Lesotho has a great advantage over other Sub-Saharan African countries that should be exploited-there is no gender education gap and girls attend school as much or even more than boy. So, HN/AIDS education in schools is an important tool not only of early prevention, but also to ensure equal access to HNIAIDS information by both sexes. 7.65. A second point is that of male circumcision (MC). Medical male circumcision is proving to be an effective mechanism for HIV prevention in other countries. Estimates of up to 60 percent effectiveness (including South Africa) are encouraging. Given the effective role in prevention, medical MC should have received more attention in the plan to encourage action that would turn it round to be an effective weapon in the fight against HN/AIDS. But as noted, this needs to include an appropriate education campaign. 7.66. Activities in support of OVC seem to be receiving countrywide support. Support with education is a best practice, while assistance at the home base is most commendable. A functional database would strengthen efforts in assisting and targeting which OVC. Research on the impact of orphans' living situations on their education and nutrition levels should guide these programs. 7.67. Voluntary counseling and testing and PMTCT and ART picked up enthusiastically in recent years, but all have been affected by shortage of staff in the overall health care system. In 2006, Lesotho had only 89 doctors for a population of almost 2 million, and only 4 of them at Scott Hospital, whose service population is 200,000 people, as an example. In May 2007, 54 percent of professional nursing posts at health centers were vacant, resulting in nursing assistants ~' with only 2 years of training performing most of the clinical ~ 0 r k . l Higher salaries and better working conditions for both nurses and doctors are the main reasons for the exodus of trained medical staff.171 Some Basotho who wish to be tested cannot because of this shortage, as well as that of laboratories and equipment. 7.68. On funding, the activities of PLWHA are said to be facing funding shortfalls, and are also hampered by information gaps and lack of sufficient data. One program whose effectiveness is unquestionable but whose sustainability may be challenged is the home-based care. The effectiveness of this program has depended on the increased burden of women's work. Efforts to bring men into this work does not seem to have taken off. Given the high infection rate in Lesotho, how sustainable would this program as currently practiced be? Malawi responded to this challenge by recruiting volunteers, both men and women, who get paid to give time to the care of patients in their homes and relieve women for other activities. Communities manage the program. 7.69. The most notable successes have been in treatment, with the increase in access to ART throughout the country. But the high infection rate has dramatically increased adult mortality in the last few years. Nevertheless, the widespread availability of ART should enable adults to live longer, which also should slow the mushrooming of the orphan population. O Help Wanted, Confronting the Healthcare worker crisis to expand access to HIV/AIDS treatment, the MSF " experience in Southern Africa, MCdecins Sans Frontikres, Johannesburg, May 2007. 17' Lesotho: Lack of healthcare workers a drain on new HIV/AIDS plan, April 27 2006, http:/lwww.irinnews.org/l.eport.aspx?reportid=58858. 169 7.70. In sum, despite progress, Lesotho faces major challenges ahead in controlling its HIV/Aids epidemic, and mitigating the impacts. Numbers remain high; gender dynamics are insufficiently recognized and acted upon; coordination remains inadequate; information and data gaps exist; and capacity challenges at all fronts persist. There is need to think both within and outside the box in efforts to contain the spread of HIV and manage mitigation of AIDS efficiently and effectively. There is room for more programs especially in prevention: education, PMTCT, and condom use, prevention and PLWHA activities at workplace, by both the private sector and government as an employer. Evidence on infection suggests that it would be beneficial to target all women, but especially young and married women, as well as married couples, for prevention efforts. 7.71. A theme in each chapter of this report is the need to collect and manage information in order to monitor progress and evaluate results. Given the particular challenge of HIV/Aids, this is even more important. What is the use of spending money and spare capacity implementing programs if there is little information on effectiveness? Failure to implement regular household surveys, for example, is literally killing people. It also prevents Lesotho from learning from good practice within Lesotho. 170 Annex to Chapter 7: Summary of Implementing Organizations Implementing organization Technical area Budget Apparel Lesotho Alliance to Fight AIDS Labor: Evaluation and advocacy * (ALAFA) Baylor School of Medicine Treatment: Provide pediatricians, hospital care Bristol-Myers Squibb Foundation Treatment: Clinic for adults and children CARE-Lesotho Impact mitigation: Home-based care Christian Health Association of Lesotho Treatment: Hospitals and health clinics (CHAL) Clinton Foundation HIV/AIDS Initiative Treatment: TA and training, drug (CHAI) procurement European Union (EU) Orphans and vulnerable children 1 1.3 million Euros Global Fund to Fight AIDS, Funding for all areas Rounds 2 and 5: $39 Tuberculosis, andMalaria (GFATM) million for HIV/AIDS Does not include funding Irish AID Donor coordination: Bilateral donors Management I Treatment Lesotho Youth Federation Prevention: Youth training MCdicins Sans Frontikres (MSF) Treatment Ministry of Education and Training Prevention: Training and education for (MOET) teachers and students, impact mitigation Scholarships for orphans and vulnerable children Ministry of Health and Social Welfare Main implementing body for all (MoHSW) HIV/AIDS programs National AIDS Commission (NAC) Overall coordination M&E Partners in Health Treatment: Rural health care Peace Corps Prevention: Youth training Population Services International (PSI) Prevention: Condom distribution, VCT, education Red Cross Lesotho Prevention: VCT, impact mitigation HBCS ovc United Nations Children’s Fund TA: Education (Life skills curriculum), (UNICEF) impact mitigation (orphans and vulnerable children) United States Government- PEPFAR Prevention Coordination of supply management Integration of HIV/TB services World Bank Management of finds and coordination: TA World Food Program (WFP) Prevention and treatment: Food relief World Health Organization (WHO) Donor coordination: Multilateral Training for health workers Note: Author unable to obtain missing budget Formation from Lesotho. 171 Tables and Figures for Chapter VI1 Table 7.1: HIV prevalence by selected characteristics, 2004 (YO) Females Males All 26.4% 18.9% No education 30.7% 28.0% Primary education 26.0% 16.6% Secondary education or above 26.9% 17.8% 1 quintile wealth index 19.6% 17.9% 2 quintile wealth index 27.9% 16.7% 3 quintile wealth index 25.5% 23.2% 4 quintile wealth index 27.3% 2 1.7% 5 quintile wealth index 28.9% 14.2% Rural 24.3% 18.2% Urban 33.1% 2 1.4% Sources: 2004 Lesotho Demographic and Health Survey The data are weighted with the sample weights recommended by the data provider. The standard errors in brackets take into account clustering at the enumeration area level. 172 Table 7.2: Determinants of HIV prevalence in Lesotho, 2004 Males Females Urban 0.0219 0.0414 [0.0303] [0.027 81 Currently married 0.0689** 0.0126 [0.0270] [0.0310] Formerly married 0.1329** 0.267 1*** [0.063 13 [0.0628] Widowler -0.0 104 -0.03 64 [0.055 71 [0.0460] >1 marriage 0.0573 0.1 141 E0.05721 [0.0763] Primary education -0.0503** -0.0249 [0.0256] [0.05 8 93 Secondarylmore education -0.0400 -0.0479 [0.0275] [0.0607] 2' quintile wealth index 0.01 13 0.0992*** [0.0282] [0.0344] 3' quintile wealth index 0.0908** 0.0906** [0.0354] [0.0378] 4' quintile wealth index 0.0825** 0.0938** [0.0386] [0.0423] 5" quintile wealth index -0.0 140 0.0674 [0.03 621 [0.0432] Catholic 0.0202 -0.0161 [O.O 1871 [0.0196] No religion -0.0289 0.0851 [0.0301] [0.0932] Age group dummies Yes Yes Region dummies Yes Yes Ecolog. zone dummies Yes Yes Observations 2232 3013 Source: DHS 2004. Notes: The data are weighted with the sample weights recommended by the data provider. The standard errors take into account clustering at the enumeration area level. Dependent variable =1 if HIV positive. Robust standard errors in brackets. * = significant at 10%; ** = significant at 5%; *** = significant at 1%. 173 Males Females Miner =1 if male is a miner 0.0843* 0.09 15* [0.0529] [0.053 71 Partnerminer = I if partner is a miner 0.0588** [0.03001 Urban 0.0248 0.0329 0.0957** [0.03051 [0.0309] [0.0397] Current married 0.0655** [0.0273] Former married 0.1348** E0.06381 Widow -0.014 1 [0.0552] More than one marriage 0.05 18 [0.0560] Primary educ -0.0504 * -0.05 11* -0.06 19 [0.026 13 [0.0265] [0.0678] Secondary plus educ -0.0412 -0.0475* -0.0443 [0.0295] [0.02871 [0.0712] Wealthinded 0.0 100 0.0104 0.0755* [0.0280] [0.0282] [0.04291 Wealthindex3 0.0849** . 0.0784** 0.0306 [0.0357] [0.0352] [0.0460] Wealthindex4 0.0772** 0.0724* 0.03 19 [0.0383] [0.0379] [0.0507] Wealthindex5 -0.0192 -0.0181 0.0378 [0.0352] [0.035 13 [0.0540] Catholic 0.0201 0.0196 -0.0208 [0.0186] [O.O 1841 [0.0240] No religion -0.0280 -0.0327 0.2689** [0.0304] [0.028 81 [0.1147] Age dummy Yes Y €3 Yes Eco dummy Yes Yes Yes Region dummies Yes Yes Yes Observations 2229 223 1 203 5 Notes: The data are weighted with the sample weights recommended by the data provider. The standard errors take into account clustering at the enumeration area level. Dependent variable = I if HIV positive. Robust standard errors in brackets. * = significant at 10%; ** = significant at 5%; *** = significant at 1%. 174 d c c w .- W - 5 (F E a 2 E f 2 C f s 0: e .- E a, - C (F z : 2 z L C V c E .- a E E z c i r a, 2 F Table 7.5: Percentage and characteristics of orphans among children age 0-17,2004 Percentage of orphans among children age 0-17 Orphan 30.95% Orphan status by age Age 0-5 18.20% Age 6-10 31.10% Age 11-15 39.10% Age 16-17 4 1.80% Distribution of Orphans and non-orphans Wealth quintile Non-orphans Orphans First 20 22.8 Second 20.4 21.4 Thud 19.1 21.7 Fourth 20.1 19.7 Fifth 20.4 14.4 100 100 Urbanhural Urban 14.4 15.2 Rural 85.6 84.8 100 100 Double orphan 17.9 Mother orphan only 13.1 Father orphan only 69 100 Orphans living with parents 54 Orphan living with grandparents 27 Orphan living with some relatives 11 Orphan living with no relatives 2 Orphans not living with parents, but living arrangements unknown 5 Source: Lesotho Demographic and Health Survey, 2004. Notes: Orphan=if mother or father orphan or double orphan. All the percentage are weighted with sample weights. 176 F M M '- x x m 0 2 F M M v, x x m 0 0 0 0 0 0 0 0 0 0 I0 a2 bl m h P m 0 0 c E m s: 0 0 0 0 0 0 E 0 G 0 v) a u s .- 5 0 0 0 0 0 0 s bt C c . 0 I m 0 r- s: e : : r- 0 v) L m a2 h r- 0 * E m .- E E a2 . L I 4 .. 2 a2 z I c Y? 0 0 c1 ai= of m h n m e Q c 0 E c. 0 a2 P Y r e 0 s V e .- c, L 3 e 0 0 0 0 0 0 0 0 - ? - i * 6; n . 3 0 1 E E 0) 10 00 0 z* m 00 s; 8 I c! m e c 0 0 E 4 4 u 0 0 m V c. 0 0 0 - - : l 0 v1 Y e m ._ e E Y a2 0 0 0 0 0 0 6 .. 7Q) a c 0 0 0 0 0 0 1 ;I u W m P e m v) E m c e 0 4- 0 U C a2 E Q) l o e e L m W ." E + I .I C 0 v) a V E e '5 U E u - E I e C a2 d 0 0 I E w L 0 v) U e m .- e E 0 c, ; .. 