LESOTHO POVERTY ASSESSMENT Progress and challenges in reducing poverty LESOTHO POVERTY ASSESSMENT: Progress and challenges in reducing poverty © 2019 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org This work is a product of the staff of The World Bank in close collaboration with staff of the Lesotho Bureau of Statistics. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and permissions The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. Photos: Shutterstock 2 Contents Boxes 4 Chapter I.2: The nature and evolution of inequality in Lesotho 60 Figures 5 A. The bottom of the population grew the fastest 60 Tables 9 B. Inequality fell but Lesotho remains an unequal Forewords 10 country 63 Acknowledgements 12 C. Inequality of asset ownership and income fell as well 64 Acronyms and abbreviations 13 D. Inequality of opportunity is high 68 Executive summary 15 Part II: Drivers and challenges of poverty Part I: Incidence, nature and evolution of reduction in Lesotho 71 poverty and inequality in Lesotho 24 Chapter II.1: Factors associated with poverty Chapter I.1: Incidence, nature and evolution of reduction 72 poverty in Lesotho 25 A. Drivers of Lesotho’s economic growth 72 A. Nationally, poverty has been significantly reduced but rural areas still lag behind 25 B. Inequality drove the decline in poverty 74 B. Who are the poor and where do they live? 30 C. Increases in formal wage incomes and expansion of social protection reduced C. High levels of economic vulnerability persist 39 inequality and poverty 77 Lesotho has made progress in reducing D.  Increase in educational attainment drove down D.  multidimensional poverty, but gaps still exist, poverty 80 especially in rural areas 44 E. Conclusion and policy discussion 81 3 Chapter II.2: The role of human capital on Part III: Vulnerability to shocks and poverty poverty reduction 82 reduction in Lesotho 124 A. Public spending on education and health is Chapter III.1: Vulnerability to economic and high but outcomes are relatively poor 82 environmental shocks 125 B. Poor education outcomes are associated A. Defining vulnerabilities in Lesotho 125 with high poverty in Lesotho 88 B. Basotho face severe environmental and C. Public spending on education and health is economic shocks 126 not fully pro-poor and affordability issues are large for poor households 92 C. Households have few coping strategies at their disposal to manage shocks 130 Education levels matter greatly for labor D.  market outcomes 96 The impact of climate-related shocks is D.  significant, especially among agricultural E. Conclusion and policy discussion 97 households 134 Chapter II.3: Labor market challenges for E. Existing social protection programs are limited poverty reduction in Lesotho 99 in the degree to which they help households mitigate the impact of shocks 136 A. Characteristics and challenges of Lesotho’s labor markets 99 F. Conclusion and policy discussion 137 B. Poor labor market outcomes are associated Chapter III.2: Social protection and poverty and with high poverty 105 inequality in Lesotho 140 C. Low productivity in subsistence agriculture A. Social protection programs in Lesotho 140 poses a challenge for poverty reduction 107 B. Public spending on social protection 143 Falling international migration and remittances D.  slow poverty reduction 116 C. Program coverage and efficiency 144 E. Conclusion and policy discussion 121 Impact of social protection on poverty and D.  inequality 150 E. Conclusion and policy discussion 151 References 152 Boxes Box 1: Consumption aggregation and poverty Box 4: Methods to find the households measurement in Lesotho 27 vulnerable to poverty 40 Box 2: Regions and other geographical identifiers Box 5: Construction of an asset index 49 in Lesotho 37 Box 6: Fill the Nutrient Gap Lesotho 52 Box 3: Small area estimation of poverty in Lesotho 39 Box 7: Estimating a multidimensional poverty index (MPI) for Lesotho 54 4 Box 8: Gender equality in Lesotho 66 Box 15: Labor laws and labor market institutions 104 Box 9: Inequality of opportunity and Lesotho Climate-Smart Agriculture Box 16:  intergenerational mobility 68 Investment Plan (CSAIP) 109 The role of inequality and growth in Box 10:  History and features of Lesotho Box 17:  poverty trends towards 2030 76 international migration 117 Demographic dividend opportunity for Box 11:  Measuring vulnerability to shocks in Box 18:  Lesotho 79 Lesotho 126 Box 12: Human Capital Index (HCI) for Lesotho 87 Strengthening the ability of social Box 19:  protection to help households manage Policy recommendations in education Box 13:  shocks 139 sector from the Public Expenditure Review 98 Box 20: Lesotho’s Social Protection Programs 141 Government prioritizes business Box 14:  development 101 Figures Figure 1: National poverty rate, 2002–2017 15 Poverty by household characteristics, Figure 12:  2002–2017 30 Figure 2: Regional comparison of poverty 15 Poverty by employment status of Figure 13:  Figure 3: Poverty rate by constituency 16 head of household, 2017 32 Figure 4: International inequality comparison 16 Poverty incidence by age group, Figure 14:  2002–2017 33 Figure 5: Poverty rate by educational attainment of household head 18 Age-gender pyramid and poverty, Figure 15:  2002–2017 33 Figure 6: Poverty rate by type of job 18 Distribution of the poor and population, Figure 16:  Figure 7: Poverty changes in absence of El Niño 20 urban-rural, 2002–2017 34 Figure 8: Relationship between poverty and Poverty incidence by region, Figure 17:  remittances at the constituency level 20 2002–2017 35 Figure 9: Impact of social protection programs Distribution of the poor and population Figure 18:  on inequality 21 across regions 36 Figure 10: Poverty incidence, 2002–2017 26 Mapping the regions used in the Figure 19:  analysis 37 International poverty rates (US$1.90 Figure 11:  in 2011 PPP terms per person per day) Figure 20: Poverty maps for Lesotho, 2017 38 in 2017, comparison to other lower middle-income countries 29 Figure 21: Identifying the vulnerable in Lesotho 41 5 The geographical location and Figure 22:  Consumption growth-incidence curves, Figure 38:  perceptions of the vulnerable, 2017 42 2002–2017 61 Sociodemographic characteristics by Figure 23:  Quintile shares and consumption Figure 39:  vulnerability and poverty status, 2017 43 relative to mean, 2002–2017 61 Income and job characteristics by Figure 24:  Shared prosperity in Lesotho, Figure 40:  poverty and vulnerability status, 2017 44 comparison to other countries 62 Changes in the proportion of the Figure 25:  Gini coefficient 2002 and 2017, Figure 41:  population with access to selected national and by district 63 basic services 45 Regional and global inequality Figure 42:  Changes in the proportion of the Figure 26:  comparisons 64 population with access to selected basic services, comparison to other Growth-incidence curves in income Figure 43:  countries 46 and assets 65 Figure 27:  Proportion of the population with Share of income inequality due to Figure 44:  access to selected basic services, different components 66 by region, 2017 47 Inequality of opportunity and its Figure 45:  Figure 28:  Proportion of the population with drivers 69 access to selected basic services, by decile, 2017 47 Figure 46: Intergenerational mobility in Lesotho 70 Figure 29:  Access to an improved roof, GDP and Private Consumption Index, Figure 47:  2002–2017 48 2002-2018 72 Figure 30:  Changes in household asset holdings, Growth decomposition, annualized, Figure 48:  2002–2017 49 2000-2017 72 Figure 31:  Poverty rate by access to basic Employment by major sector, Figure 49:  services, 2017 50 1990 – 2017, thousand people 73 Figure 32:  Multiple deprivations faced by the Growth and Value added in Lesotho by Figure 50:  poor, 2017 50 sectors, 1990-2017 73 Figure 33:  Food insecurity and malnutrition, Decomposing changes in poverty into Figure 51:  2000–2018 51 growth and distributional change 74 Share of food in total household Figure 34:  Figure 52: Poverty projections to 2030 76 consumption expenditure, 2017 52 Decomposing changes in poverty and Figure 53:  Contribution of weighted indicators to Figure 35:  inequality into income components 78 the MPI, geographic locations, 2017 57 Total dependency ratios for SACU Figure 54:  Figure 36:  Contribution of weighted indicators to countries 79 the MPI, dimensions and indicators 58 Decomposing change in poverty into Figure 55:  Figure 37:  Share of consistencies deprived in population-shift and intra-area effect, selected indicators 58 2002–2017 80 6 Decomposing change in poverty into Figure 56:  Rate of return by education level and Figure 75:  endowments and returns, 2002–2017 81 region, 2017 97 Government expenditures on Figure 57:  Labor force participation rate in South Figure 76:  education as percentage of GDP, 2018 83 African countries, 2002–2018 100 Government expenditures on health as Figure 58:  Unemployment rate in Southern Figure 77:  percentage of GDP, 2016 83 African countries, 2002–2018 100 Figure 59: SACMEQ III scores 84 Total factor productivity in Southern Figure 78:  African countries, 2002-2017 100 Figure 60: Net enrollment rates by level, 2016 85 Government spending as percent of Figure 79:  On-time, under-age and over-age in Figure 61:  GDP, 2007-2016 100 primary 85 Private/public sector source of Figure 80:  Human Capital Index, comparison to Figure 62:  household income by poverty status, other lower middle-income countries 86 2017 101 Human Capital Index components, Figure 63:  Labor force participation rate in Figure 81:  cross-countries comparison 87 Lesotho, 2002–2017 101 Education of household head by Figure 64:  Share of registered and non-registered Figure 82:  quintile, 2017 88 household businesses, 2002 and 2017 103 Poverty rate by educational attainment Figure 65:  of household head, 2017 88 Source of funding for businesses by Figure 83:  formal/informal, 2017 103 Education outcomes by poverty Figure 66:  status, 2017 89 Trends of the GDP share of the Figure 84:  services and agricultural sector, Primary repetition rates by grade, Figure 67:  2002–2017 103 2014 and 2015 89 Share of workers by type of Figure 85:  Education outcomes by poverty status, Figure 68:  employment, 2017 104 gender and region, 2017 90 Minimum monthly wages for employees Figure 86:  Figure 69: School dropout rates, 2017 91 with more than 12 months of service with the main employer 104 Government spending by level of Figure 70:  education and wealth quintiles, Bank credit to the private sector is Figure 87:  2015/16 92 lower in Lesotho than in comparable countries 104 Concentration curves by education Figure 71:  level, 2015/16 93 Poverty rates by employment status, Figure 88:  2002 and 2017 105 Government spending by gender and Figure 72:  wealth quintiles 93 Poverty rates by employment sector, Figure 89:  2002 and 2017 105 Figure 73: School dropout rates, 2017 94 Labor force composition by education Figure 90:  Figure 74: Rate of return by education level, 2017 97 level, 2002–2017 106 7 Education level by wage employment Figure 91:  Share of individuals in household Figure 110:  type, 2017 106 with remittances and poverty, by constituency 120 Figure 92: Education level by type of job, 2017 106 Share of population getting remittances Figure 111:  Figure 93: Poverty status by firm size, 2017 107 by quintiles of consumption 120 Figure 94: Poverty status by type of job, 2017 107 Figure 112: Remittances concentration curve 120 Share of population engaged in Figure 95:  Figure 113: Impact of remittances on poverty 121 farming activities, 2017 110 Figure 114: Impact of remittances on inequality 121 Share of households by type of parcel Figure 96:  cultivation, 2017 111 Frequency of shocks experienced by Figure 115:  household (last five years and in the Poverty status by type of parcel Figure 97:  last year) 127 cultivation, 2017 111 Rainfall was much lower during peak Figure 116:  Most households are engaged in more Figure 98:  pollination in 2016 127 than one agricultural activity, 2017 111 Figure 117: Inflation was high in 2016 128 Livestock breeding, and vegetable Figure 99:  cultivation are the most common Food insecure population by year, Figure 118:  agricultural activities, 2017 111 rural Lesotho 128 Share of market-oriented farmers, Figure 100:  Percentage of households affected Figure 119:  2017 112 by risk by urban/rural 129 Share of farming households that use Figure 101:  Percentage that occurred in the past Figure 120:  inputs in production, by quintile, 2017 113 five years by poverty status 129 Cereal yields in Lesotho and South Figure 102:  Proportion of households that lost Figure 121:  Africa, 1997-2017 113 income and/or assets as a result of a shock 130 Average farm sales and profits per Figure 103:  season 114 Price and weather risk are the most Figure 122:  covariate sources of risk households Share of farming households that Figure 104:  face 130 took loans for business use, 2017 115 It is households in the middle of the Figure 123:  Source of loan for farming households Figure 105:  wealth distribution whose that received a loan 115 consumption is affected most by external conditions 131 Figure 106: Average loan amount, by loan source 115 Households with education, migrants Figure 124:  Figure 107: Trends in remittances, ODA and FDI 119 and irrigation are less susceptible to rainfall losses 132 Share of individuals in household Figure 108:  with remittances from South Africa 119 How households manage shocks in Figure 125:  Lesotho 133 Average remittances received in Figure 109:  urban and rural areas, by quintile 120 8 Share of farming households exposed Figure 126:  Coverage of cash transfers, percent, Figure 133:  to shocks, by type of shock and 2017 146 poverty status, 2017 134 Transfer amount received by a group Figure 134:  Impact of rainfall shocks across types Figure 127:  as a share of the total consumption, of rural household (relative loss of 2017 146 consumption, 2017 135 Concentration curves for social Figure 135:  Poverty rates had rainfall been normal 135 Figure 128:  protection programs 148 The impact of a rainfall shock on Figure 129:  Transfer frequency in each population Figure 136:  households receiving different types group, percent 149 of public transfers 136 Simulated poverty reduction Figure 137:  Figure 130: Number of social protection programs 144 associated with social protection programs, percent 150 Spending on social protection Figure 131:  programs, percentage of GDP 144 Simulated reduction in inequality (Gini) Figure 138:  associated with social protection Top 5 social protection programs by Figure 132:  programs, percent 150 total expenditure, percentage of GDP 144 Tables Table 1: Mean household consumption per adult Table 8: Raw headcount rates, 2017 56 equivalent (real, January 2017 prices) 25 Table 9: The incidence and intensity of Table 2: The poverty lines per adult equivalent multidimensional poverty in Lesotho 56 per month, in LSL, current survey period prices 28 Table 10: Mean expenditures on health care and average cost of transportation to health Table 3: Depth and severity of poverty (percent care (LSL) 94 and percentage points), 2002–2017 28 Table 11: Share of health expenditure in total Table 4: Number of children, household size household expenditures during the last and dependency ratios 32 month (percentage of households) 95 Table 5: Depth and severity of poverty by region Table 12: Employment, unemployment and (percent and percentage points), yearly wage earnings, by educational 2002–2017 36 attainment, 2017 96 Table 6: Top 10 and bottom 10 constituencies Table 13: Coverage of social protection by poverty incidence, 2017 38 programs, by quintile and poverty status 145 Table 7: L-MPI: Dimensions, indicators, deprivation cut-offs and weights 54 9 Preface I am pleased to present the Lesotho Poverty expansion of social protection and an increase in wage Assessment: Progress and challenges in reducing incomes among the poor. Overall, Lesotho invests about poverty report. The report was produced by a World 4.5 percent of GDP annually on social assistance, and Bank team and the Lesotho Bureau of Statistics (BOS). I the programs reach a large share of the poor. However, wish to express my gratitude to BOS and particularly to although Lesotho is now more equal than its neighbors, thank the Director, Ms. Celina Molato for facilitating this with a Gini coefficient of 44.6, it remains one of the 20 work. I also wish to thank the Minister of Development percent most unequal countries. Planning, Hon. Tlohelang Aumane, for his guidance during the preparation of this report as well as the In addition, with more than a quarter of Basotho, 27.3 Principal Secretary, Ministry of Development Planning, percent, living on less than US$1.90/day (in 2011 PPP Ms. Nthoateng Lebona. terms) in 2017, a much higher number compared to other lower middle-income countries and among Southern The report reveals that much has been done to improve African Customs Union (SACU) countries, much more the lives of Basotho. About 47,000 people were lifted needs to be done to keep more Basotho out of poverty. out of poverty over a 15-year period between 2002 The report suggests that a combination of policies that and 2017, when Lesotho’s poverty rate fell by about 7 improve human capital, address high unemployment, percentage points, from 56.6 percent to 49.7 percent. increase agricultural productivity, along with those that Despite this progress, this study finds that economic build resilience against economic and environmental vulnerability is high, with more than 75 percent of shocks, would accelerate poverty reduction in Lesotho. the population either poor or vulnerable to poverty. It reveals the nature and face of poverty in Lesotho as It is my hope that this evidence-based analysis will mainly affecting people living in rural areas, with poverty enhance policy making in Lesotho and boost the levels being highest among female-headed households, country’s efforts to reduce poverty and inequality. This the less educated, the unemployed, large families, and is in line with the World Bank’s goals of reducing poverty children. and promoting share prosperity. The evidence in this report indicates that urban areas experienced greater poverty reduction following improvements in education and increases in incomes from well-paying jobs largely in the services sector. ............................................................... In rural areas, poverty stagnated due to slow growth MARIE FRANCOISE MARIE-NELLY in agricultural incomes, a fall in remittances and Country Director for Lesotho, vulnerability of the rural population to weather shocks. Botswana, Eswatini, Namibia, South Africa Furthermore, the study shows that despite the growing World Bank urban-rural poverty divide, inequality fell as a result of 10 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Foreword The Ministry of Development Planning through its Bank staff worked together to produce this poverty Department, Bureau of Statistics (BOS) is charged with assessment report. The report documents Lesotho’s the responsibility of collection, analysis, dissemination progress and challenges in reducing poverty with a focus and publication of quality official statistics. BOS is also on the period between 2002 and 2017. The assessment mandated by the Statistics Act of 2001 to coordinate is designed to support a dialogue by examining how the entire National Statistics System (NSS). Hence from the government can combat poverty and inequality time to time, the BOS undertakes Household Budget while safeguarding the poor against economic and Surveys, also known as income and expenditure surveys, environmental vulnerabilities. to provide up to date information to all stakeholders. The report also presents the drivers of poverty in BOS undertook the Household Budget Survey (HBS) both urban and rural areas through detailed analyses as a detailed module of the Continuous Multi-Purpose of the labor market, vulnerabilities to environmental Household Survey (CMS) in 2017/18. The main shocks and social protection. I therefore recommend objectives of the CMS/HBS were to provide detailed all government Ministries, development partners, household consumption and expenditure data required academia, researchers and civil society organisations to for (i) estimation and analysis of poverty and inequality, make use of this product in their individual and collective (ii) estimation of consumer price index weights and (iii) efforts to improve the livelihoods of Basotho. estimation of expenditure of private household accounts in the system of National Accounts (SNA). In addition, I take this opportunity on behalf of The Government of disaggregated data on gender and special groups Lesotho, to thank our development partners, particularly (women, youth, and people living with disability) is the World Bank, for their collaboration in this analysis required for monitoring and evaluation of development of poverty and inequality in the country as well as for agendas such as the 2030 Agenda for Sustainable providing continued technical and financial support. Development and the Africa Agenda 2063 in order to ensure the SDG motto of leaving no one behind. These data requirements resulted in the collaboration between the World Bank and the Ministry whereby the ............................................................... former provided both technical and financial support for HONORABLE TLOHELANG AUMANE undertaking of the 2017/18 CMS/HBS. BOS and World MINISTER OF DEVELOPMENT PLANNING Lesotho Poverty Assessment I Progress and challenges in reducing poverty 11 A Acknowledgements This report was prepared by Victor Sulla (EA1PV), Rasolonjatovo Andriamihamina (HAFE3), Nelly Rakoto- Precious Zikhali (EA1PV) and Daniel Gerszon Mahler Tiana (HAFE3) and Shilpa Challa (HAFE3), in the (EA1PV). The analyses were conducted in close health section from Christine Lao Pena (HAFH1), Pia collaboration with the Lesotho Bureau of Statistics. Schneider (HAFH1) and Omer Ramses Zang Sidjou The team would particularly like to thank Ms. Celina (HAFH2), in the section on vulnerabilities from Ruth Molato (Director) and Ms. Lerato Makana (Head of Hill (EA2PV), Zlatko Nikoloski (LSE) and Tao Tao National Accounts and Enterprise Surveys Division), (SURLN), in the section on multidimensional poverty who were instrumental in supporting and facilitating from Bianca Fisher and Eleni Yitbarek, in the social the work. The report benefited greatly from discussions protection section from Joachim Boko (HAFS2), with government officials, development partners, Oleksiy Ivaschenko (HSPGE), Dhushyanth Raju (HAFS3) and practitioners outside the government. The team especially gained from guidance by Hon. Tlohelang and Usama Zafar (HSPGE), in the migration section Aumane (Minister of Development Planning) and Ms. from Eriko Nishimura (IOM) and Nadege Desiree Nthoateng Lebona (Principal Secretary, Ministry of Yameogo (HSPJB), in the poverty map section from Development Planning). David Garces Urzainqui (LSE), in the gender section from Jana El-Horr (SAFS1) and Samaneh Hemat The report was undertaken under the guidance and (SAFS1), in the agricultural section from Ademola leadership of Marie Francoise Marie-Nelly (Country Braimoh (SAFA3) and Ana Reva (DECDB), in the labor Director, AFCS1), Paul Noumba Um (former Country section from Chiara Kofol (Economix), in the discussion Director, AFCS1), Janet K. Entwistle (Country of the main findings of the Fill the Nutrient Gap Lesotho Representative), Edouard Al-Dahdah (Senior Public summary report from Sabrina Kuri, Saskia Depee, Sector Specialist, EA1G2), Emmanuel Noubissie Janosch Klemm and Makhauta Mokhethi (World Food Ngankam (Country Program Coordinator), Aleksandra Organization). Richard Alm (independent consultant) Posarac (Program Leader), Erwin De Nys (Program edited the report. Leader), Pierella Paci (Practice