0 2 a2 z s 0 L I II m w h \o E m 5 L 0) M C a 0 h C 2 .- 3 L Lo Q) M m L Lo I .- M Q) f L 0 I II Y E m .- C E * Q) - 6 .. 2 a2 z F 8. LESOTHO’S LEGAL REFORM AND IMPLICATIONS FOR GENDER EQUALITY AND POVERTY REDUCTION 8.1. .Structural change brought improvements in the economic position of Basotho women. While their men were away in South Africa, for decades women managed the farms and undertook key responsibilities at the household and community level. Women were and still are the backbone of the rural economy-responsible for food security and subsistence agriculture production. More recently, as men were losing jobs in the South Africa mines, women started finding new employment opportunities in the cities as wage earners, thanks to their high education level and the rapidly expanding textile industry. With the growth of industry, increasing numbers of young women have moved to urban areas, often reproducing the same migratory patterns of male miners, who typically migrated individually leaving their family behind. Basotho women have high literacy rates (in 2001, 90.3 percent, much higher than men’s 73.7 percent, (UNESCO 2008)) and today occupy positions at the highest levels of g o ~ e m m e n t . ’ ~ ~ 8.2. However, despite major changes, the increase in ‘women’s empowerment’ was slow until 2006. Until very recently, under formal law, women married in community of property (the vast majority in Lesotho) became legal minor^."^ Customary law also denied land still denies women legal capacity tout court. This meant that ownership and decisions on disposal/expansion of enterprise, farm, and household assets remained with their husbands and sons, who, given the traditional migration patterns, often were not resident. Women were unable to enter into legally binding contracts without their husbands’ consent, and had no standing in court. While, as we will discuss in this Chapter, the legal status of Basotho women has dramatically improved over the last few years, still customary laws, entrenched in traditional gender attitudes and norms, deny the achievement of full economic and social equality between men and women. Women’s empowerment is a story of incomplete transition too. 8.3. Recent legislative developments greatly improved the legal status of Basotho women, putting them on equal footing with men in the enjoyment of most personal and economic rights. The Legal Capacity of Married Persons Act of 2006 (LCMP), along with other progressive pieces of legislation laid the groundwork for a gender-responsive legal framework that protects women’s rights. Vested with full legal capacity and liberated from male guardianship, adult Basotho women are now better equipped to play more productive, active and satisfying roles as economic and social agents. Their extremely high literacy rate and experience in resource management put them in a unique position to take advantage of the new legal setting. The LCMP and the other legislative developments provide the legal backbone which Basotho women need in order to act as autonomous socio-economic agents. 8.4. This chapter provides an account of the main gender-related legislative developments recently enacted in Lesotho and their potential impact on women’s statu^."^ The impediments that continue to hamper the status of Basotho women and hinder the positive impact of the ongoing 172 Based on the last 2007 election results, women occupy 25 percent of the seats in the National Asse‘mbly and 29 percent of the seats in the Senate-a relatively high percentage in relation not only to other Sub-Saharan countries, but to OECD countries as well (Inter-parliamentary Union, 2008) 173 Under statutory and common law, women acquire majority status when they reach 21 years. Until LCMP was enacted, women who married in community of property lost this status upon marriage. Table A8.1 in annex 8.1 summarizes the most important legislative changes that have affected women’s rights and the remaining challenges to be addressed. 183 legal reform are then set forth. Finally, options that may help address the shortfalls of Lesotho’s legal and judicial system and enhance its capacity to uphold the rights of Basotho women are discussed. A. DEVELOPMENTS GENDER-RESPONSIVE LEGAL IN LESOTHO’S FRAMEWORK 8.5. The Legal Capacity of Married Persons Act was a crucial advancement in women’s legal position in terms of recognition of their majority status, legal capacity, and social and economic rights. In an evolution that began with the establishment of the Law Reform Commission in 1997, Lesotho‘s legal framework has been amended and integrated by a number of gender-responsive pieces of legislation. The main legislative development affecting Basotho women’s legal status was the Legal Capacity of Married Persons Act of 2006 (LCMP). The LCMP repealed the marital powers enjoyed by a husband over the property and person of his wife, sanctioned the legal capacity of married women (whether or not married in community of property), and recognized their ability to perform certain activities previously barred to them. Women’75now have guardianship over minor children (and their consent is required on issues such as early marriage);’76and are able to choose their own domiciles; enter into contracts; manage, together with their husbands, the assets pertaining to the community of property;’77be parties to lawsuits; act as directors of companies; register immovable property in their own names; and perform “any other act which was restricted by any law due to the marital power before the commencement of [the] 8:6. This Act potentially means that Basotho women enjoy increased decision-making power on issues such as their own employment choices; that they can, having gained access to credit, improve their land, and invest and engage in entrepreneurial activities; that they have increased land tenure security and the capacity to manage the common assets. Women heads of households have now the legal standing to carry out economic transactions in the absence of their husbands and this is all the more relevant in a country in which old and new patterns of migration cause wives to be separated from their husbands for long periods. 8.7. Because of its innovative content and broad scope of application, the LCMP is perceived as a powerful policy framework on which to base the Government’s gender- related agenda of additional legislative changes. The LCMP is perceived by stakeholders (CSOs as well as officials) as a powerful tool for change that could catalyze additional legal (in both formal and customary settings), social, and cultural developments towards gender equality. Indeed, the potential improvements in women’s status brought about by the statute are a dramatic break with the century-old legal tradition sanctioning male supremacy. The high expectations stimulated by this recent legislation can be better understood if one recognizes that the LCMP is the result of a top-down legislative process rather than a response to a change in the social perception of gender roles. As a consequence, the legal framework is more in terms of laying a foundation for gender equality than the societal and cultural norms it might have been expected to represent. 175 The LCMP is applicable to all women in Lesotho, because it expressly encompasses both statutory and customary marriages (LCMP, Part I, Sec. 2). 176 LCMP, Sec. 19 (Amendment of Marriage Act). Early marriage, also referred to as child marriage, is the marriage of a young person (usually a girl) before the onset of adulthood as defined by the 1989 Convention on the Rights of the Child. 17’ LCMP,Part III. LCMP, Part 11, Sec. 3(3). 184 8.8. Therefore, it will be through the statute’s thorough and correct implementation and dissemination that the innovative legal framework may help reshape the cultural perception of gender relationships. Training and awareness campaigns on the LCMP are being planned as part of the U.S. Government-funded Millennium Challenge Account (MCA) Compact with the GoL. In addition, the Federation of Basotho Women Lawyers ( F D A - Lesotho) and other organizations are involved in spreading knowledge of the provisions of the act and to enhance its visibility. The MCA Compact activity’s goal is “to effectively enforce the LCMP through training and public awareness in order for Lesotho to realize socio-economic development” and to “have a ‘society that understands that in marriage husband and wife are equal partners.” (Federation of Women Lawyers, 2007). The training program aims to enable the judiciary, law enforcement agents, and banking institutions to implement the LCMP appropriately. The training therefore will target groups such as legal fraternities, financial institutions officers, police, corrections officers, and civil society organizations (CSOs). 8.9. The awareness campaign is intended to draw the attention of women, men, and community leaders to the LCMP and its implication for women’s status and participation in the economic and social spheres. Because the LCMP is written only in English, one component of the dissemination plan is to translate the act into Sesotho to make it accessible to those, mainly the rural population, who speak only this language. The Ministry of Gender, Youth, Sports and Recreation (MoGYSR) and the Law Reform Commission, which are already actively pursuing a gender-responsive agenda, are seizing the opportunity offered by the enactment of the LCMP to push for more changes, as they work with donors on additional legal reforms to foster gender equality. 8.10. The enactment of the LCMP is part of a positive trend of reforms that have modernized Lesotho’s legal system and made it more conducive to advance the status of women. Developments in the last five years include the efforts to curb sexual violence, which resulted in the enactment of the Sexual Offence Act of 2003 which consolidates and repeals existing laws on sexual offences. It criminalizes rape, including marital rape, and prescribes minimum penalties for sexual offences. Sexual violence is major problem in Lesotho, and, worryingly, on the increase’79 As already noted, the findings of a recent study on the magnitude and characteristics of sexual violence in Lesotho speak of a shockingly high level of forced sex, with 25 percent of women surveyed reporting being physically forced to have sex at some point. It was estimated that over their lifetime Basotho women experience forced sex more than three time as much as neighboring South African women who themselves suffer an alarmingly high incidence of rape.’*OHowever, sexual abuse is largely under-reported. The study further found a higher reporting of cases of sexual violence in areas where an awareness-raising program was being implemented, leading to the conclusion that education is key to curb this disturbing phenomenon. In light of the above, the creation of a Gender and Child Protection Unit of the Lesotho Mounted Police in charge of addressing sexual abuses of children and women is all the more relevant. Its work resulted in increased awareness, increased social rejection of these activities, and higher reporting of abuses. Finally, the Children’s Protection and Welfare Bill which covers, among others, children’s rights relating to forced and early marriage and other harmful cultural practices, is expected to be passes by the Parliament shortly. African Development Bank, 2005 ’79 The Lesotho Sexual Violence against Women Study was conducted from February to July 2003 in Maseru and Maputsoe by interviewing a random sample of 939 sexually active women aged 18 to 35. See Brown et al. (2006). 185 8.1 1 . Part of the positive trend of ‘women-friendly’ reforms is the establishment of reform commissions working with donors-primarily the Millennium Challenge Corporation (MCC) -- to prepare new gender-responsive legislation such as the Land Bill. 8.12. The LCMP and the other legislative changes create a “space” for women to expand their rights and, potentially, improve their own and their families’ social and economic situation. The enactment of an equitable legal framework cannot, by itself, realize the advancements in women’s status which the recognition of new rights and legal capacities seems to imply. It can, however, create an enabling environment within which women can become more and more assertive of their rights and develop a new awareness of their capacities and entitlements. 8.13. Improvement in the legal status of women is, furthermore, all the more relevant since women’s ability to contribute to social and economic advancement at the household and community level is tightly linked to their enjoyment of personal and economic freedoms. Women’s legal empowerment has a potential positive impact on issues such as women’s and children’s health and education, efficient allocation of household resources, and women’s economic productivity. The enforcement of an equitable legal framework has the potential, when coupled with women’s economic empowerment and a shift in the perception of gender roles, to enable women to gain control over personal choices, such as number and spacing of children and safe sexual practices to protect themselves from STDs and HIV/AIDS. Ultimately, women’s capacity l.0 exert a more active role in society through the exercise and enforcement of their rights could itself accelerate change in the traditional cultural perception of gender roles. B. BARRIERS TO THE IMPLEMENTATION AND ENFORCEMENT OF 4 GENDER RESPONSIVE LEGAL FRAMEWORK 8.14. However, for this to happen, legal reforms must be accompanied by effective implementation and other contextual conditions. Therefore these positive developments notwithstanding, the GoL still needs to address and overcome a number of issues that impede women’s full participation in the country’s efforts in economic development and poverty reduction. Limited access of women to justice, lack of harmonization among different pieces of legislation and survival of gender-discriminatory legal provisions in ordinary laws as well as in the Constitution itself, coupled with enduring gender-discriminatory cultural values, hinder the GoL’s gender-responsive agenda. In addition, Basotho customary law, which regulates legal matters among a large section of the Basotho population, is largely discriminatory against women and hampers the impact of the latest statutory changes. Finally, a general lack of clarity in the legal system - coupled with a low institutional capacity to embark into a comprehensive revision of the legislation - affects the capacity of legal professionals to correctly administer the law, creating inconsistencies in the interpretation and implementation of laws and statutes. Specifically, the following seven issues must be addressed and overcome: 1, Coexistence of two legal systems and the discriminatory content of customary law 2. Lack of capacity and gender sensitivity of the judiciaryAega1 professionals and government officials 3. Women’s limited access to justice and lack of awareness andor appreciation of their own rights 4. Cultural perception of gender roles 5. Lack of harmonization of legal provisions 186 6. Flaws in Constitutional protection of gender equality (clause 18(4)) 7. Failure to domesticate international treaties, such as the United Nations Convention on the Elimination of All Forms of Discrimination against Women (CEDAW) and the International Covenant on Economic, Social and Cultural Rights (ICESCR). Roles of customary and statutory law 8.15. In Lesotho, statutory law and Roman Dutch common law (together comprising the formal legal system) operate alongside traditional customary law. Under the British protectorate, Roman Dutch common law was applicable only to the white settlers and the indigenous African (Basotho) people who could demonstrate that they had abandoned their traditional way of living and had adopted a European 1ifes.tyle. The rest of the population was to be governed by customary law. On attaining her independence on October 4, 1966, the Kingdom of Lesotho continued the legal dualism, and maintained the same criteria of applicability of the two different systems. The coexistence of the two parallel legal systems creates particular confusion and inconsistencies in family, property, and inheritance law. Both legal professionals and lay people often have troubles in identifying which law applies to a case. Moreover, the discriminatory content of customary law is a major impediment to the protection of women’s rights. Box 8.1: Customary and statutory law: How are their boundaries determined? The Administration of Estates Proclamation of 1935 is an example of ambiguities in applying these two legal systems. The statute distinguishes between three types of estates: those of non- Basotho people; those pertaining to Basotho who have shown that they have abandoned the customary way of life and, if married, have solemnized their marriage under formal law; and those pertaining to Basotho who still live according to custom. The last category is expressly excluded from the application of the provisions of the Administration of Estates Proclamation. In reality, the distinction between the last two categories is difficult to make, as the demarcation between people who live a customary way of life and those who have abandoned,itis inevitably blurred, especially in the case of formally educated people in urban areas. Indeed, many who embrace a “Western” lifestyle still follow certain aspects of their traditions. In addition, even common law marriages in Lesotho usually are concluded with elements of custom, making it difficult to decide which law should prevail. 8.16. In addition, this legal duality creates contradictions that often frustrate legislators’ efforts to modernize outdated laws and bring Lesotho’s legal system in line with international standards. Cross-references between laws and derogations in favor of customary law weaken the effect of new provisions that mandate gender equality and uphold women’s enjoyment of human rights. 8.17. The dual legal system is mirrored by a dual judicial system under which statutory and customary courts work side by side. The lack of clear lines of demarcation between the two systems and the tendency for both the users and professionals of the law to resort to customary law when in doubt are major drawbacks for women. Indeed, customary law rarely upholds women’s rights, and considers them as perpetual minors under male. guardianship. 8.18. Women’s land rights are negatively affected by the duality of Lesotho’s legal system. The 1979 Land Act, as amended by the 1992 Land Amendment Order, regulates land rights, along with customary law. Under formal as well as statutory law, all land in Lesotho is vested in the Basotho nation, and it is administered by the State or by the Chiefs for the Basotho People. No person can hold title to the land other than in the form and to the extent provided for 187 by formal or customary law, Le., through leasehold or allocation.’8’ There is no individual ownership of land in Lesotho. In rural areas, however, formal land law never took root and traditional forms of land tenure are commonly applied (Pule and Thabane, 2004, p. 288). In addition, the land allocation committees, “although administering a statutory law, still apply customary law principles in land allocation”; almost all rural land in Lesotho is held under customary tenure (Matashane-Marite, 2005). Although a gender-responsive revision of the Land Bill is expected, and should ultimately result, when enacted, in a more equitable land law, its impact will not be felt if customary law will remain the main source of land regulation. 8.19. Formal law recognizes women’s rights to hold, register, and inherit land, while customary law denies women access to land in their own right. According to customary law, women acquire rights to land through their fathers, husbands, male children or male relatives pertaining to their husband’s family. The legal minority status to which customs relegate women translates into their incapacity to be the beneficiaries of land allocations and to inherit tenure rights to land. 8.20. Women-headed households, although able to access and use land, cannot dispose of , it and do not enjoy tenure security. Traditionally, widows retain the right to use the land of their deceased husbands. Under customary law, however, widows hold the land in trust for the heir (be it the eldest son or, if there are no male children, the husbands’ relatives) and can neither freely administer it nor dispose of it. In addition, there are a growing number of cases of dispossession of HIV/AIDS widows, forced by their late husbands’ relatives to vacate the land. Lack of kno\\rledge of existing protective laws prevents women from enforcing their rights to land (African Development Bank 2005). Box 8.2: Impact of HIV/AIDS epidemic on women’s customary rights to land Under customary law, widows traditionally could retain the right to access and use the property left by their husbands by joining the heir’s household (widow inheritance). The HIV/AIDS epidemic is jeopardizing the survival of this custom and widows’ access to productive resources. The stigma attached to the disease has worsened the social status of HIV/AIDS widows, who often are blamed for the deaths of their husbands, dispossessed of his land, and deprived of productive assets by their husbands’ relatives. In one rural highland community with a high prevalence of HIV/AIDS, a man died. Shortly thereafter, his wife fell sick. It became known in the village that she was HIV positive, and her in- laws accused her of having infected and killed their son. The woman was dispossessed and sent back to her parents’ house, deprived of the principal means of livelihood for herself and her children. Other issues in implementation 8.21. In the formal setting, lack of adequate gender sensitization and awareness of women’s rights by legal and other professionals, coupled with unfriendly and costly procedures, jeopardize the capacity of the judicial system to uphold women’s rights. Lesotho’s judicial system has a number of shortcomings that prevent it from functioning properly and fulfilling its purpose as a nondiscriminatory law enforcement instrument. High costs of proceedings, high centralization of the courts, biased attitudes, and magistrates’ lack of training severely limit access to justice for the poor in general and for women in particular. In addition, 18’ Urban land is allocated in leasehold by committees appointed by the Minister of Local Government, and are chaired by the Principal Chief. Rural land is allocated by Village Development Councils, comprised of the local chief and seven members elected in a public meeting (Adams et all, 1999). 