Manager, EA1PV), and Andrew Dabalen (Practice Manager, EA2PV). It Relebohile Mohlakoana and Seitebatso Tsemane ably benefitted from feedback from peer reviewers Ambar provided logistical assistance in the preparation of this Narayan (Lead Economist, EPVGE), Javier E. Baez report. Communications support was provided by Elita (Senior Economist, EA1PV) and Nistha Sinha (Senior Economist, EMNPV). Banda and Zandile Ratshitanga. Last but not least, the team would like to thank everyone at the Lesotho Bureau The following people provided inputs to the report: of Statistics and the World Bank who contributed to In the education section from Harisoa Danielle making this a truly collaborative effort. Thank you. 12 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Acronyms and abbreviations CaLP Cash Learning Partnership LSL Lesotho Maloti CCT Conditional cash transfer LVAC Lesotho Vulnerability Assessment Committee CGP Child Grant Programs MUAC Middle upper-arm circumference CMS/HBS Continuous Multipurpose Household Survey and Household Budget Survey NHSP National Health Strategic Plan CPI Consumer Price Index NISSA National Information System for Social Assistance DMA Disaster Management Authority NMDS National Manpower Development ECCD Early Childhood Care and Development Secretariat FAO Food and Agricultural Organization NSDP National Strategic Development Plan FDI Foreign Direct Investment OAP Old Age Pension GDP Gross Domestic Product ODA Official Development Assistance GHI Global Hunger Index ODI Overseas Development Institute GoL Government of Lesotho OPM Oxford Policy Management HBS Household Budget Survey OVC Orphans and Vulnerable Children HCI Human Capital Index PAP Public Assistance Program ICTs Information and Communication PPP Purchasing Power Parity Technologies SACU Southern African Customs Union IFPRI International Food Policy Research Institute SADC Southern African Development Community ILO International Labor Organization SADP Smallholder Agriculture Development IMF International Monetary Fund Project IWM Integrated Watershed Management SDGs Sustainable Development Goals Public Works SMEs Small and Medium Enterprises LCMPA Legal Capacity of Married Persons Act SSA Sub-Saharan Africa LDHS Lesotho Demographic and Health Survey TFP Total Factor Productivity LFP Labor force participation UNICEF United Nations Children’s Fund LMICs Lower Middle-Income Countries WFP World Food Programme LNDC Lesotho National Development Corporation Lesotho Poverty Assessment I Progress and challenges in reducing poverty 13 E 14 Executive summary Poverty has fallen over the past 15 years but stagnated in rural areas, adding to an already large urban-rural divide Lesotho’s poverty rate fell from 56.6 percent in Although this is low when compared with other 2002 to 49.7 percent in 2017. Lesotho’s poverty countries in Sub-Saharan Africa, it is high among other rate, measured at the national poverty line of Lesotho lower middle-income countries and among Southern Maloti (LSL) 648.88 (2017 prices) per adult equivalent African Customs Union (SACU) countries. Within SACU, per month, fell about 7 percentage points over a 15- only Eswatini had a comparable international poverty year period (Figure 1). This translates to about 47,000 rate. The corresponding rates were 18.9 percent for Basotho escaping poverty during this period. In 2017, South Africa, 16.1 percent for Botswana and 13.4 for 27.3 percent of Basotho were poor at the international Namibia (Figure 2). poverty line of US$1.90/day (in 2011 PPP terms). Figure 1: National poverty rate, 2002–2017 Figure 2: Regional comparison of poverty 70 61.3 60.7 80 60 56.6 International Poverty Rate (%) Poverty headcount rate (%) 49.7 50 60 41.5 40 28.5 40 30 Le so th 20 o 20 10 0 0 Lesotho Urban Rural Sub−Saharan African countries ordered from lowest to highest international poverty rate 2002 2017 Source: The 2017/18 Continuous Multipurpose Survey and Household Budget Survey (CMS/HBS), the 2002/03 Household Budget Survey (HBS) and PovcalNet. Urban areas recorded strong poverty reduction, 10.9 percentage points from 56.9 percent in 2002. Due while rural areas’ poverty levels stagnated, adding to increased urbanization, the number of rural poor to an already large urban-rural divide. The urban decreased from 864,000 to 801,000, while the number poverty rate decreased from 41.5 percent to 28.5 of urban poor increased from 180,000 to 196,000. This percent. In rural areas, muted consumption growth is equivalent to the share of urban poor increasing from was accompanied by stagnation in poverty rates, which 17 percent to 20 percent, suggesting that poverty will decreased marginally from 61.3 percent to 60.7 percent. continue to be a rural phenomenon in the near term. A This has resulted in a widening gap between rural and profile of Lesotho’s poor shows that poverty levels are urban poverty. Poverty fell in four out of six regions. highest among female-headed households, especially The two regions that experienced a poverty increase those headed by single women, the less educated, the are both in rural areas – Rural Mountains and Rural unemployed and the caretakers of large families and Senqu River Valley. In 2017, 67.8 percent of Basotho children. living in Rural Mountains were poor, an increase of Lesotho Poverty Assessment I Progress and challenges in reducing poverty 15 The modest decline in the national poverty rate from 29.0 percent to 21.9 percent, with a decline of masks a notable decline in extreme poverty and 8.3 in urban areas and 4.3 in rural areas. This occurred inequality. Extreme poverty, measured based on a because consumption growth between 2002 and 2017 food basket required to achieve the minimum daily was inclusive for the very poorest segments of the calorie requirement of 2,700 kilocalories (kcal) per population, leading to a reduction in the Gini coefficient adult equivalent per day, declined from 34.1 percent to of more than 7 points. Although Lesotho is now more 24.1 percent. It halved in urban areas, going from 22.2 equal than its neighbors, with a Gini coefficient of 44.6, percent to 11.2 percent, and declined in rural areas from it remains one of the 20 percent most unequal countries 37.7 percent to 30.8 percent. The poverty gap declined in the world (Figure 4). Figure 3: Poverty rate by constituency Figure 4: International inequality comparison 60 Lesotho, 2002 Maseru Pover ty rate (%) Lesotho, 2017 Gini coefficient <20 40 20−30 30−40 40−50 50−60 20 60−70 >70 0 Countries ordered from lowest to highest Gini Source: Calculations based on the 2016 Population and Housing Census, the 2017/18 CMS/HBS, the 2002/03 HBS and PovcalNet. Non-monetary indicators show some progress peers. The services sector has been the most resilient in reducing poverty, but significant gaps remain, contributor to economic growth. The contribution of especially in rural areas. Access to basic public agriculture, the sector that supports most of the poor, services has improved, but it remains unevenly remained volatile. The volatility is due in part to climate distributed across regions, with the spatial pattern of variability. The textile industry, another important sector access closely following the urban-rural divide. Rural for the livelihoods of the poor, has recently stagnated. regions tend to have smaller shares of people with Furthermore, the economic growth of the past decade access to basic services. Furthermore, access to basic had little impact on job creation, and high unemployment services is unevenly distributed between the poor and rates characterize Lesotho’s labor markets. Since 2015, non-poor. The gaps manifest in a high incidence of the economy has not grown in per capita terms due multidimensional poverty which, like monetary poverty, to challenges from political instability, environmental is higher in rural areas. The poor are simultaneously shocks and a prolonged period of slow growth in South deprived in multiple dimensions, and this reinforces and Africa that diminished SACU revenues. South Africa’s perpetuates poverty, particularly in rural areas. poor economic outlook in the near-to-medium term is expected to have a negative impact on Lesotho’s growth The persistence of poverty in Lesotho is strong trajectory. Lesotho, therefore, finds itself in need of despite solid economic growth in the past 15 years. new sources of sustainable and inclusive growth, with Lesotho’s per capita real gross domestic product a dynamic private sector that creates jobs and helps (GDP) grew at an average annual rate of 2.7 percent the country seize opportunities in regional and global in 2000–2017, faster than the country’s regional markets. 16 Lesotho Poverty Assessment I Progress and challenges in reducing poverty The reduction in urban poverty was driven by education and skills development and formal wage jobs The role of secondary and higher education fraction of GDP, Lesotho spends more on education than most countries with similar levels of income. Secondary and higher education are the main Free access to primary education was implemented in paths out of poverty. Globally, education and skills are 2000, yet primary net enrollment has remained almost reliable ways to escape poverty and this is even more stagnant at around 80 percent since 2002. Among the so in the case in Lesotho. Among individuals living in Southern and Eastern Africa Consortium for Monitoring households headed by someone who did not complete Educational Quality (SEACMEQ) countries, Lesotho had primary education, 61.3 percent were poor in 2017. the lowest net primary enrollment in 2017 at 81 percent, This number fell to 24.4 percent when the household compared to 97 percent in Namibia. Public education head had completed secondary education and 8.4 spending in secondary and higher education favors the percent when the household head had a post-secondary rich and is unequal. education (Figure 5). More than two-thirds of the poor lived in households in which the household head had Access to better paying, more stable and no education. Hence, a sustainable path out of poverty higher productivity jobs, supported poverty comes from obtaining secondary education. reduction From 2002 to 2017, urban areas saw larger increases Education tends to lift people out of poverty than rural areas in individuals with secondary because it gives individuals access to better jobs education or higher education. In 2017, more than half that provide a reliable source of income. The returns of adults in urban areas – 52 percent – had completed to education, especially post-secondary, are high, secondary education or more. This is up from 33 percent increasing the probability of being in the labor force in 2002. Rural areas also saw an increase in the share and out of unemployment. In both 2002 and 2017, the of adults with secondary education or more, but the labor force participation rate was highest among those increase was smaller – from 9 percent to 19 percent. with technical/vocational training or university or higher Increases in educational attainment can explain 2.3 education. percentage points of the reduction in poverty in urban areas, but they contributed to less than 1 percentage Skilled non-agricultural jobs provided the main point of poverty reduction in rural areas. sources of income to lift urban households out of poverty. In urban areas, the poverty rate declined by Rural poverty was less impacted by increases in 2.2 percentage points due to increases in formal wage educational attainment due to poorer educational income and 5.1 percentage points because of increases outcomes. Poor children, particularly in rural areas, face in self-employment income. Whether individuals are high repetition and dropout rates at the primary level. out of the labor market or in the labor market matters Those who make it to the end of primary school often less for poverty than the type of jobs individuals hold. face large impediments to transitioning to secondary In particular, the duration of work contracts is strongly education due to access issues and the high costs of correlated with individuals’ poverty status. Many secondary education. Access to post-primary education enterprises operate only seasonally, and rural enterprises remains a challenge for children from poor families, show this intermittent pattern more frequently than particularly in the Rural Mountains and Rural Senqu urban enterprises. Workers with seasonal contracts are River Valley regions. the poorest – 61 percent were below the poverty line in 2017 – followed by those with casual contracts and Despite the contribution of education and skills temporary jobs. The poverty rate is lowest among those to poverty reduction, Lesotho’s human capital with permanent jobs. Less than 30 percent of individuals outcomes are low considering its spending. As a with permanent jobs were poor (Figure 6). Lesotho Poverty Assessment I Progress and challenges in reducing poverty 17 Figure 5: Poverty rate by educational Figure 6: Poverty rate by type of job attainment of household head 70% 100 61% Share of population (%) 60% 80 39 49% 54 48 50% 71 69 60 Poverty rate 40% 34 40 26 30% 24% 24 20 17 18 20% 22 26 27 8% 12 13 0 10% Permanent Seasonal Contract Casual Temporary 0% ry y e y ar da on ar nd im on N co ec Pr Se -s st Extreme poor Poor Non-poor Po Source: The 2017/18 CMS/HBS. Despite the positive impact of higher income on urban and rural areas and the relatively small difference poverty reduction, the economic growth of the past in poverty rates among working and non-working 15 years had limited impact on job creation. As a individuals can be attributed to a larger share of the result, unemployment is high, estimated at 28.0 percent rural population being employed in low productivity overall and 43.2 percent among youths aged 15 to 24. subsistence agriculture and/or low-paying agricultural The unemployment challenge is exacerbated by a small jobs. In contrast, urban areas are characterized by a private sector. Its competitiveness has been limited larger share of the employed population in wage-earning by skills mismatches and lack of entrepreneurship jobs outside agriculture, including the well-paid public programs for growth-oriented businesses. In recent sector, where workers earn an average monthly salary years, Lesotho’s business climate has improved, but about seven times higher than the minimum monthly the regulations governing business and access to wage of a skilled worker in the textile sector. finance remain obstacles. The informal sector is large, with lack of access to credit and household savings Employment outcomes are also characterized by acting as barriers to formality. The weak private sector uneven gender outcomes. Women tend to have both gives the government a central role in providing formal better education and health outcomes in Lesotho, but employment. Although the public sector is one of the this has not translated into the labor market. Women main drivers of growth, it does not employ many of the have a high representation in government and the poor. In addition, the government’s contribution to GDP government has introduced policy and legislative changes is highly dependent on the cycle of fiscal policies. supporting gender equality; even so, the effectiveness of these measures in bringing about positive changes Labor productivity is low in Lesotho, especially in in the lives of women has so far been limited. Deeply rural areas and sectors with a high participation of entrenched social norms and stereotypes negatively the poor. In general, low labor productivity is associated impacting women are dominant, and the new laws have with low wages. Low productivity is especially a been stymied by limited awareness, low capacity and challenge in sectors where informality is high, including inconsistency in application. Opportunities for women agriculture, hunting and forestry, wholesale and retail to participate on equal footing in the economic life of the trade, hotels and restaurants and community, social and country remain limited. personal activities. The large gap in poverty between 18 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Rural poverty stagnated in part due to the vulnerability of the rural population to weather shocks and falling remittances Vulnerability to weather shocks Lesotho endured historically low rainfall in 2015- 2016; without it, rural poverty would have been 6 Agriculture plays a major role in Basotho percentage points lower, and the pace of national employment and livelihoods. About 71 percent of poverty reduction would have nearly doubled. In the population is involved in some form of agricultural 2015-2016, Lesotho faced one of its greatest rainfall activity. The sector is dominated by smallholder shortages in decades due to an El Niño. Rural areas subsistence production with reliance on rainfed, low responded to the shock by reducing consumption by input/low output production methods characterized an average of 23 percent. As a result, rural poverty by limited use of irrigation, improved seed, fertilizers was significantly higher than it would have been in the and pesticides. This contributes to low yields and absence of the weather shock. It is estimated that rural subsequently to widespread poverty in rural areas. In poverty would have fallen to 54.6 percent with normal Lesotho, 78 percent of those in the lowest quintile rainfall. Urban poverty, on the other hand, was less participate in agricultural activities, compared to 58 impacted by the rainfall shortage, and the poverty rate percent of the wealthiest quintile. The average farm size would only have declined 3 percentage points faster is 3.5 hectares (five hectares for non-poor households had no shock occurred. Aside from urban households and two hectares for the poor households). Most farming relying less on agriculture, the Metolong dam, which households produce for their own consumption, and was finished just before the drought, also helped supply only 3 percent produce primarily for sale. Most farming critical water to Maseru and other urban areas. In total, households engage in several agricultural activities with poverty in Lesotho would have decreased twice as fast very little specialization; livestock and cereal production over the 15 years had the shock not occurred (Figure 7). are the most common. Livestock farming is the most prevalent agricultural activity in Lesotho, where three- Except for pensions, existing social protection fourth of households raise animals. programs are limited in helping households mitigate the consumption impacts of shocks. Food social Climate change and environmental shocks are protection transfers reduced but did not eliminate the among the key challenges in the agriculture sector, impact of rainfall shocks on beneficiaries. This was also especially the prospect of frequent droughts and true for those who received help from NGOs. Receipt heavy seasonal floods. In rural areas, the higher of a pension eliminated a household’s vulnerability to prevalence of shocks with large impacts on welfare, rainfall risk. On average, rural households receiving exposes households to higher risk levels. When shocks pension support are richer than other rural households. occur, households have few coping mechanisms, These resources combine with constant and sizeable particularly for shocks that affect many households in pension income in reducing the impact of variable the same community or market at the same time. In agricultural income on consumption. many instances, most households do not take specific actions to manage the shocks, and poor households Those with access to education and irrigation are are less likely than others to act to manage shocks. less susceptible to risks. Consumption in households As a result, consumption is often reduced in response with no education drops 38 percent in the face of a bad to shocks, increasing poverty and leading to long-run shock, compared to 11 percent for households headed consequences for children’s human capital attainment. by someone with a secondary education or above. When households do employ coping strategies, most Having education makes households more resilient rely on help from family and friends. because they can diversify their sources of income in the face of deteriorating agricultural production conditions. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 19 Figure 7: Poverty changes in absence of El Figure 8: Relationship between poverty and Niño remittances at the constituency level 61.3 60.7 60 56.6 80 54.6 49.7 Poverty rate (%) 50 60 Poverty rate (%) 44.6 41.5 40 40 28.5 30 20 25.3 0 20 2002 2017 0 10 20 30 40 50 Share of individuals in household with remittances (%) Rural, actual Rural, with normal rainfall Lesotho, actual Lesotho, with normal rainfall Mountainous constituencies Urban, actual Urban, wtih normal rainfall Rural non−mountainous constituencies Urban non−mountainous constituencies Source: Calculations based on the 2016 Population and Housing Census, the 2017/18 CMS/HBS, and the 2002/03 HBS. Falling remittances Lesotho has long depended on remittances from Their geographical isolation makes it more challenging mining jobs in South Africa, but employment in to work in South Africa. This has important implications the industry has recently declined. The main reasons for the well-being of Basotho living in these areas. for decreasing remittances are mechanization and Remittances serve as a buffer when shocks occur; stagnation in South Africa’s gold mining industry and without access to remittances, this region is likely to the country’s shift toward the use of local labor. Yet, react stronger to shocks. The 2015-16 drought mostly close to 15 percent of Lesotho’s population still works impacted the lowlands and foothills. Had the drought in South Africa. Their remittances remain important to been equally severe in the mountainous region, then the economy, representing 15 percent of GDP. this region would lag even more behind than is already the case. Rural poverty would have been nearly 10 percentage points lower had remittances not declined since Non-poor rural households are vulnerable to falling 2002. The transfers are strongly progressive, with back into poverty. The high reliance of rural households significant impacts on poverty and inequality. Poor on agricultural income and remittances makes their households use the money to buy necessities and livelihoods volatile, and a shock increases the risk that invest in human capital. Urban households are less rural households that have escaped poverty may fall reliant on remittances, and urban poverty would have back. For a quarter of Basotho, the predicted probability been only 2.7 percentage points lower if remittances of being in poverty is sufficiently high to make them had remained at the 2002 levels. vulnerable to falling back into poverty. This means these households share characteristics with others that Poor households, especially the less educated and are poor and under different shock scenarios may find rural residents, are particularly likely to be dependent themselves characterized as poor. This is particularly upon remittances. Among the 30 constituencies most true among rural households, where three of four non- reliant on remittances from South Africa, 25 are rural poor households face high risks of falling into poverty. In and all have poverty rates above 40 percent (Figure contrast, only 30 percent of non-poor urban households 8). Yet, some of the very poorest constituencies in the are vulnerability to poverty. mountainous areas rely relatively little on remittances. 