188 “delay in disposing of criminal charges, lenient sentences and high rates of acquittals” (African Development Bank, 2005) create lack of confidence in the system, discourage recourse to courts, and frustrate a plaintiffs demand for expeditious justice in instances-such as domestic violence-in which late redress is often in vain and/or dangerous for the woman.’82 8.22. Lack of professionalism and specific gender training and sensitization are probably the main factors affecting the justice system’s ability to effectively respond and provide redress to women plaintiffs. Traditional patriarchal values often bias the interpretation of the legal provisions regulating issues such as women’s rights to inheritance, land ownership, and domestic violence. Decisions on such matters are rendered more on the basis of the magistrate’s personal beliefs and attitudes than according to the law, thus jeopardizing or negating women’s ability to obtain fair judgments. Many prosecutors in the Magistrate Courts have no professional legal training. As a result, a high number of cases are lost due simply to shortcomings in the presentation of evidence by unqualified prosecutors.’83 8.23. The lack of proper training of professionals working in the financial sector can result in the erroneous application of new laws. Specifically regarding the LCMP, it is reported that financial institutions at times have mistakenly denied loans to male applicants on the grounds that, according to the LCMP, they needed written consent from their wives, even when the loan was not applied for against the couple’s joint property. Such misinterpretations deepen the social rejection of the statute, particularly among men who perceive the new regulation as unduly limiting their economic freedom. 8.24. Finally, government officials who do not appreciate the role that gender issues play in the design and implementation of policies, budgets and legislation will be unable to effectively address socioeconomic matters, severely affecting GoL’s ability to advance the country’s socioeconomic situation. 8.25. Access to justice is limited also by women’s lack of awareness of, or ability to act upon, their own rights and cultural biases which, although discriminatory, often are embraced by women themselves as a result of the socialization process. Women often lack information about their rights and how to access the mechanisms to enforce them. Poor dissemination of the laws’84and the confusion created by the dual legal system make it difficult for women to properly assert and enforce their rights. In addition, statutory law is written only in English, while a large portion of the population, especially in rural areas, speaks and understands only Sesotho. Legal aid and assistance services are lacking. A legal aid unit under the Ministry of Justice, Human Rights and Rehabilitation provides in Maseru legal assistance to the poor. Its capacity to operate is hindered by a lack of resources and staff. Furthermore, the fact that the IE2 The High Court is understaffed and has an enormous backlog of cases. Judgments take months to obtain and, according to the Registrar of the High Court, they are still dealing with cases as old as from the year 2000. A report on the criminal judicial system in Lesotho commissioned by GoL found that the criminal justice system functions at an efficiency level far below expectation and, as a result, the public has lost confidence in the system. ‘83 In some cases of domestic and sexual violence, the prosecutors are police officers who do not have the same training or incentives as most of the defense lawyers. As of the period following the last consolidation (1992), laws have not been easily accessible to the public and the judicial officers. In fact, accessibility to the laws prior to the consolidated 1992 volume and even those in the 1992 volume are not guaranteed. Most volumes are out of print and the government printer lacks the capacity to keep up to date with the production and printing of the gazettes and law volumes. Not a single consolidated volume of the laws passed by parliament has been published since 1992. 189 office is located in Maseru makes the legal aid service difficult to access by a great part of its ~ potential ~ s e r s . ”Finally, cultural biases embraced by men and women prevent social acceptance of the most innovative legislation, such as the LCMP, which is perceived as endangering the customary social structure of the Basotho. In particular, legislation enacted to enable women to protect themselves against STD/HIV/AIDS infection is less effective without a shift in power relations between women and men. 8.26. In addition, lack of harmonization of legal provisions hinders the potential impact of legislative developments, perpetuating women’s disempowerment and economic and social subordination to men. Although the enactment of new laws has opened new channels for women’s participation in social and economic life, Lesotho’s legal framework is still far from being a homogeneous set of laws upholding gender equality. Discriminatory provisions scattered in old statutes (some dating back to the nineteenth century) still create legal barriers for women in areas such as access to resources, inheritance, and entrepreneurship; and hinder the impact of the new legislation. For example, notwithstanding the LCMP provision sanctioning the right of a woman to become director of a company, the Companies Act has not been amended accordingly and still prevents women from occupying such positions.’86Bank regulations still put women in the same category as minors and other persons and bodies who may be subjected to special rules for the withdrawal of money.I8’ 8.27. One area that would benefit from a comprehensive revision is inheritance law. For example, the 1992 Land Amendment Order recognizes inheritance rights of widows and eliminates the privileges of first-born male children. Nevertheless, pieces of legislation still exist- -the laws of Lerotholi being one of them-which deny women inheritance rights and countermand gender-responsive-reform interventions. The LCMP, which provides for amendments to certain laws, does not address this matter. However, some piecemeal reforms seem to be underway 8.28. The Constitution, although protective of gender-equality, contains limitations and derogations to the principle of nondiscrimination. The Lesotho Constitution of 1993 repeatedly sanctions the right to equality and nondiscrimination. Section 4 of the 1993 Lesotho Constitution guarantees fundamental human rights and freedoms to all persons regardless of sex or status. Section 18 mandates freedom from discrimination, and Section 18(3) includes sex as a prohibited ground of discrimination. Section 19 provides that everyone is equal before the law and entitled to equal legal protection. Section 26 provides for the obligation of the GoL to enact policies guaranteeing equality and justice for all citizens irrespective of, among other things, gender. Section 30 mandates equality in work and employment conditions for men and women, and the enactment of appropriate measures in favor of pregnant women. 8.29. However, the Constitution contains derogations and limitations that weaken the protection of the right to equality. First, it excludes personal law matters from the scope of the nondiscriminatory clause, providing that clause 18 (l), which mandates freedom from discrimination, shall not apply “to any law to the extent that that law makes provision (. ..) (b) for the application, in the case of persons of any such description as is mentioned in subsection (3) (or of persons connected with such persons), of the law with respect to adoption, marriage, divorce, burial, devolution of property on death or other like matters which is the personal law of Lesotho Review of Civil Legal Systems - Final Report (provided by MCC) 186 The Companies’ Act No. 25 of 1967, Sec. 144(1)@). 187 Lesotho National Development and Savings Bank Order No. 8 of 1971. 190 persons of that description?.?88In addition, although the Constitution requires that customary law, common law, and statutory law be consistent with its provisions, it also states that ?the principles of freedom from discrimination shall not apply to any law which provides (...) (c) for the application of the customary law of Lesotho with respect to any matter in the case of persons who, under that law, are subject to that law?.?89.These provisions create loopholes in the constitutional guarantee of women?s right to non-discrimination which allow discriminatory practices and provisions to resurface in the system. The survival, sanctioned by the supreme law of the country, of areas of lawful discrimination constitutes a major obstacle for the implementation of gender-responsive legal reforms and the harmonization of the legal framework. 8.30. Finally, although Lesotho is a signatory to most international instruments that sanction women?s human rights and gender equality, it has failed to translate such provisions into enforceable domestic law. 8.3 1. While, once ratified, international treaties are useful instruments to which magistrates and lawyers can resort, they need to be translated into national legislation for their provisions to be enforceable. Most treaties, although ratified, are still awaiting domestication in Lesotho.IgoThe most relevant is CEDAW. However, it should be noted, that, in August 2004, the GoL repealed the broad re~ervation?~? it had made upon ratification of this treaty. Therefore, CEDAW now can be translated into enforceable national legislation. Box 8.3: Internationaltreaties concerning women?s rights ratified by Lesotho Lesotho has ratified most of the international human rights instruments dealing with women?s rights. Those ratified are the United Nations Charter (1945), Universal Declaration of.Human Rights (1 948), International Covenant on Economic, Social and Cultural Rights (ICESCR) (1 966), International Covenant on Civil and Political Rights (1966), African Charter on Human and People?s Rights, and the Convention on the Elimination of All Forms of Discrimination against Woman (CEDAW) (1 979) 192 c. NECESSARY IMPLEMENTATION STEPS 8.32. In sum, the GoL?s action to create a gender-responsive legal framework is noteworthy. However, effective implementation of accomplished reforms and additional legal developments in matters such as inheritance and land rights, customary law, and harmonization of the legal framework are still required. Below is a list of possible implementation steps aiming at addressing and possibly correcting the distortions affecting the enforcement of Basotho women?s rights. The order of the recommendations does not necessarily ?? Lesotho Constitution, Sec. 18 (4)(b). Lesotho Constitution, Sec. 18 (4)(c). 19? In some cases, the Constitution itself prevents domestication. See Lesotho Constitution, Sec. 2, which reads: ?This Constitution is the supreme law of Lesotho and if any other law is inconsistent with this Constitution, that other law shall, to the extent of the inconsistency, be void.? 19? The original reservation, made upon ratification, read as follows: ?The Lesotho government?s ratification is subject to the understanding that none of its obligations under the Convention especially in Article 2(e) shall be treated as extending to the affairs of religious denominations. Furthermore, the Lesotho government declares that it shall not take any legislative measures under the Convention where those measures would be incompatible with the Constitution of Lesotho.? The surviving derogation concerns only conflicting provisions regarding ?constitutional stipulations relative to succession to the throne of the Kingdom of Lesotho and law relating to succession to chieftainship.? Ig2 Ratified by Lesotho on August 22, 1995. 191 reflect their importance in terms of potential impact. Further revision of the laws and statutes- including the Constitution-is, by all means, crucial; however, the process to do so is lengthy and the amount of political will required very high. For this reason, the authors highlight as priorities a set of measures based on their short-term feasibility. The authors believe that this list should be seen as a set of actions and programs which could be developed over the short and medium term. 8.33. Given the centrality of the LCMP and its role as groundbreaking advancement in the modernization of the Lesotho legal system, efforts should be made to ensure its correct interpretation and implementation. For the LCMP to have the anticipated impact on Basotho society, both professionals and lay people need to be equipped to apply its provisions. Training, sensitization and awareness campaigns targeting professionals and/or the public at large should be scaled up and planned periodically on a regular basis. Given the apparent lack of clarity among legal and other professionals on how the LCMP fits within the existing legal framework, the pieces of conflicting legislation should be revised and harmonized with the LCMP. Finally, an authoritative opinion by the Attorney General clarifying the correct interpretation and the scope of application of the LCMP should be obtained. Such opinion should serve to educate and raise awareness of the meaning - and content - of the LCMP among professionals and - provided it is given proper publicity - the public at large. The opinion, which should highlight in particular the fact that the statute supersedes any discriminatory provisions contained by any law, including customary law, on the relevant subject matter, would further contribute to the clarification of the relationship of existing legislation with the LCMP. ,8.34. As to the implementation and modernization of the whole legal system, the following suggestions may be considered. 8.35. Training of trainers is the first step towards an effective and sustainable knowledge- creation process among professionals and lay people. In the effort to spread legal awareness and literacy and understanding of the benefits of gender equality for men, women and the society as a whole, the GoL could consider the creation of a core team of local trainers. Such trainers will then have the skills and expertise to develop the relevant tools to train and educate the different target groups. 8.36. Education on the links between gender equality and socioeconomic growth is essential to demonstrate the value of investing in women?s empowerment. Educating men and women, including legal professionals and government officials, on the value of gender equality for poverty reduction and their own family well-being will facilitate acceptance of gender- responsive legislation, as a powerful instrument for gender equality, and advance the gender discourse at the social and government level. 8.37. Fostering women?s awareness of their own rights as well as their access to mechanisms, procedures, and institutions that protect such rights are key to achieve effective implementation of existing protective legislation. Given the existence in Lesotho of gender-responsive legislation, the challenge. is to make it work for the intended beneficiaries. Experiences of gender-responsive legal reforms in similar settings have shown that without parallel efforts toward education and sensitization about the new legal framework and the principles upheld by it, women?s empowerment and advancement cannot be achieved. Given the high literacy of Basotho women and their exposure to economic activities, training and education inputs can be expected to make a significant impact. On the other hand, and for the same reason, awareness and legal literacy campaigns must be innovative and not follow the usual pattern that assumes utter ignorance on the part of men and especially women. Possible steps are to: 192 1 . Scale up and expand legal literacy programs targeting women and men of all social backgrounds in both rural and urban settings. Grassroots organizations already operating in the field should be capacitated and brought into partnership with the relevant governmental bodies, as the latter usually are better equipped (vis-a-vis human resources, logistic, and procedures) to act in such settings. 2. Include functional literacy in the legal literacy campaigns, so that men and women would be equipped with skills to utilize the knowledge acquired through the legal literacy campaigns. 3. Scale up public awareness campaigns through the media. 4. Through appropriate policies, encourage the creation of legal aid clinics in partnership with universities and NGOs. 5 . With the help of international donors, strategic litigation on key issues relevant to gender equality should be promoted. Strategic litigation, by bringing to court carefully selected cases, could serve to further clarify the content and reach of gender-responsive legislation. 6. Capacitate the Ministry of Justice and other relevant institutions in charge of disseminating the law. Translation of laws in Sesotho and production of bilingual legal glossaries also should be on the agenda. 7. Introduce gender-related legal education and gender-sensitization in school and university curricula. This should be done keeping in mind the importance of early gender education for boys and girls of primary school age. Drop-out rates after primary school could frustrate the purpose of gender-related educational school programs that target only the middle school.age range.’93 8.38. Education and gender-sensitization of magistrates, legal professionals, law enforcement officers, custodians of customary law, policy makers and all other relevant professionals and officers, such as bank employees, financial agents, and civil servants, are needed to see adequate and correct implementation and protection of women’s rights. As noted, lack of proper training of professionals called on to apply and enforce the law results in inadequate protection of women’s rights. The particular complexity of Lesotho’s dual legal system as well as the coexistence of conflicting statutory provisions require adequate understanding of conflict of laws rules and rules to harmonize national law. In addition, proper training and sensitization of policy makers will enable them to actively promote change in both the formal and customary legal system. The following measures could be contemplated: 1. Increase capacity of legal, law enforcement professionals, custodians of customary law and policy makers through training sessions and workshops on (a) scope of application of statutory/common law versus customary law and (b) content and scope of application of newly enacted statutes. 2. Introduce courses on women’s rights and women studies in law school curricula. 3. Implement gender-sensitization programs for legal professionals, law enforcement officers, custodians of customary law and policy makers. ‘93 Programs targeting young children might be very effective in diientangling girls’ and boys’ socializations from predetermined patriarchal structures and role models. 193 8.39. Initiate the process for a revision of certain aspects of customary law’9J.A joint review of the laws of Lerotholi and of unwritten customary laws and practices by the GoL and traditional leaders should be encouraged in order to identify which aspects of custom should be amended and which should be retained but revisited to respond to changing circumstances. In particular, traditional norms regulating inheritance, land, and family matters should be brought in line with the recent developments in the formal legal system. Once the relevant areas are identified, the process for the amendment of customary law should be initiated in Parliament.. 8.40. Reform of outdated statutes sanctioning gender inequality and reconciliation of contradictory provisions contained in different pieces of legislation is necessary to realize the full potential of the ongoing modernization of the legislation affecting women’s rights. Although a long-term goal, feasibility studies could be conducted on the following: 1. Broaden the scope of legal reform by setting up a commission to harmonize laws that regulate the same subject matter, including inheritance, land rights, and corporate and labor issues. 2. Eliminate clauses that exempt customary law from the application of gender- responsive legislation. 3. Clarify the scope ofthe application of statutory/common and customary law. 8.4 1. Finally, given Lesotho’s good record of ratification of international gender-related treaties, domestication now should be on the agenda. All of the above should be acted upon, and in order, to give effect and meaning to the recent legal advancements that have taken place in Lesotho. However, the survival in the national Constitution of provisions substantially permitting forms of discrimination against women undermines efforts to build a coherent and harmonious legal framework supportive of women’s rights. Even though revision of the Constitution is a delicate matter that touches on issues of national sovereignty, the negative impact of discriminatory Constitutional provisions on the overall endeavor for gender equality should not be overlooked. Customary law can be amended, like any other law, by Parliament, which chiefs are part of. The chiefs sit in the senate, the higher house of Parliament, which is composed by 22 chiefs and 1 1 other members nominated by the King ’ in accordance with the advise of the Council of State. The process for the amendment of laws can be initiated by ministers, members of Parliament, and the Lesotho Law Reform Commission (Information provided by Ms. Limpho Mahase, Programme Coordinator, Gender Equality in Economic Rights, Lesotho Ministry of Gender, Youth, Sports and Recreation) 194 a k c .- C - c m .- V t - - t m .- E .- E L .- e 5( a .- c a *E - w c V c E - 01 b 3; 01 .- > - c m .- V - c a E c. d W c .- C .- V a $ ? c m .- E .- L 5( E - s z Y a * 4 LI 0 C .