20 Lesotho Poverty Assessment I Progress and challenges in reducing poverty The fall in inequality was associated with an expansion of social protection and wage income The growth in urban areas combined with poverty annually on social assistance, and the programs reach stagnation in already-lagging rural areas would suggest a large share of the poor. The Gini coefficient would increased inequality over the 15-year period. Yet be 7.8 percent higher without these programs (Figure Lesotho’s Gini coefficient fell by 7 points from 2002 9), and the expansions of the programs that occurred and 2017, supported by strong growth rates at the very between 2002 and 2017 led to a 2.6-point decline in bottom of the distribution in both urban and rural areas. the Gini coefficient. Cash transfers were particularly Although this growth did not pull many rural households effective in reducing inequality. Social assistance is also out of poverty, it did increase their welfare and reduce important for reducing the poverty gap, suggesting that inequality. government programs helped boost the welfare of the poorest segments of the population. However, leakages Expansion of social protection contributed persist, and the impact of social assistance on poverty significantly to inequality reduction could be strengthened by improved targeting efficiency. In the absence of a comprehensive national social The comprehensive social protection system security scheme, the burden to provide certain benefits contributed significantly to inequality reduction. falls fully on employers, resulting in low compliance and Overall, Lesotho invests about 4.5 percent of GDP a small proportion of the population receiving benefits. Figure 9: Impact of social protection programs on inequality 10 Impact on Gini coefficient(percent) 7.8 8 6 5.1 4 1.8 1.5 2 0.8 0.2 0 All social In-kind Labor market Cash Social Education protection subsidies programs trans fers insurance subsidies Source: The 2017/18 CMS/HBS. Increased wage income also played a points lower had rural households not suffered from dominant role in inequality reduction poor rainfalls prior to the survey. Increased wage income also played a dominant role Although inequality fell significantly over the past in inequality reduction, while a fall in remittances 15 years, high levels of inequality remain due to the and stagnation in agricultural incomes slowed urban-rural divide, public-private wage gaps and inequality reduction. Increased wage income led to a discrepancies in educational outcomes. A higher 3.1-point decline in the Gini coefficient. The reduction in education is a strong predictor of escaping poverty, but inequality would have been larger if remittances had not at the same time it is also an inequality generating factor. fallen and agricultural incomes had not been stagnant. Not all individuals are able to complete quality primary The Gini coefficient would have been an estimated 3 education and proceed to secondary education, and Lesotho Poverty Assessment I Progress and challenges in reducing poverty 21 this discrepancy causes divergences in wage incomes Sub-Saharan Africa with available estimates. This is later in life. In particular, those who manage to get a closely linked to a low intergenerational mobility that government job earn significantly higher salaries than continues to pose a challenge for reducing poverty those working in the private sector, adding to inequality. and inequality. For example, parents’ level of education In general, more than half of the inequality in income generally predicts their children’s level of education, can be attributed to differences in wage income. and it remains difficult for people born in the bottom to climb to the top and vice versa. Inequality continues to Half of the current level of inequality in Lesotho can manifest in disparities in access to basic public services be attributed to factors beyond individuals’ control, across income groups as well as geographic locations. such as place of birth, basic educational attainment, These disparities limit the ability of poor households to health and environmental shocks. This is a relatively take advantage of economic opportunities. high number – higher than the 10 other countries in Policies that improve human capital, address high unemployment, increase agricultural productivity, along with those that build resilience against economic and environmental shocks, would accelerate poverty reduction Improving human capital health-care infrastructure and soft skills development in rural regions by making sure appropriate equipment, Given that universal and free primary education health staff, medicines and other inputs are available in has largely been attained, it is important to boost facilities. Improving health outcomes in Lesotho should learning in primary education, increase secondary also involve investing in the first 1,000 days of life enrollment and completion among the poor and from conception. This should include well-coordinated improve the general efficiency of spending on multisectoral investments on nutrition-specific and education. Investmenting in Early Childhood Care sensitive interventions. and Development (ECCD) is a cost-effective strategy for reducing inequities and substantially improving Addressing high unemployment long-term human development outcomes for adults. Children who benefit from quality ECCD perform Strengthening the private sector through demand- better in primary school, repeat grades less and drop and supply-side reforms is key to creating jobs. On out less. Therefore, giving the poor the opportunity to the demand side, policies aimed at easing the process access quality ECCD will improve not only the quality of receiving business licenses, accessing finance and of education but also the cost efficiency of the system. lowering regulatory compliance costs would improve To achieve quality in primary schools, it is important to the business environment. Pursuing deeper regional improve the foundational skills of reading, writing and integration and trade dialogue with South Africa, the arithmetic. This will require improving teaching quality United States and the European Union would also be and teacher training. At the secondary level, expansion important for growing Lesotho’s private sector. Adoption of the junior secondary school network is important for of digital technology should be further prioritized in making these schools more accessible to the poor. In Lesotho. On the supply side, targeted policies that addition to putting in place the appropriate physical and boost entrepreneurship and promote skills development soft infrastructure, increasing equity of access to junior would increase the supply of skilled labor, boost labor secondary education requires that the financial cost of productivity and reduce poverty. attending such schools be reduced as much as feasible. It is important to improve migration governance Setting up a pro-poor health financing program by developing a national strategy with a capacity could help improve the health system’s equity to prevent irregular migration, promote legal outcomes. Such a program could focus on alleviating migration and mobility and ultimately enhance indirect expenses like transportation to and from synergies between migration and development. The seeking care. Special attention is needed to improve bilateral labor agreement between Lesotho and South 22 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Africa should be enhanced with the goal of facilitating knowledge and expertise and provide technical advice labor mobility and improving migration data collection, to farmers. Critical areas that need capacity development management and analysis to support evidence-based include identifying funding gaps and needs, assessing policymaking. Furthermore, reducing remittance costs, public and private financing options, developing possibly through financial inclusion and development, payment for ecosystem services programs, developing would be beneficial to the poor. Reducing barriers would bankable investment plans, project pipelines and encourage new entrants in the remittances market, and financing propositions and developing financially viable the competition would likely drive down the cost of opportunities for effective private sector engagement. sending money across borders. Building resilience against shocks through Increasing productivity in agriculture responsive social protection Improvements in agricultural productivity can be The most cost-effective way to reduce vulnerability achieved through transitioning from subsistence to poverty is to reduce exposure to shocks. to commercial agriculture, increased use of Sustainable interventions that are important for productivity-enhancing agricultural inputs and risk mitigation include land management practices, strengthening linkages between farmers and protection of forests and woodlands and improved buyers. Commercialization can be prioritized largely in management of water resources. Investments in lowlands and foothills, while the highlands would benefit irrigation and education both have the potential to do from resilient landscape, or afforestation, and farmer- that. Irrigation reduces the impact of low rainfall on managed natural regeneration to restore and replenish agricultural output, and education increases the ability less fertile land. This entails diversification from cereals of individuals to earn income in other, less-rainfall to higher-value crops, such as fruits and vegetables, that dependent sectors, either before or in response to have the potential to increase incomes and create jobs. the shock. It is also important to increase the ability of Cultivation of vegetables and fruit has good potential households to manage the risks that remain through in Lesotho because of a favorable climate, strong local better financial market development that can spread demand and opportunities for import substitution. risk beyond the immediate social network and through Agricultural commercialization will involve building social protection. Agricultural insurance could protect linkages between farmers and buyers and supporting farmers against disasters by transferring their risks local agro-dealers/extension services. Furthermore, it to the credit markets. It could also help in increasing will be necessary to train agri-entrepreneurs in business farmers’ agricultural productivity and access to financial skills, record keeping, marketing as well as on use of services. The government could look at policy options to inputs and agronomic practices. support the expansion of agricultural insurance. Investments in Climate Smart Agriculture (CSA) Social protection policies are another way offers the potential to transform Lesotho’s governments can help households manage risks. agriculture into a more productive, climate-resilient This could be accomplished by providing households a and low-emissions sector. The effective scaling up of dependable source of income that is not subject to risk CSA in Lesotho will require addressing several adoption and by scaling up benefits to provide more support in barriers, including limited implementation capacity, hard times. Being ready to scale up requires an early insufficient access to inputs and credits and insufficient warning system that provides accurate information agricultural research. There is a need to strengthen on when and where scale-up is required and triggers research and establish partnerships with international that can be used to develop clear transparent rules for research institutes to develop high-yielding, stress- scaling up support to households. It also requires pre- tolerant and climate-ready varieties. Agricultural financing arrangements to finance the added benefits. extension services should be upgraded to catalyze Scaling up is easier for existing beneficiaries already the agricultural innovation process, improve the CSA in social protection systems, so it would necessitate knowledge system, facilitate access to information, having all potential beneficiaries in a national database. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 23 I Part I: Incidence, nature and evolution of poverty and inequality in Lesotho 24 Chapter I.1: Incidence, nature and evolution of poverty in Lesotho The proportion of the population living below the national poverty line decreased from 56.6 percent in 2002 to 49.7 percent in 2017. Urban areas experienced a 13 percentage point reduction in poverty from 41.5 percent to 28.5 percent, while poverty stagnated in rural areas, decreasing marginally from 61.3 percent to 60.7 percent. As a result, the gap between rural and urban poverty further widened. Consistent with this, non-monetary poverty indicators suggest that progress has been made but gaps still exist, especially in rural areas. Poverty levels are consistently highest among female-headed household, the less educated, the unemployed, large families and children. Despite progress in reducing poverty, high levels of economic vulnerability persist. More than a quarter of the country’s population (27.7 percent) is vulnerable to falling into poverty. Only 22.6 percent of households are not vulnerable to poverty, and they are concentrated in urban areas, where half of all households have sustainably escaped poverty. A. Nationally, poverty has been significantly consumption: while consumption grew by 5.3 percent reduced but rural areas still lag behind in urban areas, rural households registered a decline of 7.3 percent. The largest contraction (34.8 percent) Overall consumption growth between 2002 and was recorded in Rural Senqu River Valley, followed by 2017 was not broad-based: in rural areas, it was Rural Mountains at 28 percent and Rural Foothills at 9.2 muted. Table 1 shows that at national level, mean percent. This could be due to the influence of the 2015- household consumption per adult equivalent, the welfare 2016 El Nino drought that, among other challenges, indicator used in this Poverty Assessment (see Box 1), resulted in crop failures and acute food security, increased by 4.5 percent in real terms between 2002 particularly in rural areas where subsistence farming is and 2017. However, not all areas experienced growth in the main source of livelihood. Table 1: Mean household consumption per adult equivalent (real, January 2017 prices) 2002 2017 Percentage change Urban 1,253.9 1,320.9 5.3 Rural 759.7 704.3 -7.3 Region Maseru Urban 1,547.4 1,422.1 -8.1 Other Urban 1,073.3 1,245.3 16.0 Rural Lowlands 713.7 795.1 11.4 Rural Foothills 716.2 650.3 -9.2 Rural Mountains 838.4 603.4 -28.0 Rural Senqu River Valley 929.2 605.8 -34.8 Lesotho 875.8 915.0 4.5 Source: Calculations based on the 2002/03 Household Budget Survey (HBS) and the 2017/18 Continuous Multipurpose Household Survey and Household Budget Survey (CMS/HBS). Lesotho Poverty Assessment I Progress and challenges in reducing poverty 25 Household consumption growth meant solid The strong poverty reduction in urban areas did poverty reduction in urban areas, but poverty levels not translate into an equally strong reduction in the remained high in rural areas.1 In urban areas, the absolute number of poor. This suggests that poverty headcount poverty rate decreased strongly from 41.5 reduction in urban areas is barely keeping up with to 28.5 percent, a 13 percentage point reduction (Figure demographic shifts. Comparing the absolute number 10). In rural areas, poverty nearly stagnated, decreasing of poor in 2002 and in 2017 reveals an increase in the marginally from 61.3 to 60.7 percent. The relatively number of poor by about 16,000 in urban areas (Figure larger decline in urban areas resulted in a widening 10). In rural areas, the absolute number of poor fell by of the gap between rural and urban poverty. At the 63,000, and this drove the 47,000 reduction in number national level, the share of the population living below of poor at the national level. the national poverty line fell from 56.6 percent in 2002 to 49.7 percent in 2017, a statistically significant decline of about 7 percentage points. Figure 10: Poverty incidence, 2002–2017 (a) Poverty headcount rate (b) Number of poor 1.200 70 1.044 Number of poor, thousands 61.3 997 60.7 56.6 1.000 Poverty headcount rate (%) 60 864 49.7 801 50 800 41.5 37.7 628 40 34.1 600 532 30.8 28.5 484 30 24.1 22.2 407 400 20 11.2 180 196 10 200 96 77 0 0 Urban Rural Lesotho Urban Rural Lesotho Urban Rural Lesotho Urban Rural Lesotho Food poverty line National poverty line Food poverty line National poverty line 2002 2017 2002 2017 Source: Calculations based on the 2002/03 HBS and the 2017/18 CMS/HBS. 1 The report uses the food poverty line (FPL) and the upper bound poverty line (UBPL), considered the absolute national poverty line, to measure poverty in Lesotho. The focus is on these two poverty lines because the lower bound poverty line (LBPL) was not computed in 2002/03, making it impossible to conduct a trend analysis. In 2017, 44.7 percent of the population lived below the LBPL and poverty rates were 24.0 percent in urban areas and 55.5 percent in rural areas. 26 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Box 1: Consumption aggregation and poverty measurement in Lesotho The main data sources for this Poverty Assessment are the 2002/03 Household Budget Survey (HBS), the 2017/18 Continuous Multipurpose Household Survey and the Household Budget Survey (CMS/HBS). In this report, 2002 refers to the 2002/03 survey year and 2017 to the 2017/18 survey year. These surveys collect detailed information on household spending and consumption as well as other indicators of wellbeing, including income, education, health, access to basic services and ownership of assets and employment. A consumption-based welfare indicator is used to measure poverty, referred to in this report as the consumption aggregate. The aggregate is constructed following guidelines provided in Deaton and Zaidi (2002). All food expenditures and self-produced food items valued at local market prices are included in the aggregate. Consumption of non-food items includes expenditures on personal care and hygiene items, clothing, utilities, transportation and other items. Notably, the following non-food items were excluded from the consumption aggregate: expenditures on durable goods, actual and imputed housing rents, expenditures on ceremonies such as weddings and funerals and hospitalization costs. Consumption from durable goods is excluded because the CMS/HBS does not estimate an asset’s annual flow of value; including the value of these lumpy purchases would distort the consumption aggregate. This approach is consistent with the way the consumption aggregate was constructed in previous poverty analyses. Imputed rents for homeowners and actual rents for rent payers were also excluded to create comparable statistics. Two adjustments are made to the consumption aggregate. The first is for household size and composition by dividing the consumption aggregate by the officially used per adult equivalent scales, which are based on calorie requirements that vary by age and sex. The second is the use of spatial and temporal deflators to account for price variation across time and space. In the CMS/HBS, deflators were used to convert households’ nominal consumption during different survey quarters and in different locations (urban/rural). These adjustments ensure the consumption aggregate is comparable across space and time. The consumption aggregates used in this Poverty Assessment are the official ones used by the Bureau of Statistics. Lesotho uses the cost-of-basic-needs (CBN) method to determine its consumption-based poverty line. It is based on a food basket required to achieve the minimum daily calorie requirement—2,700 kilocalories (kcal) per adult equivalent per day—and adjusted upward to include non-food consumption. The food poverty line (FPL), considered the extreme poverty line, is defined as the level of consumption per adult equivalent that individuals need to purchase enough food for an adequate diet. It is determined in two stages. First, a food reference basket is constructed using households from the second to fifth deciles of consumption per adult equivalent as a reference population. Second, the basket is costed to determine the level of the FPL. The estimation of the absolute poverty lines—the lower bound poverty line (LBPL) and the upper bound poverty lines (UBPL)—builds on the FPL by including an allowance for non-food consumption following the upper- and lower-bound methods of Ravallion (1998). The share of food in total consumption expenditures is computed for lower- and upper-bound reference groups and then respective non-food expenditures are added by dividing the food poverty line by the respective food shares to estimate the total lower- or upper-bound poverty line. The reference group for constructing the LBPL are households with total expenditures close to the FPL. These households sacrifice some of their basic food requirements to meet their non-food needs. The UBPL, on the other hand, uses households with food expenditure close to the food poverty line as the reference group. These households can purchase both adequate food and non-food items at the UBPL. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 27 Table 2 presents the estimated poverty lines for 2002/03 and 2017/18. Table 2: The poverty lines per adult equivalent per month, in LSL, current survey period prices Poverty Line 2002/03 2017/18 Food poverty line 84.41 352.39 Lower-bound poverty line Not estimated 572.41 Upper-bound poverty line 149.91 648.88 Source: Lesotho Bureau of Statistics (forthcoming). Not only do Basotho living in rural areas face in urban areas (6.1 percentage points) than rural areas higher risks of being in poverty, but their poverty (5.2 percentage points).3 The food or extreme poverty also tends to be more intense and severe than their line4 shows that rural areas experienced a faster decline counterparts in urban areas. The national poverty line than urban areas in both the depth and severity of indicates the depth and severity of poverty has been poverty. Rural areas recorded a faster decline in the falling faster in urban than in rural areas (Table 3). The food (extreme) poverty rate than in the national poverty poverty gap narrowed by 8.3 percentage points in rate, meaning that consumption growth in rural areas urban areas and 4.3 percentage points in rural areas.2 was able to pull a substantial number of people out of Overall, the declining trend indicates that the minimum extreme poverty and bring them closer to the national cost of eliminating poverty – i.e., closing the gap – fell poverty line. Overall, both the depth and severity of during this period. Similarly, severity of poverty eased poverty presented in Table 3 confirm that rural poverty in both urban and rural areas, with a higher reduction is much higher than urban poverty. Table 3: Depth and severity of poverty (percent and percentage points), 2002–2017   Poverty gap Squared poverty gap Change Change   2002 2017 2002 2017 (2017-2002) (2017-2002) Food poverty line Urban 9.1 3.2 -5.8 5.0 1.4 -3.6 Rural 17.5 10.6 -7.0 10.7 5.0 -5.7 Lesotho 15.5 8.1 -7.5 9.4 3.8 -5.6 National poverty line Urban 19.2 10.8 -8.3 11.6 5.6 -6.1 Rural 32.0 27.7 -4.3 20.9 15.7 -5.2 Lesotho 29.0 21.9 -7.1 18.7 12.2 -6.5 Source: Calculations based on the 2002/03 HBS and the 2017/18 CMS/HBS. 2 The poverty gap provides information on the portion of the poverty line that people require, on average, to escape poverty. It is expressed here as a percentage of the poverty line. 3 The severity of poverty is measured by the squared poverty gap, a weighted sum of poverty gaps that gives more weight to those individuals or households who fall well below the poverty line. 4 Note that the national food poverty line is interchangeably referred to as the national extreme poverty line. 28 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Despite the progress, poverty levels are relatively high for the country’s income level Lesotho’s poverty levels are relatively high for of Basotho were poor at the international US$1.90 per a lower middle-income country. Compared to person per day poverty line (in 2011 PPP terms). Poverty other countries with similar consumption levels and rates remain higher in Lesotho than several other lower initial poverty rates of about 20 percent, the pace of middle-income countries (Figure 11). Among countries in Lesotho’s poverty reduction in the past 15 years could the Southern African Customs Union (SACU), Lesotho’s be considered moderate and slower than expected. As poverty rate is comparable only to that of Eswatini. The a result, poverty levels remain relatively high for the international poverty rates were 18.9 percent for South country’s income levels. In 2017, about 27.3 percent Africa, 16.1 percent for Botswana and 13.4 for Namibia. Figure 11: International poverty rates (US$1.90 in 2011 PPP terms per person per day), comparison to other lower middle-income countries (a) Annualized change in poverty (b) Comparison to other middle-income countries 70 Annualized change in poverty rate (pct. points) 60 Poverty headcount rate (%) 4 50 40 2 30 27.3 18.9 20 16.1 13.4 0 10 Lesotho 0 Botswana Solomon Islands Honduras West Bank and Gaza Micronesia, Federated States of Bangladesh Myanmar Pakistan Morocco Bhutan Cote d'Ivoire South Africa Philippines Mauritania Nicaragua Egypt, Arab Republic of Cameroon Mongolia Moldova Georgia Ghana Vietnam Lesotho Zambia Namibia Kyrgyz Republic Ukraine El Salvador Sri Lanka Kenya Tunisia Djibouti Bolivia −2 −4 0 20 40 60 80 Initial poverty rate (%) Source: For Lesotho: Calculations based on the 2002/03 HBS and the 2017/18 CMS/HBS. For the rest of the countries: PovcalNet. Note: Values are the most recent available over the past five years (2013 to 2017). Lesotho Poverty Assessment I Progress and challenges in reducing poverty 29 B. Who are the poor and where do they live? Who are the poor? This section presents a profile of the poor in terms of the demographic characteristics of households and individuals at the national poverty line. Figure 12 shows the poverty headcount rate as well as the distribution of the poor by each characteristic. Figure 12: Poverty by household characteristics, 2002–2017 (a) Poverty headcount rate 80 71.6 Poverty headcount rate (%) 70 64.7 66.6 67.1 65.2 59.6 63.1 59.8 60.9 62.161.3 65.1 61.3 62.2 61.6 62.6 58.0 56.3 57.0 57.2 58.4 60 55.3 55.2 54.3 55.0 57.2 55.7 51.9 52.4 54.1 52.7 48.9 49.9 51.0 48.6 50 46.3 46.0 46.1 45.2 45.7 44.8 41.4 41.7 40.7 42.5 40 35.8 38.1 35.5 32.9 30 26.7 28.3 26.4 24.4 23.2 20 17.1 10 8.4 0 1 and more Married/Living together Post-secondary Single female None 0.75 to 1.00 Separated/Divorced 0.00 to 0.25 0.50 to 0.75 Secondary 1 2 1 2 3 4 5 6 0.25 to 0.50 Male 7 or more Other Widow/Widower no children Female 3 or more children Primary Never married Gender of household Marital status of Single Education of Number of children Household size Household dependency ratio head household head female household head head 2002 2017 (b) Distribution of the poor 80 70.1 71.0 68.8 70 68.6 62.2 60.4 58.5 60 Distribution of the poor (%) 53.6 53.3 48.9 50 43.1 41.5 39.6 37.2 40 36.6 36.6 33.4 36.0 31.2 35.1 30.1 29.9 30 22.7 22.2 21.9 23.0 21.3 16.7 18.9 18.9 20 15.3 13.7 15.0 14.7 10.9 12.1 12.0 8.1 9.2 8.0 7.3 9.0 10 5.3 5.0 2.4 4.2 4.6 7.1 2.6 4.1 6.3 6.8 1.8 0.91.5 1.0 0 1 2 1 2 3 4 5 6 Male None Other Single female Primary Never married Widow/Widower 1 and more 7 or more Female No children Secondary Separated/Divorced 0.75 to 1.00 Married/Living together 0.00 to 0.25 0.25 to 0.50 0.50 to 0.75 Post-secondary 3 or more children Gender of Marital status of Single Education of Number of children Household size Household dependency ratio household household head female household head head head 2002 2017 Source: Calculations based on the 2002/03 HBS and the 2017/18 CMS/HBS. 30 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Individuals living in female-headed households are households with a head who had no formal education more likely to be poor than those living in male- in 2017. In general, poverty rates decline with education headed households. In 2017, the poverty rates were levels. Further, the pace of poverty reduction increases 55.2 percent for female-headed households and 46.3 with education levels, highlighting the role of education percent for male-headed households. Between 2002 as a socioeconomic equalizer. and 2017, the rate for male-headed households fell 9.0 percentage points, compared to 4.4 percentage Basotho living in households with children are more points for female-headed households. As a result, the likely to be poor than those living in households gap widened between individuals in female- and male- without children. In 2017, the poverty headcount headed households. However, 58.5 percent of the poor rate was 30.2 percentage points higher for individuals lived in male-headed households in 2017. living in households with at least three children than for individuals living in households with no children. The Living in a household headed by a single female gap between these two groups widened considerably is associated with an increased likelihood of being from 8.9 percentage points in 2002 to 30.2 percentage poor. In 2017, the poverty rate among individuals living points in 2017. in a household headed by a single female was 57.2 percent, 11.4 percentage points higher than the rate for The incidence of poverty is higher in larger families. individuals living in households headed by either a single In 2017, the poverty headcount rate among individuals male or married head. The gap between the poverty living in one-person households was 17.1 percent, rates for these two groups widened between 2002 and compared to 67.1 percent for individuals living in 2017. households with at least seven members. The latter made up the largest share of Lesotho’s households Widows and widowers exhibit the highest rates (26.6 percent). Despite a decline in their share of total of poverty. In 2017, 59.8 percent of people living in a population, individuals living in households with at household headed by a widow or widower were poor, least six household members registered an increase in compared to 35.8 percent of those living in households poverty levels since 2002. Overall, adding members to with heads who never married. Since 2002, households a household progressively increases the probability of headed by a widow or widower had the smallest decline being poor. However, concluding that poverty increases in poverty. Individuals living in households with a divorced with household size should be done with caution, taking head had almost the same poverty rate (46.1 percent) as into consideration economies of size in household those living in households headed by someone married consumption. or living with a partner (46.0 percent). The distribution of the poor by the head’s marital status shows that most of Poverty increases with household dependency the poor (53.6 percent) lived in households with heads ratios. In Lesotho, 42.3 percent of the population lived that were either married or living with a partner. These in households with a dependency ratio higher than 1 in individuals made up the highest share in total population 2017.5 Individuals living in households with a dependency (57.9 percent) in 2017. ratio of less than 0.25 had the lowest poverty rate (26.4 percent), while individuals living in households with a Educational attainment strongly correlates with dependency ratio between 0.75 and 1 had a headcount poverty in Lesotho. In 2017, 61.3 percent of individuals rate of 65.2 percent. Between 2002 and 2017, poverty living in a household headed by someone with no formal reduction was highest among individuals living in education were poor, down slightly from 62.1 percent households with the lowest dependency ratios, while in 2002. This was about seven times the poverty rate dependency ratios above 0.75 were associated with among individuals living in a household in which the increasing poverty levels. head had completed tertiary education. In households with a head with no formal education, 68.6 percent The poor tend to live in households that are larger were poor. At 55.6 percent, most Basotho lived in and have more children. In 2017, an average poor 5 The dependency ratio is the number of dependents younger than 15 or older than 64 divided by the number of household members of working age. A low dependency ratio means the household has an adequate number of members of working age to support the dependents that in the household. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 31 person lived in a household that had about 1.5 more person lived in a household with 2.8 dependents for members than non-poor households (Table 4). The every working-age adult, compared to 2.0 for the non- larger household sizes for the poor are mainly driven poor. This means the strain on household resources is by children. As a result, dependency ratios are far significantly higher for poor households. Trends were higher for poor households. Nationally, an average poor similar in both urban and rural areas. Table 4: Number of children, household size and dependency ratios Urban Rural Lesotho Non-poor Poor Non-poor Poor Non-poor Poor 2002 Number of children, 0-6 years 0.6 0.7 1.0 1.1 0.9 1.0 Number of children, 0-14 years 1.6 2.2 2.3 2.6 2.1 2.5 Household size 5.1 6.1 5.9 6.7 5.6 6.6 Dependency ratio 2.0 2.3 2.6 2.6 2.4 2.6 2017 Number of children, 0-6 years 0.6 0.9 0.7 1.0 0.7 1.0 Number of children, 0-14 years 1.3 2.3 1.6 2.5 1.4 2.5 Household size 4.0 5.5 5.0 6.1 4.5 6.0 Dependency ratio 1.9 2.8 2.2 2.8 2.0 2.8 Source: Calculations based on the 2002/03 HBS and the 2017/18 CMS/HBS. Employment is associated with lower poverty unemployed heads (Figure 13). Considering the industry rates, but poverty rates are relatively high even of employment, individuals living in households with a among the employed. In 2017, the poverty rate was head employed in the agriculture sector had the highest 40.1 percent among individuals living in households poverty rate (62.1 percent). It was 33.9 percent for headed by an employed person, 22.4 percentage points industry and 33.4 percent for services. This highlights lower than the rate (62.5 percent) for households with the heightened poverty risk for agricultural households. Figure 13: Poverty by employment status of head of household, 2017 Poverty headcount rate (%) 70 62.5 62.1 60 50 42.4 40.1 40 33.9 33.4 30 20 10 0 Not employed Services Employed Construction Industry Agriculture Employment Industry of employment status Source: Calculations based on the 2017/18 CMS/HBS. 32 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Children are particularly prone to poverty. In 2017, Children aged 0-5 had a headcount rate of 53.2 percent. children aged 6-14 had a poverty rate of 60.9 percent, These data cause concern because child poverty has a 2.2 percentage point increase from 58.8 percent in life-long implications. Poverty is also high among the 2002 (Figure 14). The age cohort made up the largest elderly: 52.0 percent of those aged above 65 lived below share of Lesotho’s poor both in 2002 and 2017. The 15- the poverty line in 2017. 19 age cohort registered a poverty rate of 55.0 percent. Figure 14: Poverty incidence by age group, 2002–2017 70 100% 6.1 7.3 60.2 2.4 65+ 58.2 58.8 60.9 90% 2.7 2.5 Poverty headcount rate (%) 60 57.0 57.6 57.6 2.7 55.0 55.3 55.2 56.3 3.6 3.2 60-64 53.2 52.3 51.0 52.4 52.0 3.8 3.3 48.9 80% 4.0 4.2 55-59 Distribution of the poor (%) 50 45.8 45.6 4.5 5.3 43.3 43.9 41.7 40.7 43.7 44.5 70% 5.4 50-54 6.2 37.9 40 7.3 6.5 45-49 60% 11.0 8.4 40-44 30 50% 11.3 35-39 12.2 20 40% 30-34 30% 25-29 10 24.4 26.1 20-24 20% 0 15-19 65+ 15-19 25-29 40-44 55-59 20-24 30-34 35-39 45-49 50-54 60-64 6-14 0-5 10% 12.6 12.9 6-14 0% 0-5 2002 2017 2002 2017 Source: Calculations based on the 2002/03 HBS and the 2017/18 CMS/HBS. The gender age pyramid shows a population that third of the population in both 2002 and 2017. At 21.3 is young and with higher poverty rates than adults. percent in 2017, children aged 6-14 accounted for the (Figure 15). Individuals aged 0-14 made up around a highest share of total population. Figure 15: Age-gender pyramid and poverty, 2002–2017 2002 2017 90-95 90-95 80-85 80-85 70-75 70-75 Age in years Age in years 60-65 60-65 50-55 50-55 40-45 40-45 30-35 30-35 20-25 20-25 10-15 10-15 0-5 0-5 8 6 4 2 0 2 4 6 8 8 6 4 2 0 2 4 6 8 Share in total population (%) Share in total population (%) Poor females Poor males Females Males Poor females Poor males Females Males Source: Calculations based on the 2002/03 HBS and the 2017/18 CMS/HBS. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 33 Where do the poor live? Most Basotho poor live in rural areas. The rural share poverty reduction in urban areas than in rural areas. As of the population fell 10.7 percentage points from 76.5 Figure 16b shows, urban areas have higher consumption in 2002 to 65.8 percent in 2017, but the rural share of than rural areas. The data suggest that improvements in poverty decreased only 2.4 percentage points from 82.2 rural areas and/or increasing mobility to urban areas are to 80.4 percent (Figure 16a). This reflects the stronger central to future poverty reduction. Figure 16: Distribution of the poor and population, urban-rural, 2002–2017 (a) Distribution of the poor (b) Welfare distribution, 2017 100% 90% .001 National poverty line 80% 70% 65.8 60% 82.8 80.4 76.5 Density 50% .0005 40% 30% 20% 34.2 10% 17.2 19.6 23.5 0% 2002 2017 2002 2017 0 0 1000 2000 3000 4000 5000 Distribution of the Poor Distribution of Population Monthly expenditure (per adult equivalent) Urban Rural Urban population Rural population Source: Calculations based on the 2002/03 HBS and the 2017/18 CMS/HBS. The nation’s decline in poverty was not equally Mountains region were poor at the national poverty line, distributed across the six regions. Poverty fell in four a 10.9 percentage point increase from 56.9 percent in out of the six regions (see Box 2 for details on how 2002. In the Rural Senqu River Valley, 67.9 percent of these regions are defined). The two that experienced the population was poor in 2017, up from 55.5 percent increases in poverty are both rural areas – Rural in 2002. Although still disappointingly high, the poverty Mountains and Rural Senqu River Valley (Figure 17). rate was lowest in Maseru Urban at 24.7 percent. In 2017, 67.8 percent of Basotho living in the Rural 34 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Figure 17: Poverty incidence by region, 2002–2017 80 70 Poverty headcount rate (%) 60 50 40 30 20 10 0 Rural Rural Maseru Other Rural Rural Rural Senqu Maseru Other Rural Rural Rural Senqu Urban Urban Lowlands Foothills Mountains River Urban Urban Lowlands Foothills Mountains River Valley Valley Food poverty line National poverty line 2002 17.6 25.0 38.2 43.8 33.7 34.2 33.7 46.3 62.4 66.8 56.9 55.5 2017 9.7 12.4 25.8 33.1 37.2 35.2 24.7 31.4 54.4 63.6 67.8 67.9 Source: Calculations based on the 2002/03 HBS and the 2017/18 CMS/HBS. Not only did poverty rise in the Rural Mountains depth and severity of poverty, with the depth of poverty and Rural Senqu River Valley regions, the situation based on the national poverty line actually increasing in of the poor in these regions also worsened between both regions. The poor in Maseru Urban had the least 2002 and 2017. Table 5 shows that the two regions depth and severity of poverty in the country. lagged the rest of the country in terms of reducing the Lesotho Poverty Assessment I Progress and challenges in reducing poverty 35 Table 5: Depth and severity of poverty by region (percent and percentage points), 2002–2017   Poverty gap Squared poverty gap Change (2017- Change (2017-   2002 2017 2002 2017 2002) 2002) Food poverty line Maseru Urban 6.8 2.5 -4.4 3.7 1.0 -2.7 Other Urban 10.4 3.8 -6.6 5.8 1.8 -4.1 Rural Lowlands 17.7 8.4 -9.3 10.7 3.9 -6.9 Rural Foothills 20.0 10.8 -9.3 12.2 5.3 -6.9 Rural Mountains 15.8 13.3 -2.5 9.8 6.3 -3.5 Rural Senqu River Valley 16.8 13.5 -3.3 10.2 6.8 -3.4 National poverty line Maseru Urban 15.0 9.5 -5.5 8.9 4.7 -4.2 Other Urban 21.7 11.9 -9.9 13.3 6.2 -7.1 Rural Lowlands 32.5 23.7 -8.8 21.2 13.1 -8.1 Rural Foothills 36.0 29.1 -6.9 23.8 16.5 -7.3 Rural Mountains 29.1 32.4 3.3 19.0 18.9 0.0 Rural Senqu River Valley 29.3 32.3 3.0 19.5 19.0 -0.5 Source: Calculations based on the 2002/03 HBS and the 2017/18 CMS/HBS. Rural Lowlands accounts for the highest share of 2002. The distribution of the poor did not change much the poor. The region’s poverty rates were 42.9 percent between 2002 and 2017. Rural Lowlands consistently in 2002 and 35.3 percent in 2017 (Figure 18). This is accounted for the biggest share of the country’s poor, partly due to the region’s relatively high population followed by Rural Mountains and then Rural Foothills. share – 32.2 percent in 2017, down from 38.9 percent in Figure 18: Distribution of the poor and population across regions 100% 4.3 4.4 10.0 7.3 90% 20.4 20.3 16.6 80% 22.7 70% 12.8 9.7 15.1 60% 12.5 50% 32.2 38.9 40% 42.9 35.3 30% 19.6 20% 14.5 11.9 12.3 10% 14.6 5.3 7.3 9.0 0% 2002 2017 2002 2017 Distribution of the poor Distribution of population Maseru Urban Other Urban Rural Lowlands Rural Foothills Rural Mountains Rural Senqu River Valley Source: Calculations based on the 2002/03 HBS and the 2017/18 CMS/HBS. 36 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Box 2: Regions and other geographical identifiers in Lesotho This report’s main geographical division uses the six regions of Lesotho – Maseru Urban, Other Urban, Rural Lowlands, Rural Foothills, Rural Mountains and Rural Senqu River Valley. The latter four contain the four different agro-ecological zones of Lesotho after parsing out urban areas. Peri-urban areas are kept in the rural regions. These six regions represent the lowest level at which the 2017/18 CMS/HBS survey is representative. The different characteristics of the six regions make them interesting units of analysis on such topics as infrastructure access, the impact of weather shocks and proximity to South Africa. These six regions have been used in prior surveys in Lesotho; for example, Housing and Population Censuses, Labor Force Surveys and the Demographic and Health Survey. Figure 19 maps each household interviewed in the 2017/18 CMS/HBS by region, as well as the location of the four agro-ecological zones. Figure 19: Mapping the regions used in the analysis Maseru Region Maseru Urban Other Urban Rural L owlands Rural F oothills Rural Mountains Rural Senqu River Va lley International Boundary ‘Lowlands’ Zone ‘Foothills’ Zone ‘Mountains’ Zone ‘Sequ River Valley’ Zone Agro Ecological Zone Name Source: Lesotho Ministry of Public Works and Transport (right figure). Note: Each dot of the left figure represents the location of a household surveyed in the 2017/18 CMS/HBS. Ten districts make up Lesotho’s first administrative level. They are divided into 78 community councils. Lesotho also has 80 constituencies, used for electoral purposes. In addition to the six main regions, these subdivisions will be used in this report when more detailed geographical analyses are needed. A national poverty map demonstrates heterogeneity poverty rates at the constituency level. This means across space, with the country’s 10 districts having areas closer to the border with South Africa tended pockets of both wealth and extreme poverty. For to exhibit lower poverty rates. This could be capturing instance, seven of the 10 constituencies with the accessibility to economic opportunities in South Africa lowest poverty rates were in the Maseru district, which as well as accessibility to better-developed markets. also had two of the poorest 10 constituencies. Figure 20 Such access is likely to be associated with higher flows reveals a very interesting pattern of poverty in Lesotho: of remittances from Basotho living and/or working in in 2017, peripheral areas were associated with reduced South Africa. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 37 Table 6: Top 10 and bottom 10 constituencies by poverty incidence, 2017 Poverty Poverty headcount Constituency District headcount rate (%) Constituency District rate (%) Top 10 Bottom 10 Hloahloeng Mohale’s Hoek 78.3 Maseru Maseru 15.0 Semena Thaba-Tseka 75.6 Khubetsoana Berea 18.2 Maliepetsane Mafeteng 73.8 Abia Maseru 19.3 Ketane Mohale’s Hoek 73.4 Mabote Berea 21.5 Senqu Mokhotlong 73.2 Stadium Area Maseru 22.1 Thaba-moea Thaba-Tseka 72.2 Hlotse Leribe 23.9 Mosalemane Berea 71.8 Lithoteng Maseru 25.4 Thaba-putsoa Maseru 71.0 Thetsane Maseru 26.2 Makhaleng Maseru 70.9 Maama Maseru 27.2 Mashai Thaba-Tseka 69.7 Lithabaneng Maseru 27.5 Source: Calculations based on the 2016 Population and Housing Census and the 2017/18 CMS/HBS. Figure 20: Poverty maps for Lesotho, 2017 (a) Poverty headcount rate, by district (b) Poverty headcount rate, by constituency Butha−Buthe Leribe Maseru Berea Pover ty rate (%) Mokhotlong <20 Poverty rate (%) 20−30 <40 Maseru 30−40 Thaba−Tseka 40−50 40−50 50−60 Mafeteng 50−60 >60 Qacha’s Nek 60−70 Mohale’ s Hoek >70 Quthing Source: Calculations based on the 2016 Population and Housing Census and the 2017/18 CMS/HBS. 38 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Box 3: Small area estimation of poverty in Lesotho Unmasking poverty’s heterogeneities and patterns across Lesotho’s subnational levels requires a methodology that maps poverty across space—small area estimation. It relies on the detailed analysis of two main data sources: a household survey and a population census. In Lesotho, these are found in the 2017/18 Continuous Multipurpose Household and Household Budget Survey (CMS/HBS) and the 2016 Population and Housing Census. The CMS/HBS contains detailed modules on consumption expenditures. Due to the relatively small sample size, however, the available information only covers six regions in the country. By contrast, census data are available for all households and provide exact information on the distribution of demographic or other characteristics at highly disaggregated levels, such as districts and constituencies. However, it does not include the detailed information on consumption or income required to produce reliable indicators of the level and distribution of welfare. To overcome these shortcomings, the Elbers, Lanjouw and Leite (2008) methodology combines the strengths of both sources of information. First, the household survey is employed to develop an imputation model for the welfare measure in question – in this case, consumption expenditure per adult equivalent – that relies exclusively on characteristics common to survey and census. This model is then applied to the census data to obtain an imputed value of consumption expenditure for each household, allowing poverty rates to be computed. The local estimates of poverty that result from repeated application of this process are thus imputation-based and susceptible to errors. Nevertheless, the methodology has been empirically validated (Elbers, Lanjouw and Leite, 2008) and implemented in several developing countries. The acquired experience shows that the resulting estimates are reliable and precise enough to be useful for the purposes of policy design (Bedi, Coudouel and Simmler, 2007). Poverty estimates were calculated for all 80 constituencies in Lesotho. C. High levels of economic vulnerability non-poor individuals to falling back into poverty due, persist for example, to environmental shocks, price shocks, unemployment or other adverse events. Box 4 provides What household and individual characteristics are details on the methods used to identify households associated with a reduced likelihood of falling back vulnerable to poverty in Lesotho. Data on welfare into poverty? To answer these questions, this section self-perceptions support the results in identifying the identifies households not vulnerable to poverty – a group vulnerable: non-vulnerable households broadly perceive that has successfully escaped poverty – and explores that they do not belong