- Y 0 m L ." v Il , e *z z TECHNICAL APPENDIX 1 HBS 1994-95 AND 2002-03 DATA QUALITY AND COMPARABILITY This appendix covers technical issues in the design and implementation of the two surveys which affects the analyses which can be done, and quality of the results. These issues have been mentioned in the body of the report as well, but are elaborated here. 1. Data Quality During the preparation of the Lesotho poverty analysis, World Bank staff and consultants found inconsistencies and, overall, a low level of reliability in the data. We interviewed Bureau staff and consultants on the procedures followed during the stages of field implementation, data capturing, and data handling; and reviewed consultant reports. The information we gathered suggests that, while the quality of the 2002-03 HBS data is better than that of the 1994-95 survey, both suffer more than usual from quality issues. These affect the interpretation of the results. Data Capturing The HBS data was not consistently captured (transferred from survey to computer) by best practice procedures such as double entry or other methods to guard against human error during keying of data. The HBS 2002-03 data was captured initially and verified concurrently, but due to time pressure and capacity constraints, this procedure later was abolished and only a single entry without verification was done. Consistency checks between captured data and a few questionnaires from both 1994-95 and 2002-03 indicate substantial entry errors. The data quality report by Roberts and Woolard (2001) on the 1994-95 dataset found data entry errors in 5 of 8 questionnaires. Six questionnaires from HBS 2002-03 had been left behind because entry errors had been found in them. A check of these six after they were corrected revealed additional data entry errors in all of them. Related to the much-delayed keying of the HBS 2002-03 data,'95parts of the data are missing in the final dataset. For instance, local price data were collected but never keyed in. Only income data at the household level are available in the final dataset. The income data at the individual level reportedly was keyed but left out in the conversion to the SPSS format. Data Completeness and Consistency Although the 1994-95 data had a planned sample size of 5100 households, the final sample is 3 784 households, totaling 20449 individuals. Reportedly, a large number of the households had to be dropped after data collection to bad data quality. The final sample has rural and urban response rates of 72.4 and 76.5 percent, respectively. Similarly, HBS 2002-03 had a planned sample size of 6882 households, whereas the final sample is 5992 households, corresponding to 27678 individuals. Rural and urban response rates were 100.0 and 75.0, respectively, corresponding to an overall response rate of 87.1 percent. In comparison, the 2004 Lesotho DHS had a planned sample of 9903 household and interviewed 8592, corresponding to 86.8 percent. A large number of the dropped households were no longer were occupied by anyone. Excluding the dropped 19' The 2002-03 HBS data w as not entered at HQ until years after collection, reportedly because data-entry staff were dissatisfied with their remuneration for the task. 199 LSMS surveys carried out in other countries around the households, the response rate was 95.2.‘96 same time show average response rates of 88.7 (United Nations 2005). Some basic checks across variables in different sections of HBS 2002-03 related to consumption and well-being show signs of incomplete information. For instance: 28 percent of households that reported their main income source as “Farming” did not report any sold produce, consumption of own agricultural products, nor expenditure on agriculture. The corresponding numbers for urban and rural were 20 percent and 29 percent, respectively. Seasonality does not seem to be the explanation for this pattern as the percentage varies from 19 to 40 over the months. 26 percent of households that reported having at least 1 family member working as a subsistence farmer did not report any sold produce, consumption of own agricultural products, nor expenditure on agriculture. 49 percent of households that reported that they had a household business did not report any sold products, consumption of own products, nor expenditure on the business. 0 47 percent of households that reported their main income source as “Other household business” did not report any sold products, consumption of own products, nor expenditure on the business. 40 percent of households whose main source of income was remittances did not report receiving any remittance in the last month. 0 40 percent of households in 2002-03 indicated that they did not have any monetary income at all. Households with no income are only weakly related to overall consumption per capita. In 1994-95, 68 percent of households reported no monetary income. Both the large number of households dropped in 1994-95 due to poor quality and the lack of consistency across variables in the 2002-03 survey suggest that field implementation was less than optimal. On the other hand, these weaknesses could be related to the survey instruments, which in many cases do not follow standard and best practices (see Grosh and Glewwe, eds. 2000). For example, the questions on labor force status and sector of economic activity have a confusing structure and skip pattern that does not enable capturing standard labor market concepts. In addition, the definition and questions used to define a household member do not make clear who actually lives and consumes in the household In addition, the consumption sections of the 1994-95 and 2002- 03 survey differ significantly so that comparison over time is not possible (see below). Data Handling The data quality report from Roberts and Woolard (2001)197 on the 1994-95 dataset raised a number of additional quality issues related to data handling. The main problems identified were (1) large number of duplicates, (2) wrong weights, (3) human errors in sorting monthly and 196 It is common for DHS to have higher response rates than expenditure surveys because keeping a diary or recording expenditures is more complicated than the DHS. 197 The report is a short-status report based on a three-day visit. 200 annual data, and (4) bad quality of income and domestic own consumption data. All of the above problems also were encountered during the HBS 2002-03 preparation. Joint efforts between BoS and post-data collection technical assistance from donors addressed problems 1-3 in both datasets. The fourth point could be addressed only during survey implementation. Both the 1994-95 and 2002-03 datasets have inconsistencies between their theoretical stratification and the distribution in the datasets. The stratification and sample described in chapter 1 of the 2002-03 Household Budget Survey Report Lesotho (BoS 2007) do not match the actual 'datasets. Both datasets contain observations in strata in which there should be no observations. For instance, in 2002-03 Thaba Tseka reported 24 households in the Lowlands, even though Thaba Tseka is solely a Mountain area. Furthermore, both surveys have mismatches between reported numbers of households and actual numbers of households in each stratum. Both the 2002-03 and 1994-95 HBS datasets have been corrected for outliers and missing data. These processes and the impact on the data remain undocumented. 2. Comparability To track poverty over time, a time-consistent measure of poverty is needed. An inconsistent measure can substantially over- or under-estimate poverty, leading to erroneous conclusions. This appendix compares the two surveys to see whether they are designed ways that can be expected to give time-consistent measures of poverty. Examples from India have shown that even small changes in survey designs can lead to significant changes in measured consumption, thereby invalidating comparison over time (Scott and Amenuvegbe 1990, Deaton and Grosh in Grosh and Glewwe 2000, Deaton and Kozel2005). Survey Design Both surveys are designed to be national surveys and use two-stage cluster sampling with geographic areas used as strata. The sampling frame was changed from the 1986 census to the 1996 census between the 2 surveys. With the change of frames, some recoding of boundaries and definition also took place. The reclassification did not relate only to primary sampling units (PSUs), enumeration areas (EAs), and definition of urban and rural. It also included a change in the classification of the Thaba Tseka district from 1994-95, when it was considered a part of both the Rural Mountains and the Senqu River Valley agrological zone, to being classified as only Mountain after 1996 (agrological zones are used for stratification). These changes could impact the comparison between specific areas within the country over time, but should not significantly impact the comparability at the national level way since they both are national representative surveys. Both surveys were intended to run for 12 months (June 1994-May 1995 and November 2002-October 2003), but delays and initial problems meant that HBS 2002-03 ran for 14 months. Questionnaire Design The surveys have the following major similarities and differences. Both surveys: Used a 4-week diary in combination with recall questions to collect consumption data, Had weekly supervision visits with literate households and two visits a week with illiterate households. Transferred the diary into a preprinted schema from which data was keyed to a computer. 20 1 The surveys differ as follows: The diaries in the surveys do not have identical questionnaires to record item, unit, quantity, value and type of transfer. Particularly recording types of transfer varies as a new structure was used in 2002-03 (Table A2.1). For instance: o There is no separate recording for farming and other household businesses in 1994-95, only a joint recording of household businesses. HBS 2002-03 has separate reporting for farming and business. o Operating cost for households (could be both farming and nonfarming) was recorded in the diary in 1994-95, but by recall in 2002-03. In addition, separate business and farming data were not collected for 1994-5, but they were in 2002- 03. The overall design of schemas for collection consumption data used in the two HBSs are not the same. A number of consumption goods were not collected with the same reference period. The overall headlines and subheadlines do generally not follow the same outline, and some headings exist in one year’s survey but not in the other. See the differences in Table A2.2. Observations on the differences are that: o There is a shift in a large number of variables from 4-week diary to l-month recall. It is not clear how the diary was used for the l-month recall questions. o Education, transportation, and communication expenditures were collected using a 12-month recall in 1994-95 but a 4-week diary in 2002-03. Transport and communication increased from 0.4% of overall consumption in 1994 to 5.0% in 2002-03. During the same periods, education decreased from 4.1% to 1.7% of household consumption, respectively. The reduction in educational expenditure could be explained by the roll-out of free education between the two surveys, but the decreases in the other two categories are hard to explain. o Household textiles were recorded via l-month recall in 1994-95, but 12-month recall in 2002-03. Recreational expenditures were recording using 12-month recall in 1994-95, bu# 4-week diary in 2002-03. Longer recall periods often lead to lower estimates of consumption (Scott and Amenuvegbe 1990) o The 1994-95 dataset has consumption for each product by type of consumption (bought, gifts-in-kind, own production). The 2002-03 set has own production lumped in a few overall headings and only one category for gifts-in-kind, which includes both food and nonfood. Furthermore, the gifts-in-kind data was transferred from the diary to the one category labeled “Giftshransfers in kind” located in the income section, not in the consumption section as normal. It cannot be ruled out that under the latter headline some element of monetary transfer could be included. Gift in kind increases from 0.9% of total consumption expenditures in 1994-95 to 9.4% in 2002-03. Food aid for the drought in 2002- 03 is believed to be included in gifts-in-kind, which could explain some of this increase. The two surveys use the same definition of what constitute a household and a household member. However, the first few questions used to identify household members are not 202 identical. The change of question order could impact the recording of household members and size. That in turn would influence the consumption per capita. As a result, we conclude that the surveys in 1994-95 and 2002-03 were different in a number of key areas that can have significant impact on measured consumption and thereby the comparability of the consumption aggregates. Real Consumption Growth One check on comparability can be a comparison of the relationship of the components of the aggregate