to the poorest segments of the characteristics that make them different from those society (Figure 22a).6 In addition to the direct benefit of vulnerable to falling into poverty. In this way, the section no longer being vulnerable to falling into poverty, such examines the extent to which the observed poverty households tend to have advantageous impacts on a reduction in the past 15 years is a sign of persistent country’s tax base, savings, governance and more. progress. It considers the vulnerability of currently 6 The 2017/18 CMS/HBS asked households to place themselves on a six-step ladder, where the first step indicated the poorest people and the sixth step indicated the richest people. This question can assess whether households categorized as not being vulnerable to poverty actually feel they have escaped the bottom of the distribution. The poor and vulnerable for the most part think they are on the second step, while the non-vulnerable mostly believe they are on the third step. Hence, the non-vulnerable do believe they belong to the middle of the distribution (Figure 22a). Lesotho Poverty Assessment I Progress and challenges in reducing poverty 39 Box 4: Methods to find the households vulnerable to poverty The academic literature includes several methods to separate households vulnerable to poverty from the ones not vulnerable to poverty. For example, Lopez-Calva and Ortiz-Juarez (2014) use two-period panel data to predict the probability of households falling into poverty between the first and second round. They regress several sociodemographic characteristics and the occurrence of shocks on a binary indicator of whether households fall into poverty. This method is not feasible for countries where panel data is unavailable, which includes Lesotho. An alternative approach used by Chaudhuri et al. (2002) predicts both the mean consumption level and the variance of consumption for each household using only a cross-section of data. Combined with an assumption about normality, the predicted variance can be used to assess each household’s consumption level under different shocks and the probability the household falls below the poverty line. The probability of being poor declines with consumption, so this method can be used to identify a line above which households are no longer deemed vulnerable to poverty. Günther and Harttgen (2008) propose using multilevel models to study economic vulnerability. They extend the method by Chaudhuri et al. by allowing for groups of households to have different relationships between the outcome and explanatory variables. Notably, Günther and Harttgen allow for certain explanatory variables – in the present case, sociodemographic variables and shock variables – to have different impacts on welfare based on the households’ community. This method is relevant if communities respond to shocks differently based on location. The model also allows the breakdown of shocks by whether they are idiosyncratic or covariant. In this Poverty Assessment, the Günther and Harttgen method is used to derive a probability of being poor for each household. Lesotho’s constituencies are used as the community-level variable to measure how households might differ in how they transform sociodemographic variables and shocks into welfare. The household variables in the model, inspired by Günther and Harttgen, include the number of children and the head’s age, educational attainment, industry, sector of work and main source of income. In addition, community level variables are included, such as the constituency’s access to improved water, electricity, improved sanitation and house ownership. Finally, the model incorporates household and community level exposures to shocks. By running the above-described model and using the same variables to predict the squared residuals, an assumption about normality leads to predictions about the likelihood each household will be poor in the next year. Figure 21a shows the resulting predictions as well as the fitted line from them as a function of monthly consumption for urban and rural households separately. Often the predicted probabilities of being poor are used to create a vulnerability line above which households are considered non-vulnerable to poverty. This approach works well if the predictions decline sharply in consumption and if predictions do not vary much by other factors than consumption. The latter is not the case for Lesotho. For a given consumption level, rural households are predicted to be much more likely to be poor. This suggest they are exposed to more risk and that creating a fixed monetary threshold above which households are considered non-vulnerable would classify too many rural households and too few urban households as non-vulnerable. Instead, a probability threshold above which households are considered unlikely to be poor is set. Günther and Harttgen (2008) suggest using 29 percent, which is equivalent to considering households as vulnerable if they have a 50 percent or higher probability of falling below the poverty line at least once in the next two years. Using this 29 percent threshold, the number of vulnerable households is likely an upper bound. If more relevant variables had been available, the predictions would have been better, and the fitted line would have been steeper. Likewise, if the presence of shocks was larger in this cross-section than can be expected in the future – likely due to the presence of the El Niño prior to the fieldwork – then it is plausible that the number of vulnerable is overestimated. To the extent the current model overfits the data, on the other hand, the 29 percent threshold could downward bias the true number of vulnerable. 40 Lesotho Poverty Assessment I Progress and challenges in reducing poverty A simple logit model predicting poverty status was used as a robustness check. This approach gave very similar predicted probabilities of being poor (Figure 21b). The finding bolsters confidence that using a mixed model and predicting the variance separately while assuming normality is robust to a simpler approach. Figure 21: Identifying the vulnerable in Lesotho (a) The probability of being poor as a function (b) Comparing the predictions from Günther of consumption and Harttgen (2008) with a logit model 100 100 Vulnerable Günther and Harttgen (2008) approach Probability of being poor (%) 80 Probability of being poor (%) 80 60 60 20 29 40 20 40 0 100 200 649 1k 2k 5k Monthly consumption 0 Rural: fitted line Rural: household prediction 0 20 40 60 80 100 Probability of being poor (%) Urban: fitted line Urban: household prediction Logit model Source: Calculations based on the 2017/18 CMS/HBS. Note: The left figure shows the predicted probability of being poor as a function of monthly consumption, adjusted by the adult equivalence scale and expressed in 2017 prices. Households to the left of the vertical line at 649 are poor. Households above the horizontal line of 29 and to the right of the vertical line are not poor but considered at risk of falling into poverty. The right figure shows the predicted probabilities of being poor using the Günther and Harttgen (2008) approach and a logit model. The fact that the two predictions cluster around the green 45-degree line suggests that the main methodology is robust to a simpler approach. Despite progress in reducing poverty, more than Nearly all individuals in Rural Mountains, Rural 75 percent of Lesotho’s population is either poor Foothills and Rural Senqu River Valley are poor or or vulnerable to poverty. Figure 22b shows that 27.7 vulnerable to poverty. More than half of the population percent of the population was identified as vulnerable in Maseru is not vulnerable to poverty (56.9 percent), to falling into poverty in 2017. Combining this with the and the figure is nearly half in other urban areas (45.2 population share that is either poor or extremely poor percent). In rural areas, nearly all non-poor households means that 77.4 percent of the population was either resemble poor households. Less than 5 percent is not poor or vulnerable in 2017. In rural areas, 31.1 percent vulnerable to poverty in the Rural Mountains, Rural of the population was vulnerable to poverty. This is in Foothills and Rural Senqu River Valley. In these areas, addition to the 60.7 percent of the rural population that most of the households that have successfully escaped is already in poverty. By comparison, 21.3 percent of the poverty in the past two decades exhibit characteristics urban population was vulnerable to poverty. similar to those that have not. As a result, those out of poverty are at risk of falling back into it. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 41 Figure 22: The geographical location and perceptions of the vulnerable, 2017 (a) Households’ self-perceived welfare (b) Population by vulnerability and poverty status, by region level 100 3.3 3.6 3.6 60 8.3 13.2 22.6 80 33.1 28.5 28.6 Share of households (%) Share of population (%) 31.1 45.2 50.2 40 56.9 32.3 27.7 60 32.7 30.6 30.5 29.9 20 23.4 40 21.3 28.6 25.5 18.4 20 17.3 18.9 0 15.0 33.1 35.2 37.2 30.8 24.1 25.8 1 2 3 4 5 6 Poor Rich 11.2 9.7 12.4 0 0.0 0.0 Imagine six steps, where on the bottom, the first step, stand National Urban Rural Maseru Other Rural Rural Rural Rural poorest people, and on the highest step, the sixth, stand the rich Urban Urban Low− Foothills Senqu Moun− On which step are you today? lands River tains Valley Extremely poor Poor Vulnerable Non−vulnerable Extremely poor Poor Vulnerable Non−vulnerable Source: Calculations based on the 2017/18 CMS/HBS. Non-vulnerable households do not necessarily of vulnerable households have six or more members, belong to the middle class. Predictions of households while only 15 percent are one-person households. successfully escaping poverty do not necessarily mean Hence, large households are less likely to permanently that their income levels are far above the poverty line. escape poverty. Some households may have stable incomes that are just above the poverty line. A daily consumption of about The likelihood of not being vulnerable increases $10 per person per day (2011 PPPs) – equivalent to a with education levels: 60 percent of household monthly consumption per adult equivalent higher than heads among the non-vulnerable have secondary 2000 Maloti – has been used to define middle-class line or higher educational attainment. Poor households, in prior contexts. With this line, only 30 percent of the on the other hand, have similar educational attainment non-vulnerable belong to the middle class. The rest still to the economically vulnerable. More than two-thirds of have low incomes but are not very likely to fall back into vulnerable household heads have not completed primary poverty. In contrast, 6 percent of the vulnerable belong school (Figure 23b). Only 7 percent of household heads to the middle class. These households currently have among the economically vulnerable have obtained a relatively high consumption levels but are vulnerable to secondary education or more. In contrast, 60 percent shocks and thus at risk of falling back into poverty. of household heads among the non-vulnerable have secondary education degrees or more, and one in Smaller households are less likely to be vulnerable. five have post-secondary education. The data suggest Figure 23a shows the household size by economic status secondary education offers a sustainable path out of – extremely poor, poor, vulnerable or non-vulnerable. For poverty. Schooling tends to lift people out of poverty the non-vulnerable, nearly a third of households are one- and vulnerability because educated individuals have person households, and only 11 percent of households access to better jobs that can generate a stable income. have five or more members. By comparison, 22 percent 42 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Figure 23: Sociodemographic characteristics by vulnerability and poverty status, 2017 (a) Household size by vulnerability and poverty status 40 40 Share of households (%) 29 30 26 24 22 20 21 20 18 20 17 18 16 15 15 16 14 12 11 13 8 9 10 8 5 3 0 1 2 3 4 5 6+ 1 2 3 4 5 6+ 1 2 3 4 5 6+ 1 2 3 4 5 6+ Extremely poor Poor Vulnerable Non−vulnerable (b) Educational attainment of household heads by vulnerability and poverty status 80 71 Share of household heads (%) 67 64 60 39 40 25 25 23 22 22 20 14 10 7 6 0 1 0 0 None Secondary None Secondary None Secondary None Secondary Primary Tertiary Primary Tertiary Primary Tertiary Primary Tertiary Extremely poor Poor Vulnerable Non−vulnerable Source: Calculations based on the 2017/18 CMS/HBS. Non-vulnerable households derive their income Among various wage jobs, non-vulnerable from wages and non-agricultural businesses. Two- households are most likely to hold private sector thirds of the income of non-vulnerable households wage jobs or government jobs. Various kinds of comes from formal wages, compared to less than wage jobs exist, and it is important to understand half for vulnerable households (Figure 24a). Non- which of them are particularly associated with not vulnerable households are almost entirely independent being vulnerable to poverty. Figure 24b shows that of agricultural income and social assistance – both 18 percent of non-vulnerable household heads have only constitute 4 percent of their total income. For the a government job. By contrast, less than 5 percent of vulnerable, these two components are a quarter of their the economically vulnerable and the poor work for the income. This discrepancy suggests that relatively few government. Non-vulnerable household heads are also households manage to permanently escape poverty much more likely to work in the private sector. Nearly a through agriculture. Non-vulnerable households, in turn, third of those in non-vulnerable households have private are more likely to rely on self-employment income, sector wage jobs, while less than 15 percent of the which makes up 16 percent of their total income. vulnerable have such jobs, suggesting these types of Lesotho Poverty Assessment I Progress and challenges in reducing poverty 43 jobs are likely to shift households permanently out of much less likely to be inactive. Only 11 percent of the poverty. Among the non-vulnerable, 20 percent are self- non-vulnerable household heads are inactive, compared employed, compared to 10 percent for the vulnerable. to a third of the vulnerable. Household heads among the non-vulnerable are also Figure 24: Income and job characteristics by poverty and vulnerability status, 2017 (a) Share of total income by vulnerability and (b) Type of employment by vulnerability and poverty status poverty status 80 18% of household heads Government wage job among the non−vulnerable work for the government 67 Private sector wage job Share of total income (%) 60 49 Private household wage job 43 37 Other wage job 40 Self−employed 19 20 20 17 16 15 Household agriculture 13 14 14 12 12 12 9 9 7 4 4 2 2 4 3 Unemployed 0 Wage Agri− Business Social Transfer Other Inactive culture 0 10 20 30 40 Share of household heads (%) Extremely poor Poor Vulnerable Non−vulnerable Extremely poor Poor Vulnerable Non−vulnerable Source: Calculations based on the 2017/18 CMS/HBS. D. Lesotho has made progress in reducing multidimensional poverty, but gaps still exist, especially in rural areas This section complements the preceding analyses by that the portion of people with at least basic water exploring levels and trends in selected non-monetary services7 increased by 5.2 percentage points from 66.4 dimensions of poverty and well-being. The choice of percent in 2000 to 71.6 percent in 2015. The coverage non-income indicators is guided by the availability of rate was 87.4 percent in urban areas, compared to 65.7 data and relevance to Lesotho. These indicators include percent in rural areas. The gap between urban and housing conditions, improved drinking water, sanitation rural areas remained constant at around 22 percentage facilities, electricity, education and health services, points. Household survey data for 2017 indicate that asset ownership and food security. 82.2 percent of the population lived in households with access to a source of improved drinking water Access to basic services has increased, but within a 30-minute roundtrip from the dwelling unit. rural regions lag behind The coverage rates were 95.8 percent in urban areas and 75.1 percent in rural areas. According to NSDP-II, Despite strides in broadening access to improved several factors contributed to the lag in rural areas— water and sanitation services in the past 15 years, reduced investments, climate change and declining rural areas lag in terms of coverage. Figure 25a shows yields or drying-up of water sources. 7 This indicator encompasses both people using basic water services and those using safely managed water services. Basic drinking water service is defined as drinking water from an improved source, provided collection time is not more than 30 minutes for a round trip. Improved water sources include piped water, boreholes or tubewells, protected dug wells, protected springs and packaged or delivered water. 44 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Figure 25: Changes in the proportion of the population with access to selected basic services (a) At least basic drinking water service (b) At least basic sanitation service 100 100 90 90 80 80 Share of population (%) Share of population (%) 70 70 60 60 50 50 40 40 30 30 20 20 10 10 0 0 2000 2002 2004 2006 2008 2010 2012 2014 2001 2003 2005 2007 2009 2011 2013 2015 2000 2002 2003 2004 2005 2006 2007 2008 2009 2010 2012 2013 2015 2001 2011 2014 Lesotho Urban Rural Lesotho Urban Rural (c) Electricity (d) Access in 2017 100 95.8 100 90 90 82.2 80 Share of population (%) Share of population (%) 80 75.1 70 70.3 70 60 60 50 50 45.1 44.4 45.5 40 40.6 40 30 30 25.2 20 20 10 10 0 0 2000 2002 2003 2005 2006 2008 2009 2011 2012 2014 2015 2001 2004 2007 2010 2013 At least basic At least basic Electricity drinking water sanitation Lesotho Urban Rural services services Lesotho Urban Rural Source: World Development Indicators for 2000–2015 and the 2017/18 CMS/HBS for 2017. Access to sanitation services has increased rapidly compared to an increase of 27.2 points to 45.7 percent since 2000. Figure 25b reports data on the proportion in urban areas. Household survey data for 2017 indicate of people using at least basic sanitation services— 45.1 percent of the population lived in households with i.e., improved sanitation facilities not shared with access to an improved sanitation facility not shared with other households.8 Starting from a low of 7.2 percent other households. The corresponding coverage rates nationally in 2002, the percentage of the population were 44.4 percent in urban areas and 45.5 percent in with access to improved sanitation services registered rural areas. The NSDP-II recognizes the importance of a 36.6 percentage point increase to 43.8 percent in a “programmatic approach aimed at increasing access 2015. Although still lagging, rural areas experienced an to water and improved sanitation and hygiene” in the increase of 38.6 points from 4.5 percent to 43.1 percent, country’s agenda to reduce poverty. 8 This indicator encompasses both people using basic sanitation services and those using safely managed sanitation services. Improved sanitation facilities include flush/pour flush to piped sewer systems, septic tanks or pit latrines, ventilated improved pit latrines, compositing toilets or pit latrines with slabs. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 45 Expanding access to electricity among the percent. Recent household survey data indicate that population has been slow. Between 2000 and 2015, 40.6 percent of the population lived in households with Lesotho achieved a 25.7 percentage point increase from access to electricity in 2017. In urban areas, this figure 4.3 percent to 30.0 percent in national electricity access was 70.3 percent, compared with 25.2 percent in rural rates. Most of the expansion occurred in urban areas, areas. Access to electricity remains concentrated in where the proportion of the population with access to urban areas, and it has been identified by NSDP-II as a electricity increased from 13.6 percent to 69.6 percent. constraint to improving Lesotho’s investment climate as In rural areas, the share rose from 2.0 percent to 20.0 well as livelihoods of the population. Figure 26: Changes in the proportion of the population with access to selected basic services, comparison to other countries (a) At least basic drinking water (b) At least basic sanitation service (c) Electricity service 100 100 100 90 90 90 80 Share of population (%) 80 80 Share of population (%) Share of population (%) 70 70 70 60 60 60 50 50 50 40 40 40 30 30 30 20 20 20 10 10 10 0 0 0 Lesotho Middle income Sub-Saharan Africa Low income Lower middle income Eswatini South Africa Botswana Lesotho Lower middle income Namibia Middle income Botswana Sub-Saharan Africa Low income Eswatini South Africa Namibia Botswana Low income Eswatini South Africa Lesotho Lower middle income Middle income Namibia Sub-Saharan Africa 2000 2015 2000 2015 2002 2015 Source: World Development Indicators. Lesotho lags other lower middle-income countries services in 2017, while Rural Senqu River Valley had the in providing basic services. Figure 26 compares lowest share with basic sanitation services and access Lesotho’s coverage rates in water, sanitation and to electricity. These two regions also had the highest electricity to those in other SACU countries and regions. national poverty rates in 2017. Interestingly, Maseru Lesotho falls behind an average lower middle-income Urban, with the highest access rates for improved country for all three indicators and lags all other SACU drinking water services and electricity, had the second countries with regard to electricity and improved drinking lowest share of population with at least basic sanitation water services. For access to improved sanitation services. This underscores the general challenge facing services, Lesotho is ahead of only Namibia. the country regarding expanding access to improved sanitation facilities and suggests that the relatively low The spatial pattern of access to basic public services monetary poverty rates in Maseru Urban do not correlate is stark and closely follows the urban-rural divide. perfectly with improving non-monetary dimensions of Rural regions tend to have lower proportions of people poverty. Overall, increasing access to basic services in with access to basic services (Figure 27). Rural Mountains rural regions should be a priority for the government. had the lowest share with access to basic drinking water 46 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Figure 27: Proportion of the population with access to selected basic services, by region, 2017 11.4 Rural Senqu River Valley 29.7 83.1 14.6 Rural Mountains 41.0 63.6 12.8 Rural Foothills 39.9 71.2 37.5 Rural Lowlands 53.1 80.5 67.6 Other Urban 50.1 93.4 73.9 Maseru Urban 36.7 99.1 0 20 40 60 80 100 Share of population (%) Electricity At least basic sanitation services At least basic drinking water services Source: Calculations based on the 2017/18 CMS/HBS. Access to basic public services correlate negatively population. For the top 10 percent, 55.7 percent had with poverty levels, with lowest access among the access to improved sanitation facilities. The access gap poorest segments of the population. In 2017, 72.1 between the poor and rich is widest for electricity: only percent of the poorest 10 percent of the country’s 9.9 of the poorest 10 percent had access to electricity population had access to an improved water source in 2017, compared to 78.8 percent among the richest – 22.1 percentage points lower than the proportion 10 percent. Among the poor in 2017, access was 75.5 among the richest 10 percent (Figure 28). Similarly, at percent for improved drinking water sources, 39.9 29.2 percent, access to improved sanitation facilities percent for improved sanitation facilities and 23.7 was lowest among the poorest 10 percent of the percent for electricity. Figure 28: Proportion of the population with access to selected basic services, by decile, 2017 (a) At least basic drinking water (b) At least basic drinking water (c) Electricity service service 100 100 100 90 90 90 Share of population (%) 80 Share of population (%) 80 80 Share of population (%) 70 70 70 60 60 60 50 50 50 40 40 40 30 30 30 20 20 20 10 10 10 0 0 0 Bottom 40% Richest decile Poorest decile Non-poor Poorest decile Bottom 40% Bottom 40% Richest decile 2 3 4 5 6 7 8 9 Richest decile Poorest decile Poor Non-poor Non-poor 2 3 4 5 6 7 8 9 2 3 4 5 6 7 8 9 Poor Poor Consumption per adult equivalent Consumption per adult equivalent decile Consumption per adult equivalent decile decile Source: Calculations based on the 2017/18 CMS/HBS. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 47 Housing conditions and asset holdings have improved In 2017, 72.0 percent of peopled lived in a house with with an improved roof increases with income levels. an improved roof, up marginally from 70.0 percent In 2017, 94.1 percent of people in the richest decile in 2002. Urban households are more likely than rural lived in houses with an improved roof, 46.5 percentage residents to live in houses with an improved roof (Figure points higher than the proportion among the poorest 10 29a).9 As expected, the likelihood of living in a house percent (Figure 29b). Figure 29: Access to an improved roof, 2002–2017 (a) By geographic location (b) By decile 59.3 44.1 Bottom 40% 65.4 Rural Senqu River Valley 66.8 61.0 Poor 67.3 28.0 Rural Mountains 32.2 Non-poor 82.9 74.0 51.6 Richest decile 94.1 Rural Foothills 57.4 79.8 9 84.4 82.7 74.5 Rural Lowlands 78.0 8 83.1 93.2 73.3 Other Urban 96.4 7 78.5 69.6 97.9 6 75.2 Maseru Urban 99.7 68.6 68.0 60.0 5 74.8 Rural 61.8 68.6 4 69.5 95.2 61.1 Urban 97.7 3 67.1 59.8 Lesotho 72.0 2 63.2 70.2 47.6 Poorest decile 61.7 0 20 40 60 80 100 0 20 40 60 80 100 Share of population (%) Share of pooualtion (%) 2017 2002 2017 2002 Source: Calculations based on the 2002/03 HBS and the 2017/18 CMS/HBS. The growth in household consumption between around three. Consistent with patterns for monetary 2002 and 2017 was accompanied by improvements poverty, asset holdings are on average higher in urban in asset holdings. Insofar as they capture material than in rural regions (Figure 30a). As expected, asset deprivation, ownership of physical assets is frequently holdings tend to be higher among richer households used to examine households’ welfare status (Box 5). (Figure 30b). In 2017, the richest decile had an average In 2002, an average Basotho household owned around of around five out of 17 asset types, close to three times two of 17 asset types. In 2017, this had increased to the ownership rate for the poorest decile. 9 An improved roof is defined to include: corrugated iron, wood, cement fiber, ceramic tiles, cement and roofing shingles. Traditional thatched roofs are not considered improved because they harbor pests and disease and require significant maintenance. 48 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Figure 30: Changes in household asset holdings, 2002–2017 (a) By geographic location (b) By decile 1.9 Bottom 40% 2.1 Rural Senqu River Valley 1.2 1.2 Poor 2.3 1.7 1.4 Rural Mountains 0.5 Non-poor 3.9 2.2 Rural Foothills 2.5 5.2 1.3 Richest decile 3.0 3.1 9 4.3 Rural Lowlands 1.7 2.3 8 3.7 4.1 2.1 Other Urban 2.7 7 3.4 1.9 Maseru Urban 4.3 3.1 3.9 6 1.7 2.6 5 2.7 Rural 1.3 1.7 4 2.7 4.2 1.4 Urban 3.2 3 2.2 1.3 Lesotho 3.1 2.1 1.7 2 1.2 Poorest decile 1.6 0 2 4 6 1.0 Average number of household assets owned (out of 17) 0 2 4 6 Average number of household assets owned (out of 17) 2017 2002 2017 2002 Source: Calculations based on the 2002/03 HBS and the 2017/18 CMS/HBS. Box 5: Construction of an asset index The index is constructed by counting the number of assets types a household owns from a specified set of durable assets (excluding a car/truck). For this analysis, a set of 17 assets was identified as being common in the 2002 and 2017 datasets. The list includes: radio, telephone (either landline or cellular), television, tractor, electric/gas/coal stove, scotch cart, computer/laptop, bicycle, motorcycle/scooter, generator, refrigerator, fan, air conditioner, water heater, washing machine, video player and camera/video camera. For each durable asset, a dummy variable was created, valued at one if a household owns at least one of that asset and zero otherwise. The total asset ownership index for each household was computed by adding up the dummy variables. Given that the set being analyzed includes 17 items, the index ranges from zero (none of the items) to 17 (at least one of each item). A household owning five out of the 17 items, for example, gets a score of 5. The poor experience multiple deprivations people who had no access. The gap was widest for electricity: the group with access had a poverty rate Lack of access to basic services is associated with of 29.1 percent, while the group without access was high poverty rates. Figure 31 reveals wide gaps at 63.7 percent. These patterns underscore poverty as between poverty rates among individuals with and a barrier to access to basic services and a contributor without access to basic services. In 2017, for instance, to and/or a result of resource inequality. In addition, 45.7 percent of people who had access to an improved the patterns highlight the need for the government to drinking water source were poor, 22.7 percentage address the constraints, for example, of affordability or points below the poverty rate of 68.3 percent among infrastructure, which limit access by the poor. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 49 Figure 31: Poverty rate by access to basic services, 2017 80 68.3 69.2 70 63.7 60 Povetry headcount rate (%) 54.4 50 45.7 44.0 42.1 40 29.1 30 20 10 0 Improved Improved Electricity Improved roof Improved Improved Electricity Improved roof drinking water sanitation facility drinking water sanitation facility source source Access No access Source: Calculations based on the 2017/18 CMS/HBS. The monetary poor are simultaneously deprived percent for an improved roof. Considering the whole in multiple dimensions. Figure 32a illustrates the population shows that 16.2 percent of the population multiplicity of deprivations facing Lesotho’s poor. At were simultaneously deprived in improved sanitation, least 24 percent of the monetarily poor, measured at the electricity and improved roof in 2017. Almost 10 percent national poverty line, were affected by one of the four of the population did not have access to an improved deprivations tracked in Figure 32. For example, 24.5 drinking water source, sanitation and electricity. On percent of the poor had no access to improved water, average, deprivation levels tend to be higher in rural while 60.1 percent had no access to improved sanitation areas than at the national level (Figure 32b). facilities. It was 76.3 percent for electricity and 39.0 Figure 32: Multiple deprivations faced by the poor, 2017 (a) Lesotho (b) Rural No improved No improved drinking water drinking water source source 24.5% 28.6% Poor No improved Poor 76.3% 60.1% 59.8% No improved No electricity (49.7% of the sanitation facility No electricity 80.6% (60.7% of the rural sanitation facility population) population) 39.0% 46.2% No improved roof No improved roof Source: Calculations based on the 2017/18 MS/HBS. 50 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Food insecurity is prevalent Hunger is a serious problem in Lesotho. The 2018 Lesotho’s GHI score is driven by relatively high Global Hunger Index (GHI) gives the country a score stunting rates – about 33 percent in 2014 – among of 23.7, ranking 78th among 119 qualifying countries.10 children of up to 5 years of age. This underscores the Encouragingly, the country’s GHI score has been importance of addressing nutrition in the early stages of generally trending downwards (Figure 33). Comparison a child’s life to achieve sustainable human and economic to other SACU countries indicates the country performed growth in Lesotho. According to the 2016 Cost of Hunger slightly better than Botswana and Namibia in 2018. All Study for Lesotho, the country lost about 7.1 percent SACU countries except for South Africa suffered from of total GDP because of child undernourishment. The serious hunger problems in 2018, with South Africa’s money is lost through increased healthcare spending, situation was classified as moderate. additional burdens on the education system and lower productivity in the workforce. Figure 33: Food insecurity and malnutrition, 2000–2018 (a) GHI scores (b) Malnutrition 50 50 Share of children under age five (%) 45 45 Global hunger index (GHI) score 40 40 35 35 30 30 25 25 20 20 15 10 15 5 10 0 5 2000 2005 2010 2018 South Africa 18.1 20.8 16.1 14.5 0 Eswatini 2004 2009 2014 28.9 27.6 26.7 22.5 Lesotho 32.5 29.7 26.3 23.7 Underweight 16 13 10 Namibia 30.6 28.4 30.9 24.3 Stunting 44 39 33 Botswana 33.1 31.2 28.4 25.5 Wasting 5 4 3 Sub-Saharan Africa 43.6 38.1 32.2 29.4 World 29.2 27 23.1 20.9 Overweight 7 7 7 Source: For GHI scores: von Grebmer et al. (2018). For malnutrition: Lesotho Ministry of Health and ICF International (2016). Household spending on food accounts for a high (Figure 34). In general, Lesotho faces considerable share of total spending, highlighting the challenge challenges in addressing food security for poor and of food insecurity in the country. The average share vulnerable households, the majority of which are of food spending in total household consumption women and children (World Bank 2019f). This challenge expenditure was 63.6 percent in 2017, rising to as high is compounded by a high HIV/AIDS rate, which has had as 78.4 percent among households in the poorest decile a significant negative impact on agricultural productivity. 10 The GHI is produced annually by the International Food Policy Research Institute to measure and track hunger across countries and regions. The GHI is a multidimensional tool that combines three equally weighted indicators: 1) the prevalence of undernourishment; 2) the prevalence of underweight children age 5 and below; and 3) the mortality rate of children under age 5. The index is used to rank countries on a 100-point severity scale, with zero indicating “no hunger” and 100 being the worst. Values less than 10 indicate “low hunger,” values between 10.0 and 19.9 reflect “moderate hunger,” values between 20 and 34.9 indicate “serious hunger,” values between 35.0 and 49.9 are for “alarming hunger” and values exceeding 50 point to “extremely alarming” hunger problems (von Grebmer al. 2018). Lesotho Poverty Assessment I Progress and challenges in reducing poverty 51 Figure 34: Share of food in total household consumption expenditure, 2017 (a) By geographic location (b) By decile Rural Senqu River Valley 74.3 Bottom 40% 73.2 Poor 71.9 Rural Mountains 72.6 Non-poor 58.0 Rural Foothills 71.0 Richest decile 47.1 9 56.6 Rural Lowlands 66.6 8 62.3 Other Urban 57.4 7 63.3 6 66.6 Maseru Urban 51.9 5 67.7 Rural 69.5 4 69.8 Urban 54.9 3 70.3 2 75.3 Lesotho 63.6 Poorest decile 78.4 0 20 40 60 80 100 0 20 40 60 80 100 Share of food in total household consumption Share of food in total household consumption expenditure (%) expenditure (%) Source: Calculations based on the 2002/03 HBS and the 2017/18 CMS/HBS. Box 6: Fill the Nutrient Gap Lesotho The results of the Fill the Nutrient Gap (FNG) Lesotho analysis, launched in August 2019, highlight the triple burden of micronutrient deficiencies, undernutrition and overweight/obesity and discusses its social and economic determinants and implications. The key findings from the analysis are: 1. Despite the improvement Lesotho has made in the health sector, malnutrition remains widespread. Over a third of children under age 5 are stunted and unlikely to reach their full mental and physical potential. In addition, the rising overweight and obesity rates of adults contribute significantly to the country’s public health problems. The malnutrition burden is a challenge throughout all wealth quintiles and geographic districts. Child undernutrition cost Lesotho an estimated $200 million annually, equivalent to 7 percent of its GDP, and exacerbates the impact of HIV. 2. Economic access is low. More than half of Basotho households (56 percent) would not be able to afford a diet that meets their nutrient requirements, if they made the optimal dietary choices, and the proportion rises to more than 70 percent in the mountainous regions (Mokhotlong, Thaba-Tseka and Qacha’s Nek). 3. Low dietary diversity indicates supply- and demand-side issues. About 80 percent of dietary energy comes from such staple foods as maize and starchy roots, with a small proportion from fresh fruit, vegetables and animal source foods. The limited consumption of fresh foods rich in essential nutrients is both an issue of supply (production, availability) and demand (affordability, choices). On the supply side, staples dominate production, with more than 85 percent of total agricultural output focused on cereals (primarily maize) and potatoes. Livestock is typically considered an asset used for income generation but rarely for consumption, and recent economic growth has not led to notable increases of availability of animal-sourced foods, vegetables or fruits. On the demand side, the finding that all wealth quintiles consume mainly staples low in essential nutrients suggests factors beyond affordability issues, and people are not choosing enough nutritious foods. Agricultural productivity, already low, is expected to decrease further as climate conditions worsen, and this will have adverse impacts on crop yields, further impacting availability. 52 Lesotho Poverty Assessment I Progress and challenges in reducing poverty 4. Lesotho’s private sector has yet to reach its full potential. Lesotho depends on South Africa to supply its internal market. The agricultural value chain is dominated by government initiatives, and participation of micro, small and medium agricultural enterprises along the chain is limited. Access to credit for agriculture, and subsequent growth and investment, is very low, making it difficult for farmers to increase their own production beyond subsistence farming. No agricultural insurance is available for farmers to take risks, make investments and diversify beyond staple food production. 5. Adolescent girls and breastfeeding women are at higher risk of micronutrient deficiencies. The cost of the diet analysis, part of the FNG assessment, identified the most challenging nutrients in meeting the needs of these target groups: iron, calcium, folic acid and vitamin C. Micronutrient deficiencies among these groups have adverse consequences on their health and the health of their (future) children; for example, high anemia rates in children aged 6-8 months indicate low iron stores from mothers. This is compounded by sub-optimal breastfeeding and complementary feeding practices. 6. Rural and remote areas bear the brunt of food insecurity and the subsequent malnutrition burdens. This is in part due to nutritious diets’ higher costs and worsened by seasonality and limited access to markets. Dietary diversity is low everywhere, but particularly in the more remote areas of mountainous regions and the Rural Senqu River Valley. 7. Reliance on social safety nets is relatively high because of limited economic opportunities, but access to social safety nets has not been able to ensure food security. As reported in the 2016 Lesotho Vulnerability Assessment and Analysis Report (LVAC), many rural households face survival and livelihood deficits, indicating that social safety nets only partly meet needs, an issue that may be related to targeting and the extent of assistance provided. 8. The primary school feeding program has universal coverage, and it provides a strategic entry point for improving nutrition during a critical stage of life by ensuring that meals meet children’s nutritional needs. The program has two added benefits: it offers an opportunity to explore how school meals can be used as a platform to reach adolescent girls with nutrition interventions, and it can stimulate production of diverse foods by ensuring steady demand. 9. No single intervention will eliminate malnutrition in Lesotho, and action is required from all sectors to prevent malnutrition. Priority sectors include health, education, agriculture (both small and commercial scale) and social protection. Identified interventions include universal coverage of iron and folic acid supplements for adolescent girls and women of reproductive age, improvement of infant and young child feeding practices, support for fresh school meal and investment and improvement in poultry and vegetable value chains. Source: Lesotho Food and Nutrition Coordinating Office, WFP, IFAD, UNICEF, FAO. 2019. “Fill the Nutrient Gap Lesotho. Summary Report.” https://www.wfp.org/fillthenutrientgap. Note: The FNG analysis was conducted by the Food and Nutrition Coordinating Office (FNCO) and the World Food Programme (WFP), with the International Fund for Agricultural Development (IFAD), the United Nations Children’s Fund (UNICEF) and the United Nation’s Food and Agriculture Organization (FAO). To learn more about the FNG concept and methodology, see Bose et al. (2019). Lesotho Poverty Assessment I Progress and challenges in reducing poverty 53 Like monetary poverty, the incidence of multidimensional poverty is high and concentrated in rural areas Box 7: Estimating a multidimensional poverty index (MPI) for Lesotho The Alkire-Foster (AF) method is used to estimate an MPI for Lesotho (L-MPI). The approach complements monetary poverty measures by identifying and counting the number of overlapping deprivations experienced simultaneously by individuals or households. It is built on three premises: selection of poverty dimensions and indicators; identification of the poor based on set criteria (which involves setting cutoffs or poverty lines to determine poverty/deprivation status); and aggregation of information through a poverty index. Each dimension and each indicator within a dimension are equally weighted. Any person who fails to meet the deprivation cutoff is identified as deprived in that indicator. A person is identified as multidimensionally poor if they are deprived in at least one-third of the weighted MPI indicators. In other words, a person is multidimensionally poor if the person’s weighted deprivation score is equal to or higher than the poverty cutoff of 33.3 percent. Following the AF methodology, the MPI is calculated by multiplying the incidence of poverty (H) and the average intensity of poverty (A). More specifically, H is the proportion of the population that is multidimensionally poor, while A is the average proportion of dimensions in which poor people are deprived. So, MPI = H x A, reflecting both the share of people in poverty and the degree to which they are deprived. A detailed explanation of the method is presented in Alkire and Foster (2011) and an application of the method is presented in Alkire and Santos (2014). The choice of dimensions and indicators for estimating the MPI was guided by the 2018 Global MPI and its dimensions and indicators, the country context and data availability. The 2018 Global MPI is an international measure published in 2018 by Oxford Poverty and Human Development Initiative (OPHI) and the United Nations Development Program (UNDP). Like the Global MPI, the L–MPI consists of three dimensions and 10 indicators. The two, however, differ in terms of the indicators used under the health dimension, owing to unavailability of nutrition and child mortality indicators in the 2017/18 CMS/HBS dataset. Two alternative indicators are used under the health dimension. The first is the distance to the nearest health facility. The second draws from the Fill the Nutrient Gap (FNG) report, launched in August 2019, and compares household per capita spending on food to the per capita value of a nutritious diet reported in the FNG report. Table 7 describes the dimensions and indicators used. Equal weights across dimensions is assumed, along with equal weights across indicators within each indicator. The data source for the analysis is the 2017/18 CMS/HBS. Table 7: L-MPI: Dimensions, indicators, deprivation cut-offs and weights Dimension Indicator Deprived if… Weight Access to a health Household is located more than five kilometers 1/6 facility from the nearest health facility. Health Household per capita spending on food is less than 1/6 Nutrient gap the per capita value of a nutritious diet as reported in the FNG. No household member aged 10 years or older has 1/6 Years of schooling completed six years of school. Education Any school-aged child is not attending school up to 1/6 School attendance the age of completing class 8. 54 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Dimension Indicator Deprived if… Weight Household cooks with dung, wood, charcoal or 1/18 Cooking fuel coal. Household’s sanitation facility is not improved 1/18 Sanitation (according to SDG guidelines) or it is improved but shared with other households. Household does not have access to improved 1/18 drinking water (according to SDG guidelines) or Drinking water safe drinking water is at least a 30-minute roundtrip Living walk from home. standards Electricity Household has no electricity. 1/18 At least one of the three housing materials for roof, 1/18 walls and floor are inadequate; the floor is of natural Housing materials and/or the roof and/or walls are of natural or rudimentary materials. Household does not own more than one of these 1/18 assets: radio, TV, telephone, computer, animal cart, Assets bicycle, motorbike or refrigerator and does not own a car or truck. Source: Authors’ presentation. One attraction of the MPI estimated here is that it is decomposable by space and population attributes. This makes it a powerful tool for not only identifying the poor and where they live but also for guiding targeted policy interventions to what contributes to poverty in those areas, allowing resources to be channeled properly. Nationally, the deprivations are highest in the nu- the deprivations are highest in cooking fuel, followed trient gap, followed by electricity. Table 8 reports the by electricity and then the nutrient gap. This suggests people who are deprived in each of the 10 indicators. that attention to nutrition and electricity will be required These “raw” headcounts consider all deprivations and more widely than just among the poor. Consistent with not just those among the poor. In urban areas, the depri- the spatial patterns of monetary poverty, deprivations vations are highest for improved sanitation, followed tend to be highest in Rural Mountains and Rural Senqu by the nutrient gap. In rural areas, on the other hand, River Valley. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 55 Table 8: Raw headcount rates, 2017 Other Urban Rural Senqu River Valley Mountains Lowlands Foothills National Maseru Urban Urban Rural Rural Rural Rural     Access to a health Health care facility 32.8 8.2 45.5 2.0 12.9 33.4 50.3 59.3 61.2 Nutrient gap 64.5 50.6 71.8 51.9 49.6 67.3 71.8 78.3 76.6 Years of schooling 7.5 4.0 9.3 3.6 4.2 5.4 13.6 13.2 12.0 Education School attendance 3.0 0.5 4.3 0.1 0.7 1.2 5.4 9.4 5.5 Cooking fuel 54.7 13.0 76.4 6.2 18.1 61.1 88.8 92.2 91.3 Sanitation 54.9 55.6 54.5 63.3 49.9 46.9 60.1 59.0 70.3 Living Drinking water 17.8 4.2 24.9 0.9 6.6 19.5 28.8 36.4 16.9 standards Electricity 59.4 29.7 74.8 26.1 32.4 62.5 87.2 85.4 88.6 Housing 43.8 14.0 59.3 7.5 18.8 36.2 69.3 88.3 82.0 Assets 33.4 16.7 42.1 14.4 18.4 30.6 40.0 61.1 52.6 Source: Calculations based on the 2017/18 CMS/HBS. About half of Lesotho’s population was identified areas, Table 9 shows the intensity of multidimensional as multidimensionally poor in 2017. In 2017, 51.6 poverty is high in both urban and rural areas. An average percent of the population was deprived in at least one- multidimensionally poor person is deprived in 41.8 third of the weighted indicators in Table 7. At 69.5 percent of the weighted indicators in urban areas and percent, the incidence of multidimensional poverty in 49.8 percent in rural areas. This implies that the extent rural areas was 51.8 percentage points higher than urban of multidimensional poverty is very high for the few area’s 17.7 percent. Although the multidimensional multidimensionally poor individuals in urban areas. poverty headcount rate is significantly lower in urban Table 9: The incidence and intensity of multidimensional poverty in Lesotho Multidimensional poverty   headcount rate (%) Intensity (%) MPI Lesotho 51.6 48.8 0.252 Urban 17.7 41.8 0.074 Rural 69.5 49.8 0.346 Maseru Urban 11.8 39.8 0.047 Other Urban 22.2 42.8 0.095 Rural Lowlands 53.3 44.1 0.235 Rural Foothills 77.1 51.9 0.400 Rural Mountains 88.5 54.4 0.481 Rural Senqu River Valley 87.2 52.1 0.454 Source: Calculations based on the 2017/18 CMS/HBS. 56 Lesotho Poverty Assessment I Progress and challenges in reducing poverty There is a clear spatial dimension to multidimensional caution around formulating policies or interventions poverty in Lesotho. Rural Mountains had the highest based only on the poverty headcount rate. In this case, multidimensional poverty headcount rate in 2017 at 88.5 the multidimensional poverty headcount rate hides the percent, followed by Rural Senqu River Valley at 87.2 situation of the multidimensionally poor. The result, percent (Table 9). This is consistent with the regional coupled with the finding that Maseru Urban had the poverty story revealed by the monetary indicators of lowest monetary poverty in 2017, suggests better poverty. Maseru Urban had the lowest incidence and performing provinces do have pockets of intense intensity of multidimensional poverty. Even so, the multidimensional poverty. Even so, rural regions intensity of multidimensional poverty was high at 39.8 contributed most to multidimensional poverty levels in percent. This is of policy relevance because it supports 2017 (Figure 35). Figure 35: Contribution of weighted indicators to the MPI, geographic locations, 2017 2.8% Maseru Urban 10.1% 13.6% 7.4% Other Urban Rural Lowlands Urban 29.8% Rural 31.4% Rural Foothills 89.9% Rural Mountains 15.1% Rural Senqu River Valley Source: Calculations based on the 2017/18 CMS/HBS. The nutrient gap slows progress toward reducing MPI). This was followed by deprivation in access to a multidimensional poverty. Figure 36 shows what health facility at 19.2 percent. This underscores the each indicator contributed to multidimensional poverty importance of nutrition and equitable access to quality in 2017. The nutrient gap was the most important factor healthcare facilities in reducing multidimensional in multidimensional poverty (29.7 percent of estimated poverty in Lesotho. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 57 Figure 36: Contribution of weighted indicators to the MPI, dimensions and indicators (a) Dimensions (b) Indicators 100% 100% 90% Assets 90% Contribution to MPI (%) Contribution to MPI (%) 80% 80% Housing 70% 70% 60% Living standards 60% Electricity 50% 50% Education 40% 40% Drinking water Health 30% 30% 20% Sanitation 20% 10% 10% 0% Cooking fuel 0% Maseru Urban Rural Mountains Other Urban Rural Senqu River Valley Rural Lowlands Rural Foothills Urban National Rural Rural Mountains Other Urban Rural Senqu River Valley Maseru Urban Rural Foothills Rural Lowlands Urban National Rural School attendance Years of schooling Nutrient gap Source: Calculations based on the 2017/18 CMS/HBS. Maps further support the spatial dimension of least deprivation rates. Deprivation levels vary greatly poverty and deprivation in Lesotho. The rural areas, by indicator, with the vast majority having access to particularly in the mountainous regions, are more improved water, but only 36 percent having access likely to suffer deprivations and subsequently poverty to electricity and 20 percent having access to internet (Figure 37). Maseru and the lowlands tend to have the (Population and Housing Census 2016). Figure 37: Share of consistencies deprived in selected indicators (a) Share without access to improved sanitation (b) Share without access to improved water sources facilities 58 Lesotho Poverty Assessment I Progress and challenges in reducing poverty (c) Share without access to electricity (d) Share without access to internet (e) Share without completed primary school (f) Share not knowing how to read and write Source: Calculations based on the 2016 PHC. In sum, this chapter documents the progress of poverty in the past 15 years, poverty is persistent Lesotho has made and the challenges it still faces and deeply entrenched in Lesotho, especially in rural in reducing poverty. It shows that despite progress in areas. This suggests that improvements in rural areas is reducing both monetary and non-monetary dimensions central to future poverty reduction. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 59 Chapter I.2: The nature and evolution of inequality in Lesotho Historically, inequality has been very high in Lesotho. Yet, consumption of the bottom 40 percent of the population grew by 2.2 percent annually between 2002 and 2017—much faster than the top 60 percent of the population, which rose only by 0.1 percent annually. This boost in shared prosperity caused a decline in inequality. Although Lesotho is now more equal than its neighbors, with a Gini coefficient of 44.6, it remains an unequal country. In addition to high inequality of outcomes, Lesotho faces high inequality of opportunity. Factors such as place of birth, parents’ education, health shocks and environmental shocks contribute to half of the current level of inequality in Lesotho. Relatively low intergenerational mobility exacerbates high inequality in Lesotho. A. The bottom of the population grew the 2002 and 2017, while it grew nearly 2 percent annually fastest in urban areas (Figure 38b). This gap clearly shows the growing urban-rural divide. The 40 percent wealthiest The fastest consumption growth has been rural households saw a decline in their consumption experienced at the very bottom of the population.11 between 2002 and 2017. Part of this decline was due to Nationally, the poorest 10 percent of Basotho had the drought experienced in 2017, which impacted rural annualized growth rates above 4 percent between 2002 households most severely. and 2017 (Figure 38a). In contrast, households around the median managed annualized consumption growth The bottom 20 percent more than doubled their rates of only around 1 percent, while the top 10 percent share of total consumption. In 2002, the bottom experienced a reduction in their consumption levels. quintile was responsible for 2.4 percent of total This suggest that the distribution of welfare became consumption—well below the 20 percent that would more equal during this period. indicate an equal distribution. By 2017, this share had more than doubled to 5.4 percent (Figure 39a). At the Consumption growth was more favorable among same time, the share of the total consumption value the urban population. While the bottom 10 percent held by the top quintile fell from nearly 60 percent to grew equally quickly in both urban and rural areas, the 45 percent. A typical individual in the top quintile had distribution’s middle in urban areas had much stronger 14 times the consumption of a typical individual in the growth rates than the middle in rural areas. Median bottom quintile in 2002; by 2017, this ratio was eight, a consumption in rural areas did not change between massive reduction over 15 years. 11 A simple way to determine whether the welfare distribution has gotten fairer in recent decades is to look at how fast each part of the distribution grew. This can be accomplished with growth incidence curves, which show how fast each part of the distribution grew between two points in time. Growth incidence curves come in two types: Non-anonymized curves track the growth in consumption of specific households over time. This type requires panel data and cannot be conducted for Lesotho. Anonymized curves rank all households according to their consumption levels at two points in time and observe how much household consumption at the xth percentile of the distribution grew. 60 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Figure 38: Consumption growth-incidence curves, 2002–2017 (a) National growth-incidence curve (b) Rural and urban growth-incidence curves 6 6 Annualized growth in consumption (%) Annualized growth in consumption (%) 4 4 2 2 0 0 −2 −2 0 20 40 60 80 100 Rank in national distribution of consumption 0 20 40 60 80 100 Rank in rural/urban distribution of consumption National Rural Urban Source: Calculations based on the 2002/03 HBS and the 2017/18 CMS/HBS. Note: The left figure ranks households according to their consumption and shows the annualized growth rates in consumption at each part of the distribution. The right figure replicates this for urban and rural households, separately. Figure 39: Quintile shares and consumption relative to mean, 2002–2017 (a) Quintile shares (b) Consumption relative to mean 59.4 10x mean 60 Consumption relative to mean 5x mean Share of total consumption (%) 45.1 2x mean 40 1x mean 23.9 20.6 1/2 of mean 20 15.6 10.0 11.4 5.4 6.3 1/5 of mean 2.4 1/10 of mean 0 1 2 3 4 5 Poorest Wealthiest 0 20 40 60 80 100 Quintile Quintile Rank 2002 2017 2002 2017 Source: Calculations based on the 2002/03 HBS and the 2017/18 CMS/HBS. Note: The left figure shows each quintile’s share of total consumption. The right figure ranks households according to their consumption level and plots each household’s consumption relative to mean consumption. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 61 Fewer households have consumption levels far If positive over a specified time period, it indicates the below the national average. In 2002, 15 percent of distribution is becoming more equal; if negative, less the population had consumption less than a fifth of equal. Lesotho’s shared prosperity premium was 1.9 the mean; in 2017, only 6 percent did – a significant percentage points between 2002 and 2017. decline (Figure 39b). In contrast, about 35 percent of the population exceeded mean consumption in both Lesotho is alone among its Southern Africa neighbors 2002 and 2017, suggesting little change in relative with healthy positive readings on both shared consumption at the top of the distribution. prosperity and the shared prosperity premium. The Global Database on Shared Prosperity (World Bank Consumption growth at the bottom boosted shared 2019a) contains estimates of shared prosperity in 93 prosperity between 2002 and 2017. The World Bank countries. Lesotho is one of only 14 with both shared aims to boost the growth of the poorest 40 percent in all prosperity and shared prosperity premiums above countries. In Lesotho, the annualized growth rate of the 1.5 (Figure 40b). It is also the only country among its bottom 40 percent – also called shared prosperity – was neighbors in this group. In Namibia, consumption grew 2.2 percent between 2002 and 2017 (Figure 40a). This is slower for the bottom 40 percent than the mean. In much faster than the mean growth of 0.3 percent. The Botswana, the consumption of the bottom 40 percent difference between the growth of the bottom 40 percent hardly grew but it still did better than the mean, which and the mean is called the shared prosperity premium. saw consumption fall. Figure 40: Shared prosperity in Lesotho, comparison to other countries (a) Shared prosperity (b) Shared prosperity and shared prosperity premium 10 10 Namibia 5 5 Shared prosperity (pct.) Shared prosperity (pct.) Lesotho Lesotho Mozambique Botswana 0 0 South Africa −5 −5 −10 −10 Countries ordered by shared prosperity −4 −2 0 2 4 Shared prosperity premium (pct. points) Source: Calculations based on the 2002/03 HBS and the 2017/18 CMS/HBS and the Global Database on Shared Prosperity (World Bank 2019a). Note: The left figure ranks countries according to their shared prosperity estimates. The right figure plots shared prosperity and shared prosperity premiums. 62 Lesotho Poverty Assessment I Progress and challenges in reducing poverty B. Inequality fell but Lesotho remains an All regions experienced substantial declines in unequal country inequality levels between 2002 and 2017. In Maseru Urban, for example, the Gini coefficient fell from 52.4 As a result of faster growth at the bottom of the to 40.9. In 2002, some regions had higher inequality consumption distribution, Lesotho saw a decline levels than the country as a whole; in 2017, all regions in its Gini coefficient of consumption per adult had substantially lower inequality than the country as a equivalent. In 2002, Lesotho was one of the few whole. This implies that today’s gaps are more a result countries in the world with a Gini coefficient above of inequality between regions – for example, between 50. Since then, the pro-poor growth has induced a fall urban and rural areas. Hence, the decline in the national in inequality, and the coefficient stood at 44.6 in 2017 Gini coefficient cannot be attributed to convergence (Figure 41). in income levels across regions. In fact, the widening urban-rural divide that was apparent in the growth incidence curves has offset some of the decline in the inequality. Figure 41: Gini coefficient 2002 and 2017, national and by region 60 55 52.6 53.1 51.9 52.4 50.7 49.9 50 48.9 Gini coefficient 45 44.6 41.8 41.8 40 40.9 41 40.5 39.5 35 LesothoMaseru Other Rural Rural Rural Rural Urban Urban Low− Foot− Moun− Senqu lands hills tains River Valley 2002 2017 Source: Calculations based on the 2002/03 HBS and the 2017/18 CMS/HBS. Note: The figure shows the Gini coefficient nationally and across regions in 2002 and in 2017. Lesotho is now the least unequal country among its 42a). This is a change from the early 2000s, when neighbors but remains to one of the 20 percent most inequality in Lesotho was more on par the rest of the unequal countries in the world. The Southern Africa region. Despite this progress, Lesotho remains among region is the world’s most unequal, with Gini coefficients the 20 percent most unequal countries worldwide. This above 50 in Eswatini, Botswana, Mozambique and means more than 80 percent of countries for which we Namibia – and astonishingly above 60 in South Africa. have household surveys have lower rates of inequality Lesotho is the only country in the Southern African (Figure 42b). region with a Gini coefficient well below 50 (Figure Lesotho Poverty Assessment I Progress and challenges in reducing poverty 63 Figure 42: Regional and global inequality comparisons (a) Regional comparison of Gini coefficients (b) Global comparison of Gini coefficients 65 65 63 63 60 60 Lesotho, 2002 59 Lesotho, 2017 Gini coefficient 58 Gini coefficient 55 54 40 53 52 53 50 51 20 47 45 45 40 0 Countries ordered from lowest to highest Gini Lesotho Eswatini Botswana Mozam− Namibia South bique Africa Around 2002 Latest estimate Source: PovcalNet and the 2017/18 CMS/HBS. Note: The left figure ranks neighboring countries according to their latest Gini estimate, while also showing the Gini coefficients for a survey around 2002. The right figure ranks all countries according to their Gini coefficient as measured in the latest available poverty survey. C. Inequality of asset ownership and income fell as well The robustness of the inequality decline in not possible to compute the growth in their asset consumption can be assessed by looking at ownership. Among those whose asset growth can be inequalities in income and assets. Income computed, the poorest individuals had the largest gains. aggregates that show each household’s total monthly Median households tripled their number of assets net income broken down into six components have between 2002 and 2017, but the number of assets of been constructed for both the 2002 and 2017 surveys. the wealthiest households increased by less than 50 The components are formal wage income, agricultural percent (Figure 43a). income, self-employment income, social assistance, remittances/transfers and other income. Agricultural Incomes also grew most at the bottom of the and self-employment income are net of operating distribution. The bottom 20 percents’ incomes grew costs and include the market value of own-produced many times faster than the top half (Figure 43b). In consumption. In addition, asset ownership indices were contrast to the growth-incidence curve of consumption, calculated based on the total number of 16 assets the however, all households increased their income – all by households own. Ownership of these 16 assets were at least 5 percent annually. Why the discrepancy? The comparable across the two surveys. inflation used for the income aggregate (the growth in the consumer price index) is different from the inflation The growth-incidence curve for assets supports used for the consumption aggregate (the growth in the pattern observed with consumption: the the poverty line). The two measures increased at very distribution’s bottom has grown the fastest. The different rates over the 2002-2017 period. bottom 30 percent had no assets in 2002, so it is 64 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Income inequalities are driven by large wage Lesotho tend to have both better education and health discrepancies. Income inequality can be broken down outcomes, this need not be translated through the by how much each income component added to total labor market. There is relatively high representation income inequality. Doing so reveals that nearly two- of women in government and the government has thirds of all income inequality can be attributed to introduced policy and legislative changes supporting differences in formal wage income in 2017 (Figure gender equality; however, the effectiveness of these 44). This is partly driven by the high public-private measures in bringing about positive changes in the lives wage gaps. Self-employment income contributed 30 of women has been limited. Deeply entrenched social percent to income inequality, while remittances, social norms and stereotypes that negatively impact women protection and agricultural income were all pro-poor and are dominant and reinforced by the limited awareness of thus hardly contributing to inequalities. new laws, low capacity and inconsistency in application. As a result, the opportunities for women to participate The fact that consumption, income and assets on equal footing in the economic life of the country are measured at the household level may mask remain limited. For more information, see Box 8. important gender inequalities. Although women in Figure 43: Growth-incidence curves in assets and income (a) Assets (b) Income 30 300 Annualized growth in income (%) Growth in number of assets (%) 25 20 200 15 100 10 5 0 0 20 40 60 80 100 0 20 40 60 80 100 Rank in asset distribution Rank in income distribution Source: Calculations based on the 2002/03 HBS and the 2017/18 CMS/HBS. Note: The left figure shows the growth in number of assets while the right figure shows the annualized growth in income by percentile. Growth in assets is not defined below the 30th percentile because 30 percent of individuals in 2002 lived in households with no assets. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 65 Figure 44: Share of income inequality due to different components 80 Share of total income inequality (%) 66.2 60 40 28.3 20 2.4 0.5 0.2 2.5 0 rs e e e n e om m fe m io om co ns ct co nc te nc in in tra ti ro li e er s/ en lp ra ag ce th tu ym cia w O an ul al So plo r ic itt rm m em Ag Fo Re lf- Se Source: Calculations based on the 2017/18 CMS/HBS. Box 8: Gender equality in Lesotho Lesotho’s Constitution provides for equality and justice to males and females and provides for fair wages and equal remuneration for work of equal value. The current law on land matters is gender neutral in its provisions. The Legal Capacity of Married Persons Act (LCMPA), along with other progressive 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 in theory better equipped to play more productive and active roles as economic and social agents. However, there is anecdotal evidence that the implementation of the LCMPA has been problematic, perhaps because it came from a top-down political act in a cultural context still widely seen as patriarchal. Females outperform males in terms of education levels in Lesotho, but earnings are systematically lower. The 2017/18 CMS/HBS data suggest that net enrollment in primary school (grades 1 to 7) for girls was 92.4 percent in 2017, while boys were at 88.4 percent. Net enrollment in secondary school was 51.9 percent for girls and 31 percent for boys. The female-to-male completion ratio in primary school was 127 percent in 2014. However, some schools have reintroduced fees to cope with insufficient funding, leading to disparities in enrollment rates between poor and rich wealth quintiles, especially for boys. Female represent 54 percent of the literate population in Lesotho. Females averaged 5.9 years of schooling in 2018, compared to 4.9 years for males. Access to the internet was 57 percent for females and 43.4 for males. Poverty among rural women in Lesotho intersects with other severe disadvantages they face in high rates of HIV/ AIDS, maternal mortality and gender-based violence (GBV). While Lesotho has the second highest prevalence of HIV/AIDs globally, adolescent girls and young women between the ages of 15 to 24 are three times more likely to be infected than males of the same age group.12 In fact, 49 percent of the poorest women and 52 percent of 66 Lesotho Poverty Assessment I Progress and challenges in reducing poverty all women aged 15-49 believed a woman is not justified in refusing to have sex with her husband if she knows he has sex with other women (World Bank 2016b). Moreover, women also take on the full-time responsibilities of staying home to care for HIV-ailing elders. The high birth rate among young women is particularly critical because this group also experiences the highest maternal mortality rate.13 Moreover, 86 percent of women have experienced a form of GBV at least once in their lifetime, and research indicates that GBV is likely common but goes unreported. These are among the leading factors in the exclusion of women from the labor market and their low rates of participation in the economy.14 While Lesotho continues to have one of the highest rates of maternal mortality in the world, significant progress has been made over the past decade with the opening of the Queen Mamohato Hospital in 2015. Access to delivery care has also been improving. The DHS 2014 shows sustained improvement in maternity care. The proportion of women with babies delivered by a health professional (nurse or doctor) rose from 61.5 percent in 2009 to 77.9 percent in 2014. The proportion of women who delivered in a health facility increased from 58.7 percent in 2009 to 76.5 percent in 2014. The proportion of women who received antenatal care from a health professional improved from 91.8 percent 2009 to 95.2 percent in 2014. Attempts to bolster the status of women have had limited impacts. Lesotho has a relatively high representation of women in government,15 and the government has introduced policy and legislative changes that support ending violence against women, income equality and more generally the equal status of women.16 The effectiveness of these measures in bringing about positive changes in the lives of women has been limited. Deeply entrenched social norms and stereotypes negatively impacting women are dominant and reinforced by limited awareness of new laws, low capacity and inconsistency in application. As a result, the opportunities for women to participate on equal footing in the economic life of the country remain limited. For instance, the government introduced The Land Act 2010 to improve gender equality in land allocation and inheritance. Among other measures, it introduced provisions for inheritance of property by widows, joint titling of property and equal inheritance rights for both married and unmarried women. However, customary law prevails where women are regarded as minors,17 leading to many restrictions on their ability to access and make decisions around land and contributing to their inability to access to credit and loans.18 12 Latest data available is from 2016. Gender Data Portal, World Bank. 13 In fact, among young women (aged 20-24), 14 percent gave birth by age 18. There are also high rates of adolescent fertility at 89 births per 1,000 in women between ages 15-19. 