to the whole. Breaking down the consumption aggregates by type of expenditure reveals unexpected changes in food and nonfood between the two years (Table A2.4). A first and massive difference is the consumption reported as gifts-in-kind. The average reported value increased approximately 1800 percent from 1994-95 to 2002-03. Unfortunately, in 2002- 03 we are not able to further decompose this value since gifts-in-kind was reported in a single variable that included both food and nonfood. Table A2.4 also shows that food consumption in real terms declined between the two surveys. However, it appears that the reduction in food consumption was offset by an increase in gifts-in- kind, possibly food aid. In addition, the overall growth was driven by growth in rural nonfood consumption. In sum, both the design of the two questionnaires as well as an analysis of the aggregates and their composition argues against drawing any conclusions about the evolution of household consumption or poverty in'lesotho over time. 203 Table A2.1 Coding used to record transfers in the diaries 1994195 2002103 Transaction Transaction Purpose 1. Cash expenditure Bought Private 2. Consumption of own produce Own produce Farming 3. Payment in kind from employer From own business Business 4. Gift-in-kind given Income 5. Gifts-in-kind received , Gi Wtransfer received 6. Sales of goods or services from GiWtransfer given household- operated informal business 7. Purchases for household-operated informal business 8. Cash taken from own business . Table A2.2 Method of collecting consumption data 2002103 1994195 4-week diary Food, beverage and tobacco, Clothing and footwear, Food, beverage and tobacco House services, Health, Communications, Recreation, Papers, books and stationery, Holiday package gifts, Education, Restaurants, hotels, Miscellaneous 1-month recall Clothing and food wear, Gross rent, fuel and power, Furniture, furnishings, household equipment and operations, Household utensils, Household operations, Writing, drawing equipment and supplies (not to be used in schools), Medical care and health services, Personal care, miscellaneous goods and services, Domestic Non-Consumption Expenditures (Cash), Other Cash Domestic Expenditure 1- month recall Furniture, furnishing and floor covering, Household Furniture, Household appliances, transport and textiles, Household appliances, Telephone equipment, communications, education, recreation, Vehicles, Recreational supplies, Jewelry, Travels abroad entertainment, and cultural services 204 Table A2.3 - Mean per adult equivalent expenditure in real terms (Maloti) 1994195 2002103 YOchanee Urban 370.28 289.85 -2 1.72 Rural 162.09 178.10 9.88 Region Maseru Urban 384.73 357.66 -7.04 Other Urban 338.15 248.10 -26.63 Rural Lowlands 187.37 164.72 -12.09 Rural Foothills 165.13 169.24 2.49 Rural Mountains 108.82 198.62 82.52 Rural Senqu River Valley 122.57 213.55 74.24 Total 182.21 202.71 11.25 Sources: HBSs and authors’ calculations using BoS original consumption aggregate. Table A2.4 - Mean per adult equivalent expenditure in real terms: Decomposed by type of expenditure (Maloti) National Rural Urban National 1994195 2002lQ3 Change 1994195 2002103 Change 1994195 2002103 Change Food total 104 86 -1 8 102 79 -23 129 112 -17 Food bought 67 73 6 61 63 2 123 108 -15 Food own produce 37 14 -24 40 16 -24 6 4 -2 Nonfood total 78 97 19 60 79 18 24 1 162 -80 Nonfood bought 77 94 17 59 75 16 24 1 162 -79 Nonfood own produce 0 3 3 0 4 3 0 0 0 Gifts-in-kind 1 19 18 1 21 20 0 16 15 Total consumption 182 203 20 162 178 16 Sources: HBSs and authors’ calculations using BoS original consumption aggregate. 205 Table A2.5 Difference in Poverty Line Using Bos and Real Fixed Poverty Line (%g Real consumption and one fixed poverty line Poverty headcount rate Distribution of the poor Distribution of population 199419 200210 Chang 199419 200210 Chang 199419 200210 Chang 5 3 e 5 3 e 5 3 e Urban 16.6 22.2 5.6 3.6 14.5 10.8 9.7 22.0 12.4 Rural 47.0 37.1 -9.9 96.4 85.5 -10.8 90.3 78.0 -12.4 Region Maseru 14.8 17.6 2.8 2.3 4.4 2.1 6.7 8.4 1.7 Urban Other 17.0 25.0 8.0 2.1 10.1 8.0 5.5 13.6 8.2 Urban Rural 38.3 38.1 -0.2 31.3 42.9 11.6 36.0 38.0 2.1 Lowlands Rural 50.9 43.1 -7.8 25.8 16.7 -9.1 22.3 13.1 -9.3 Foothills Rural Mountain 61.7 32.2 -29.5 26.0 20.9 -5.1 18.6 22.0 3.4 s Rural Senqu 50.4 34.8 -15.5 12.5 5.1 -7.5 10.9 4.9 -6.0 River Valley Total 44.1 33.8 -10.3 100.0 100.0 0.0 100.0 100.0 0.0 ~~ Real consumption and two poverty lines Poverty headcount rate Distribution of the poor Distribution of population 199419 200210 Chang 199419 200210 Chang 199419 200210 Chang 5 3 e 5 3 e 5 3 e Urban 15.7 22.2 6.5 3.6 14.5 10.9 9.7 22.0 12.4 Rural 45.4 37.1 -8.3 96.4 85.5 -10.9 90.3 78.0 -12.4 Region Maseru 13.6 17.6 3.9 2.2 4.4 2.2 6.7 8.4 1.7 Urban Other 16.6 25.0 8.4 2.1 10.1 7.9 5.5 13.6 8.2 Urban Rural 36.3 38.1 1.9 30.7 42.9 12.2 36.0 38.0 2.1 Lowlands Rural 48.4 43.1 -5.3 25.4 16.7 -8.8 22.3 13.1 -9.3 Foothills Rural Mountain 60.9 32.2 -28.1 26.6 20.9 -5.7 18.6 22.0 3.4 S Rural Senqu 50.1 34.8 -15.3 12.9 5.1 -7.9 10.9 4.9 -6.0 River Valley Total 42.5 33.8 -8.7 100.0 100.0 0.0 100.0 100.0 0.0 Sources: HBSs and authors’ calculations using BoS original consumptiqn aggregate. 206 REFERENCES Adair, T. 2007. “Desire for Children and Unmet Need for Contraception among HN-Positive Women in Lesotho.” Macro International Inc. (March) Adams et all, 1999. Land Tenure and Rural Livelihoods in Southern Africa. Natural Resources Perspectives, No. 39, February 1999 African Development Bank, 2005. Lesotho Multi-Sector Country Gender Profile (ADB/BD/IF/2005/270 - ADF/BD/IF/2005/225), Afro1 News. 2008. “More Aid for Lesotho’s Numerous Orphans.” 2008. ww.afrol.com/articles/24954. April 15. -. Alderman, H. J. Hoddinott and B. Kinsey and 2003. Long tern Consequences of Early Childhood Malnutrition. World Bank. Washington DC: Processed Ansell, N., and L. van Blerk. 2004. “Children’s Migration as a Household/Family Strategy: Coping with AIDS in Lesotho and Malawi.” Journal of Southern A$-ican Studies 30 (3) (September). AVERT. 2007. “HIV/AIDS in Lesotho.” Last updated Nov. 9. http://www.avert.org/aids- lesotho.htm Boehm, C. 2004. “Fields, Mines and Garments: Production, Social Reproduction and Changing Livelihood Pathways in Lowland Lesotho.” PhD diss. Institute of Anthropology, Faculty of Social Sciences, University of Copenhagen. Bowsky, Sara. 2004. “Matla ea Lesotho ke Sechaba: Hlahlobo ea Kokelo ea Bakuli ma Haeng”/”Lesotho’s Strength Is Its People: A Rapid Appraisal of Home- and Community-Based Care. CARE-Lesotho.’’ Family Health International with Support from USAID Regional HIV/AIDS Program. April. Brown, L et all 2006. “Sexual Violence in Lesotho”.Studies in Family Planning, Volume 37 Number 4 December, pp. 269-280. Bureau of Statistics (BoS). 1988. Incomes, Expenditure and Consumption of Basotho Households, Tabulation Report from the Household Budget Survey 1986-87. -. 1996. Population Census Statistical Table. Vol. 11. -. 2000a. End-Decade Multiple Indicator Cluster Survey (EMICS). -.2000b. Lesotho Agricultural Census 1999-2000. .2001 Lesotho Demographic Health Survey. -. 2002. Core Welfare inaicator survey. . - 2007.2002-03 and 1994-95 Household Budget Survey, Analytical Report. Vol. I, 207 Carter, M., R. and May, J. 1999. “One Kind of Freedom: Poverty Dynamics in Post-Apartheid South Africa.” University of Wisconsin-Madison Agricultural and Applied Economics Staff Paper 427. Colvin, Mark, Genieve Lemmon, and Julian Naidoo. 2006. “Private Sector Response to H N and AIDS in Lesotho.” Apparel Lesotho Alliance to Fight AIDS. May. Corno, Lucia, and Damien de Walque. 2007. “The Determinants of H N Infection and Related Sexual Behaviors: Evidence from Lesotho.” World Bank Policy Working Paper. December. Deaton , A., and V. Kozel. 2005. “Data and Dogma: The Great Indian Poverty Debate.” The World Bank Research Observer Deaton and Grosh in World Bank 2000a . “Designing Household Survey Questionnaires for Developing Countries: Lessons from 15 years of the Living Standard Measurement Study.” Eds. M. Grosh and P. Glewwe. De Lange, Izak 2008 Lesotho Markets Study. De Walque, D. 2006. “Who Gets AIDS and How? The Determinants of H N Infection and Sexual Behaviors in Burkina Faso, Cameroon, Ghana, Kenya and Tanzania.” World Bank Policy Research Working Paper 3844. . 2007. ”How Does the Impact of an HIV/ADS Information Campaign Vary with Educational Attainment? Evidence from Rural Uganda.” Journal of Development Economics 84: 6 86-7 14. D f D , 2000, Annexe 5, Making Markets Work for the Poor. A Framework Paper, DflD Economics and Policy Research Dept and Business Partnerships Development Dupas, Pascaline. 2006. “Relative Risks and the Market for Sex: Teenagers, Sugar Daddies and HIV in Kenya.” EHESS-PSE Dzimba, J., and M. Matooane. 2005. “The Impact of Stock Theft on Human Security: Strategies for Combating Stock Theft in Lesotho.” Chap. 7. http://www.issafiica.ordpu bs/Books/ManYFacesNovO5/Chap7.pdf DMA-WFP. 2006. Uncovering Chronic, Persistent Vulnerability to Hunger in the Southern Lowlands and Senqu River Valley. Report of the DMA-WFP Targeting exercise. Lesotho. (March). European Community 2007. Action Plan for the Development of a Joint Concept Note and Implementation Plan, European Community Support to Lesotho HJY and AIDS Response: Empowerment’of Orphans and Vulnerable Children. Draft. Prepared by PITT. April. FAO/WFP 2003 Crop and Food Supply Assessment Mission 2003. Federation of Women Lawyers (FDA). 2007. Training and Awareness Plans on the Legal Capacity of Married Persons, Act No. 9 of 2006. May 4. 208 Foster, V., and C. Araujo. 2004. “Does Infrastructure Reform Work for the Poor? A Case Study from Guatemala.” World Bank Policy Research Paper 3 185. December. Gakidou, Emmanuella, and Gary; King. 2006. “Death by Survey: Estimating Adult Mortality without Selection Bias from Sibling Survival Data.” Demography: 569-85. Gay, J and Hall. D 2002. Poverty and Livelihoods in Lesotho, 2000: More than a Mapping Exercise. Sechaba Consultants Gill-Wason, D. 2004. Evolving Livelihood Strategies of Rural Basotho: 1993-2002. LRAP Discussion Paper 5. CAFE Lesotho. Global Development Solutions, LLC 2004. Value Chain Analysis of Selected Strategic Sectors in Lesotho. Volume 2: Field Study. Global Fund HIV and AIDS Bi-Annual Progress Report, Round 2 Phase 11, Jan-June 2007, Global Fund TB Bi-Annual Progress Report Round 2 Phase 11, Jan-June 2007, Global Fund HIV and AIDS Bi-Annual Progress Report for Round 5 Global Fund Support in Lesotho. Reporting Period Nov 2006-April2007. The Global Fund to Fight AIDS, Tuberculosis, and Malaria Website: http://www.theglobalfund.org/search/docs/2LSOH~246~197~gsc.pdf. Lesotho, Government of (GoL). National Disaster Management Plan the Disaster Management Act of 1997. . 2004. “Guidelines to Prevent Mother to Child Transmission of HW.” June. www.health.gov.ls/docu1nents/Revised%2OPMTCT%2OGUIDEL~ES.pdf . 2005a Education Sector Strategic Plan, 2005-201 1, 113. . 2005b. Poverty Reduction Strategy Paper (PRSP) 2004/2005-2006/2007. Ministry of Finance and Development Planning. Maseru. . 2007a Country Report Lesotho; “Capacity Needs Assessment Report on Disaster Risk Management” Disaster Management Authority, Office of the Prime Minister, Maseru. .2007b National Action Plan for Capacity Development in Disaster Risk Reduction (2007- 15). Maseru. . 2007c. “Sustaining Treatment for AIDS in Lesotho.” Ministry of Health and Social Welfare. Maseru. Hua, H. 2007. “Report on Primary School Enrollment in Lesotho.” MoET. Maseru IMF 2005. Staff Report for the 2005 Article N Consultation. IMF. 2006. World Economic Outlook database. Washington, DC. 209 Inter-parliamentary Union. 2008. Parline database, available online at http://www.ipu.org/parline- e/parlinesearch.asp Itano, Nicola. 2005. “In Lesotho and Swaziland, AIDS Activates Women.” Women’s e News. www.womensenews.org/article.cfm/dyn/aid/2502/ 101 HIV Aids “Expert’ Patients Lighten Load for Clinic Staff.” 2006. October 18. http://www.iolhivaids.co.za/index.php?fArticleId=349303 0 Interview with key stakeholder from Global Fund. Interview with key stakeholder from MoHSW. Interview with key stakeholder from US government. Interview with key stakeholder from Government of Lesotho. Interview with key stakeholder at European Union. Key Indicators ofthe Labour Market Programme (KILM) database, compiled by the International Labour Organization (ILO), KoL (Kingdom of Lesotho). 2000a. Lesotho Agricultural Census, Ministry of AgricultureBureau of StatisticsMoF, Maseru. -. 2007 Lesotho Agricultural Situation Report, 198 1/82-2004/2005. Ministry of AgricultureBureau of StatisticsMinistry of Finance (MoF). Maseru, Lesotho. -.2007b Local Governance, Decentralization and Demand-Driven Service Delivery, Volume 1 :Main Report, World Bank/GTZ/FAO. Lawson, D. 2007. “A Gendered Analysis of ‘Time Poverty’: The Importance of Infrastructure.” Global Poverty Research Group Working Paper Series. www.g;prg.org/wbs Lesotho Demographic Health Survey. 2004. Leboela, T., and S.D. Turner. 2003. “The Voice of the People: Report on Community Consultations for the National Vision and the Poverty Reduction Strategy Paper.” Ministry of Development Planning and United Nations Development Programme, Maseru. Leduka, R.C. 2002. “Illegal Urban Settlements and the Transformation of Urban Property Rights in Lesotho.” Lesotho Social Science Review 8 (1 and 2). . 2004. “Informal Land Delivery Processes and Access to Land for the Poor in Maseru, Lesotho.” In “Informal Land Delivery Processes in African Cities.” Working Paper 5. International Development Department (IDD), School of Public Policy, University of Birmingham. www.idd.bhain.ac.uk/ . 2006. “Explaining Informal Land Delivery Processes and Institutions in African Cities: Conceptual Framework and Emerging Evidence.” South Afiican Review ofSocioZogy 3 7 (1). . 2007. “Recycled Fable or Immutable Truth? Reflections on the 1973 Land-Tenure Reform Project in Lesotho and Lessons for the Future.” Afiica Today 53 ( 3 ) . 210 Leduka, R.C., and J. Gay. 2007. “Lesotho Country Social Assessment: A Literature Review.” Processed. Maseru. Lesotho Demographic Health Survey. 2004. Lesotho National Development and Savings Bank Order No. 8 of 197 1 Lethola, R.S. 2005. “Land and Livelihoods in Southern Lesotho.” Repeat Study. “Livelihoods Recovery through Agriculture Programme.” Discussion Paper 8. CARE. Limiio N. and A. J. Venables. 2001 Infrastructure, Geographical Disadvantage, Tran sport Costs and Trade World Bank Economic Review, Vol 15, No 3. LVAC (Lesotho Vulnerability Assessment Committee). 2006a. “Lesotho Food Security and Vulnerability Monitoring Report, May-June 2006.” ,2006b. Lesotho Livelihoods Profile Report. (July) Magrath, P. 2005. “Changing Household Structures and Gender Relations in Lesotho.” LRAP (Livelihoods Recovery through Agriculture Programme) Discussion Paper 7. CARE Lesotho- South Africa, Maseru. Makhotla L.and S. Hendriks 2004. Do Home Gardens Improve the Nutrition of Rural Pre- schoolers in Lesotho? Development South Africa Vol21, No 3, pp 575 -581. Marsland, N., M. Metz, M. Phoofolo and M. Sejanamane, 2004. Food and Nutrition Insecurity in Lesotho: Problems, Trends & Responses. Matashane-Marite,K. 2005. Women’s Rights and Participation - Including Women’s Access to Land and Inheritance, and the Role of Lobbying and Grassroots Organizations in Lesotho. United Nations. Division for the Advancement of Women (DAW). EGM/WPD-EE/2005/EP .3 MCC (Millennium Challenge Corporation). 2006. Mission to Lesotho. (September). MoA/BS. 2005. Lesotho Agricultural Situation Report, 2000/01-2003/04, MOA and Food SecurityMnistry of Finance and Planning. Maseru’. Mokitimi, N. 2006. “Analysis of Security of Tenure under the Customary Land Tenure System of Lesotho.” Research Report 36. Institute of Southern African Studies. Lesotho. Medecins Sans Frontihes. 2007. “Help Wanted. Confronting the Healthcare Worker Crisis to Expand Access to HIV/AJDS Treatment: The MSF Experience in Southern Africa.” Johannesburg. May. MoHSW, 2008, Annual Joint Review Report 2007/2008 National AIDS Commission (NAC). 2006. National AIDS Policy. . National HIV and AIDS Strategic Plan (2006-201 1); 21 1 Okidi, J., and G. Mugambe. 2002. “An Overview of Chronic Poverty and Development Policy in Uganda.” CPRC (Chronic Poverty Research Centre) Working Paper 11. Open Society Initiative for Southern Africa. www.osisa.or9Jnode/5685 Partners in Health. httu://~~v.pili.or~/where/Lesotho/lesotho-org.litinl Pule and Thabane. 2004. Lesotho’s land tenure regimes: experiences of rural communities and the calls for land reform. Journal of Modern African Studies, 42,2. PSI website: http://www.psi.org/where we work/lesotho.html Rapid Appraisal for HIV/AIDS Program Expansion Lesotho, September 5-13, 2004, United State Agency for International Development (USAID) and Centers for Disease Control and Prevention (CDC). Quinlan, T., and M. Wallis. 2003. “Local Governance in Lesotho and the Central Role of Chiefs.” In D.I. Rsy and P.S. Reddy (eds) 2003. Grassroots Governance: Chiefs in Africa & the Afro- Caribbean, Calgary. University of Calgary Press. Ravallion, M. 1992. “Poverty Comparisons: A Guide to Concepts and Methods.” Living Standards Measurement Study (LSMS) Working Paper 88, World Bank. . 2004. “Pro-Poor Growth: A Primer” World Bank Policy Research Working Paper No. 3242. World Bank. Lopez, J.H and Serven,L. 2006 “A Normal Relationship? Poverty, Growth, and Inequality” World Bank Policy Research Working Paper 3814. The World Bank. Southern African Development Community (SADC). 2003. “Assessment and Review: HIV and AIDS Behaviour Change Communication Botswana, Namibia, Lesotho, Swaziland.” March. http://w.aidsportal.org/repos/SADC%20behaviour%20communication.pdf. Safilios-Rothschild. C 1985. “The Persistence of Women’s Invisibility in Agriculture: Theoretical and Policy Lessons from Lesotho and Sierra Leone.” Economic Development and Cultural Change 33 (2): 299-3 17. Scott, C., and B. Amenuvegbe. 1990. “Effect of Recall Duration on Reporting of Household Expenditures: An Experimental Study in Ghana.” Social Dimensions of Adjustment (SDA) Working Paper 8905. World Bank. Sentebale website: http://www.sentebale.ordhome/PreviousGrants.htmI Speech of Ambassador Rapolaki at OIC International Fifth Annual African Women’s and Children’s Health Symposium, University of District of Columbia, USA, Nov. 2004 http://w,Iesothoemb-usa.gov.Is/news/Speech Lesotho-Amb, Fiminization HIV AIDS.htm Statement of commitment by Lesotho’s Church Leaders on AIDS http://data.unaids.orcz/Dub/ExtemalDocument/2007/200707 18 lesothodeclaration en.pdf 212 Sustaining Treatment for AIDS in Lesotho, Ministry of Health and Social Welfare, Presentation by Dr Mphu K Ramatlapeng, Minister of Health and Social Welfare, Lesotho, October 2007. Turner, S. 2001. Livelihoods in Lesotho, CARE Lesotho. April 5. . 2003. The Southern African Food Crisis. Lesotho Literature Review. April 29. CARE Lesotho. ,2004. Livelihoods and Sharing: Trends in a Lesotho Village 1976-2004. CARE Lesotho. UN Office for the Coordination of Humanitarian Affairs. 2006. “Lesotho: Lack of Healthcare Workers a Drain on New HIV/AIDS Plan.” Apr. 27. http://wvw. irinnews.ora/reuort.aspx?reuortid=5885 8 . 2007. “Lesotho: New Plan to Reduce HIV Infections in Children.” April. 1 633 http://www,irinnews.org/Report.asUx?ReportId=7 United Nations. 2005. “Household Sample Surveys in Developing and Transition Countries.” Studies in Methods Series F No. 96. Department of Economic and Social Affairs, Statistics Division. UNAIDS, Lesotho 2005 UNGASS Country Report, Status of the National Response to the 2001 Declaration of Commitment on HIV/AIDS, 1/03-12/05, UNAIDS Lesotho 2008 UNGASS Country Report. Draft. UNAIDS website: http://www,unaids.or~en/CountrResuonses/Countries/lesotho.asp UNAIDS, Report on Global AIDS Epidemic, 2006. United Nations Children’s Fund Revised country programme document, Lesotho (2008-20 12), October, 2007: http://www.un icef.org/about/execboard/files/07-PL3 6-Lesotho(2).udf UNICEF 2005. Speaking Out ! Views of Young Basotho on Gender, Sexuality, HIV/AIDS, Life Skills and Education in Lesotho, Lesotho Youth Identity Study, Maseru UNICEF. 2006. “Lesotho: New Guidelines Set for Care of Orphans and Vulnerable Children.” March. UNICEF 2008 State of the World’s Children UNESCO 2008 Institute for Statistics, Lesotho Country Profile, available online at http://stats.uis.unesco.ora/unesco/TableViewer/document.as~x?Re~o~Id=l98&IFLannu age=eng. USAID Health Profile: Lesotho HIV/AIDS: http://www.synergyaids.co1~rofiles_Web~eJune2OO5 .pdf 213 US Government Rapid Appraisal for HIVIAIDS Program Expansion Lesotho September, 2004, United State Agency for International Development (USAID) and Centers for Disease Control and Prevention (CDC). US Government. 2008. PEFPAR-Funded Partners in Lesotho. January. Wason, D., and D. Hall. 2004. Poverty in Lesotho I993 to 2002. An Overview o f Household Economic Status and Government Policy. Chronic Poverty Research Centre (CPRC) Working Paper 40. Institute for Development Policv and Management, School of Environment and Development, University of Manchester, UK. http://www.chronicpoverty.orrr/contact/index.html William J. Clinton Foundation: htt~://~v.clintonfoi~ndation.orrr/searcli?~=cache:KGWzD~~v3uaY : http://www.clintonfoundation.org/cf-pgm-hs-ai- countries 1 1.htm+lesothoandsite=clintonfoundationandoutput=xml_no_dtdandclient=clintonfound ationandaccess=pandproxystylesheet=clintonfoundation World Bank. 2000a. “Designing Household Survey Questionnaires for Developing Countries: Lessons from 15 years of the Living Standard Measurement Study.” Eds. M. Grosh and P. Glewwe. . 2000b. Poverty Monitoring Database. http//wbln008 1.worldbank.org/dg/povertys.nsf/ World Bank 2004. Agricultural Policy and Capacity Building Project, ICR, ESSD, Southern Africa Department, The World Bank World Bank 2005a. Country Economic Memorandum, Growth and Employment Study, Report no. 26130-LSO. April 21. PREM I, Africa Region. World Bank 2005b. “Managing Agricultural Production Risks: Innovations in Development Countries. Agriculture and Rural Development (ARD). World Bank. 2007a. World Development Report 2007: Development and the Next Generation. World Bank 2007b, “Socio-Economic Differences in Health, Nutrition, and Population: Cross- Country Report,”. World Bank 2007c, Lesotho Mission Aide Memoire, Health and HN/AIDS Mission, October 29-November 10,2007 World Bank. 2008. “2005 International Comparison Program, Tables of Final Results.” WFP (World Food Programme)http://www.wfp.org/country brief/indexcountrv.asp?cou1it1~=426 WFP (World Food Programme). 2006. Lesotho Fact Sheet: Community and Household Surveillance (CHS), March WHO (World Health Organization). Know Your Status Campaign Plan. www,who.int/hiv/capacitv/icasa 1esotho.ppt . http://www.who.int/3 by5/newsitem6/en/ . www.who.int/ 214 . http://www.who. int/hiv/HIVCP LSO.pd f WHO (World Health Organization). 2007. World Health Statistics. -. 2007b New Data on Male Circuincision and HIV Prevention: Policv and Pronramine Implications. 215