14 This is also visible in a women’s unemployment rate of 32.1 percent, far exceeding the 21.3 percent for men. (World Bank 2016b). 15 Representation of women in Lesotho’s government is high compared to other Sub-Sharan countries. In 2016, the proportion of seats held by women in national parliaments was 25 percent, and the proportion of women in ministerial level positions was 18.2 percent (World Bank Gender Data Portal). 16 See UN Women Lesotho Country page. 17 In 2006, the Legal Capacity of Married Persons Act repealed customary laws that give husbands legal authority over wives and/or their property. 18 In fact, land is linked to key development challenges in Lesotho, i.e. HIV, malnutrition and violence against women. The UN HRC and UN Women find that in the context of HIV, women’s rights to inheritance and property are “… a crucial factor in reducing women’s vulnerability to violence and HIV, as well as empowering women to cope with the social and economic impact of the epidemic at the household level.” Evidence also suggests that countries where women lack landownership rights have 60 percent more malnourished children and the lack of access to credit correlates with 85 percent more malnourished children (United Nations 2013). Lesotho Poverty Assessment I Progress and challenges in reducing poverty 67 D. Inequality of opportunity is high Nearly half of all inequality is due to individuals’ Figure 45a illustrates the inequality of opportunity, region of birth, human capital achievements showing everyone’s predicted consumption based during childhood, serious health problems and on factors beyond individual control, ordered by environmental shocks. Nearly half of all inequality – 46 these predictions. If these factors did not matter, percent – can be attributed to relatively small number of all households would have the same predicted factors beyond individuals’ control (see Box 9 for more consumption – i.e., there would be no gradient. To the details on how this is measured). This remarkably high contrary, some individuals can expect to have five times number is higher than the 10 other countries in Sub- the consumption of others based solely on where they Saharan Africa with available estimates (Beegle et al. grew up and other factors outside of their control. 2016). Box 9: Inequality of opportunity and intergenerational mobility Inequality of opportunity Inequality could stem from the fact that individuals in wealthier households induce more effort or because they had access to human, social, and financial capital that gave them a head start in life. The latter part is often considered more problematic because it is beyond individuals’ control. It is frequently dubbed inequality of opportunity, and it can be expressed as a fraction of total inequality (Roemer, 1998). The higher this fraction, the more inequality is driven by sources beyond individual control. The share of inequality attributable to inherited opportunities can be estimated using machine-learning methods. Such methods use a range of variables that plausibly lie beyond individual control to predict individuals’ consumption. Examples of these variables include the districts where individuals grew up, human capital acquired during childhood, serious health problems and environmental shocks. The more these variables are able to predict consumption levels, the more inequality of opportunity is present.19 In this report, a method called conditional inference random forests is used to make such predictions (Brunori et al, 2018). Intergenerational mobility To put the unfair inequalities in a global context, Lesotho can be compared to other countries on intergenerational mobility. Narayan et al. (2018) contains estimates of absolute and relative mobility in education for more than 140 countries, looking at individuals born in the 1980s or later who have completed their schooling. One measure of absolute mobility in education is the share of individuals who have strictly more education than their parents. A measure of relative mobility in education is (one minus) the rank correlation between parents and children’s education levels in a cohort. The higher the correlation, the more parents’ rank in society predicts their children’s rank and the lower the relative mobility. When relative mobility is higher, some children are, on expectation, off to a disadvantaged start even before they are born. 19 Health shocks and environmental shocks are not necessarily fully beyond individuals’ control because behavior and actions can lower the risk of certain shocks and mitigate the consequences when they occur. This could mean that the share of inequality due to factors beyond individual control is overestimated. However, this is unlikely to be the case in practice because many other factors beyond individual control are not included in the analysis due to a lack of data – such as the quality of the education received. 68 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Figure 45: Inequality of opportunity and its drivers (a) Predicted welfare based on factors beyond (b) Relative importance of factors determining individual control individuals’ opportunities 3000 1.0 Importance for welfare 0.8 0.6 Predicted welfare (Maloti per month) 0.4 0.2 2000 0.0 Death of household member Difficulties seeing Age Diabates Death of livestock High blood pressure Rise in food prices Drought Any short-term illness Any chronic illness Birth location Gender HIV/Aids Year of preschool Read and write abilities Years of schooling 1000 Line of equality Demographics Health Education Shocks 0 Households ordered from lowest to highest predicted welfare Source: Calculations based on the 2017/18 CMS/HBS. Note: The left figure shows individuals predicted consumption based on circumstances beyond individual control. The right figure shows the relative importance of the various factors beyond individuals’ control in terms of how much they predict individuals’ consumption level. The importance is scaled by assigning the most important variable a value of 1. Differences in educational achievement during We can trace how early inequities occur by looking childhood are the most important contributor to at intergenerational mobility, a concept related unequal opportunities. In this context, the number to inequality of opportunity. Two distinct concepts of years of schooling individuals receive prior to the of intergenerational mobility exist: Absolute mobility start of high school is an important indicator. (Figure measures whether children are doing better than their 45b). Furthermore, the number of years of preschool parents, while relative mobility measures the strength education predicts individuals’ consumption and thus of the connection between parents’ outcomes and generates unfair inequities – as do individuals’ birth children’s outcomes. The latter is closely connected location, age and recent exposure to drought. The fact to the notion that all individuals ought to have equal that the number of years in preschool matters suggests opportunities, no matter their background. inequality begins early in life, so a viable path to inequality reduction must ensure that all children complete basic education levels. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 69 In a global context, Lesotho is in the bottom quartile the bottom of the distribution to climb to the top. When in terms of relative intergenerational mobility (see it comes to absolute mobility, Lesotho looks somewhat Box 9 for more details on how this is measured). 20 better. Half of all Basotho have strictly more education The correlation between parents’ years of schooling than the average of their parents, and a much greater and their children’s years of schooling in Lesotho fraction have at least as much education as their parents is 0.54, which is quite high when compared to other (Figure 46b). As a result, many children surpass their countries (Figure 46a). This means that parents’ level parents’ living standards, but this happens across the of education to a large extent predicts their children’s distribution, so most fail to bypass others within their level of education, making it difficult for people born in generation along the way. Figure 46: Intergenerational mobility in Lesotho (a) Relative intergenerational mobility (b) Absolute intergenerational mobility Economies, ranked from highest to lowest Economies, ranked from highest to lowest Lesotho Lesotho 0 .2 .4 .6 .8 0 .2 .4 .6 .8 1 1 − rank correlation between parent and child’s level of schooling Share of children with strictly more education than their parents Source: Narayan et al. (2018) and the 2017/18 CMS/HBS. Note: The right figure ranks countries according to their level of absolute mobility; i.e., the share of individuals that have strictly more education than their parents. The left figure ranks countries according to their level of relative mobility, measured as (one minus) the rank correlation between parents and children’s levels of education in a cohort. 20 Estimating intergenerational mobility requires information from two generations. Ideally, this is obtained through long panels, or alternatively by respondents reporting the educational attainment for their parents. Neither is available for Lesotho. Instead, respondents that reside in the same household as their parents are relied upon. Narayan et al. (2018) show that using only co-residents yield similar estimates to relying on panels or retrospective questions. 70 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Part II: Drivers and challenges of poverty reduction in Lesotho 71 Chapter II.1: Factors associated with poverty reduction The service sector, productivity improvements and demographic changes contributed to economic growth between 2000 and 2017. Despite moderate growth rates, the fall in poverty was driven by reduced inequalities rather than growth. The poverty and inequality declines can be explained by expansion in social protection, improvements in formal wages associated with improved education levels and a demographic dividend. Poverty and inequality reduction slowed because of the negative impacts of weather shocks and falling remittances – both more pertinent to the rural population. Lesotho’s growth and poverty reduction prospects are severely impaired by a number of vulnerabilities. Human capital and well-functioning labor markets are keys to further reductions in poverty and inequality. A. Drivers of Lesotho’s economic growth 4 percent, and GDP per capita increased at an average annual rate of 2.7 percent. However, growth slowed to Lesotho is a lower middle-income country with a 1.5 percent in 2018. Although an average growth rate of moderate growth rate. Lesotho had a gross national 4 percent is relatively high compared to other countries income per capita of US$1,380 in 2018. Government in the region, the growth rate falls short of the National spending accounted for 50 percent of GDP. From 2000- Strategic Development Plan (NSDP) target rate of 7 2017, Lesotho’s GDP grew at an average annual rate of percent. Figure 47: GDP and private consumption, Figure 48: Growth decomposition, annualized, 2002-2018 2000-2017 180 3 2.7 Index, 2002 = 100 160 140 2 Percent 120 1 0.6 100 0 80 ! -0.3 60 -1 -0.6 20 2 20 4 20 6 20 8 20 0 20 2 20 4 20 6 18 0 0 0 0 1 1 1 1 Productivity Employment rate 20 GDP Private consumption Participation rate Demographic change Source: Authors’ calculations based on WDI and IMF databases. Note: Private consumption includes households final consumption expenditures and non-profit institutions serving households final consumption expenditures, converted to index with 2002 = 100. 72 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Private consumption growth was significantly importance of agriculture. From the expenditure side, lower than GDP growth, suggesting a low GDP demographic changes and productivity improvements to poverty elasticity. As presented in Figure 47, were the main contributors to GDP growth between private consumption increased by approximately 25 2000 and 2017, while stagnation in employment and percent between 2002 and 2018, while GDP grew labor force participation had negative impacts on growth by 67 percent. Private consumption’s much slower (Figure 48). On the production side, the services sector growth compared to other GDP components suggests has been the most significant contributor to growth. a relatively slow speed of poverty reduction relative The contributions of the agriculture and mining sectors to GDP growth. Despite continuing economic growth, have been very volatile due to weather shocks and private consumption has fallen since 2012. fluctuations in commodity prices. The services sector grew fast in Lesotho while value added in agriculture Structural transformation occurred in Lesotho, remained low (Figure 49 and Figure 50). This trend is characterized by a rapid increase in value added and especially pronounced since 2010. employment in the services sector and a decline in the Figure 49: Employment by major sector, Figure 50: Growth and value added in Lesotho by 1990 – 2017 sectors, 1990-2017 700 4 000 10 million constant 2010 USD %, annual growth rate 600 Thousand people 500 3 000 5 400 300 2 000 200 0 1 000 100 0 0 -5 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 1990 1993 1996 1999 2002 2005 2008 2011 2014 2017 Agriculture Industry Services Services Industry Agriculture Value Added Growth (RHS) Source: Authors’ calculations based on WDI and IMF databases. Lesotho’s modest economic growth is under threat implying the government will find it hard to further because the country’s macro-fiscal outlook is weak stimulate growth. In addition, the majority of the rain- in the near-to-medium term (World Bank 2018e). fed dependent rural population has been exposed to This stems largely from the country’s exposure to the droughts. SACU revenues have declined from 25 percent sluggish South African economy, climate-related shocks of GDP in FY2014/15 to 13.6 percent in FY2016/17— and rising public debt. Lesotho’s economic performance mostly due to South Africa’s slower growth. For Lesotho, is dependent on South Africa, and the country’s poor the decline in SACU revenues narrowed its fiscal space near-to-medium term economic outlook is expected to and led to a considerable decline in fiscal buffers. The have a negative impact on Lesotho’s growth trajectory. country’s fiscal deficit reached a record 9.5 percent of This will restrict Lesotho’s already limited fiscal space, GDP in 2016 before receding to 3.5 percent in 2018. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 73 B. Inequality drove the decline in poverty when using the poverty gap to make the decomposition, suggesting that the very bottom of the distribution saw About three-quarters of the poverty decline can be significant gains in living standards between 2002 and attributed to distributional changes, while growth 2017. Growth’s relatively small role is surprising given can explain only a quarter of the decline in poverty. Lesotho’s high average growth rates from 2002 to 2017. Figure 51 breaks down the change in the poverty rate One possible explanation for the small contribution is between 2002 and 2017 by whether it was driven by that growth only benefited the top of the distribution. growth in mean consumption per adult equivalent or by However, this is not the case, given that consumption a more equitable distribution of consumption.21 Using was more evenly distributed in 2017 than in 2002. the food poverty line, which captures extreme poverty, Instead, only a relatively small share of real GDP growth 70 percent of poverty reduction came from distributional per capita was passed through to consumption. changes rather than growth. The picture is reinforced Figure 51: Decomposing changes in poverty into growth and distributional change Food poverty line National poverty line Contribution to poverty reduction (%) Contribution to poverty reduction (%) 100 100 19 17 30 24 80 80 60 60 81 83 40 40 70 76 20 20 0 0 Poverty rate Poverty gap Poverty rate Poverty gap Growth Distributional change Source: Calculations based on the 2002/03 HBS and the 2017/18 CMS/HBS. Growth may have failed to reduce poverty increase in share of GDP was financial and insurance significantly because it went to the wrong sectors. activities, which may not trickle down to the rural poor. As shown in the previous subsection, final household In contrast, the sectors the poor tend to rely on, such consumption expenditures’ share of total GDP as agriculture and textiles, saw stagnation or decline in decreased markedly between 2002 to 2017, suggesting their shares of total GDP. These observations points to that growth in household consumption may not have challenges of converting income growth into sustainable mirrored growth in GDP. The sector that saw the largest poverty reduction. For sustained reductions in poverty, 21 There are several ways to decompose the changes in poverty and understand what drove its decline between 2002 and 2017. Since the decline was relatively modest for such a long time span—a decrease from 56.6 percent to 49.7 percent—it is also likely that some factors triggered increases in poverty. This chapter uses decomposition techniques to break down the changes in poverty. Many techniques describe what poverty would have been in the most recent data if one aspect from the earlier time had remained unchanged. Such a counterfactual scenario allows for an isolated assessment of how this one factor contributed to the changes in poverty. However, these techniques cannot identify the causal drivers of poverty, and they paint only a broad picture of the correlations involved. They serve to motivate a deeper dive into certain areas of Lesotho’s economy. The first decomposition asks what poverty would look like with the new mean consumption level but the old distribution and vice-versa (Datt & Ravallion 1992). This methodology creates hypothetical scenarios of what poverty would have been had the distribution not changed or had the mean consumption level not changed. 74 Lesotho Poverty Assessment I Progress and challenges in reducing poverty it is vital that growth in GDP per capita translates into between 2002 and 2017. To consume, households increased consumption. This also matters for future need to engage in agriculture or enterprises, or receive poverty reduction (Box 10). income through transfers or wages. In other words, there is a close relationship between household The fact that inequality was behind a large part of income (including the value of consumption from own poverty reduction means that fully understanding production) and aggregate consumption. Looking closer why poverty fell requires exploring what happened at the components of aggregate income can yield to inequality. To this end, it is important to look at lessons on why poverty and inequality fell and suggest how the components of household income changed ways to spur further reductions. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 75 Box 10: The role of inequality and growth in poverty trends towards 2030 Based on assumptions about growth in GDP per capita and the distributional nature of that growth, poverty in Lesotho can be projected to 2030. Poverty projections should always be analyzed with caution due to the many factors and uncertainties that shape the future. Yet, stylized scenarios about future developments may sometimes shed light on how poverty could evolve and the kinds of policies that could spur such developments. If real increases in GDP per capita match the World Economic Outlook (WEO) forecasts of about 1 percent per year, Lesotho’s poverty rate would decline to only 45 percent by 2030 (Figure 52a). This modest decrease would be insufficient to meet the first Sustainable Development Goal, which calls for halving the poverty rate by 2030. More optimistic scenarios are plausible if growth rates accelerate or inequities diminish. Higher than expected growth rates could substantially reduce poverty. Policies to spur growth, such as reforms to make labor markets more efficient or increase educational achievements, could help accelerate poverty reduction. If the annual growth rate were 2 percentage points above the forecasts, suggesting annual growth rates around 3 percent, then poverty might decline to 38 percent by 2030 (Figure 52b). Poverty could fall substantially if inequalities are reduced. The previous projections assumed growth would accrue to everyone equally. But suppose growth lowers the Gini coefficient by 2 percent annually, which is realistic in historical perspective (Lakner et al. 2019). Such a change could happen with targeted social-protection programs and policies to mitigate the consequences of shocks (particularly environmental ones) on vulnerable populations. Reducing the Gini coefficient by 2 percent a year through 2030 would bring the poverty rate down to 33 percent. Figure 52: Poverty projections to 2030 (a) Baseline scenario (b) Optimistic scenarios 60 60 55 55 50 Poverty rate (%) 50 Poverty rate (%) 45.3 45 45 40 37.6 40 35 35 33.4 30 2017 2020 2025 2030 30 2017 2020 2025 2030 2% annual fall in Gini 2 pct. point higher annual growth Source: Calculations based on the 2017/18 CMS/HBS and the WEO, Spring 2019. Note: The left figure shows poverty projections to 2030 if GDP/capita grows according to WEO forecasts and inequality remains unchanged. The right figure shows poverty projections to 2030 if GDP/capita exceeds WEO forecasts by 2 percentage points per year, or if the Gini coefficient decrease by 2 percent per year using a linear growth-incidence curve to change inequality. Both figures assume that 75 percent of growth is passed through to consumption, a figure close to the Sub-Saharan Africa average. 76 Lesotho Poverty Assessment I Progress and challenges in reducing poverty C. Increases in formal wage incomes and poverty. Rural poverty would be nearly 10 percentage expansion of social protection reduced points lower if remittances and transfers had remained inequality and poverty at 2002 levels, while urban poverty would have been 2.7 percentage points lower. Increased wages and social assistance were the primary drivers of poverty reduction.22 Since 2002, Stalling agricultural incomes increased rural poverty the shares of formal wage income and social assistance and inequality. The increase in poverty rates in rural have increased considerably at the bottom of the districts in recent years is due in part to weather related distribution. Wage income was only a quarter of total shocks, especially the 2015/16 El Niño that impacted incomes in 2002, but it accounted for nearly half of all agricultural households the most. Because of this shock, incomes in 2017. Likewise, social assistance, particularly falling agricultural incomes led to a 4.1 percentage point at the bottom of the distribution, increased tremendously increase in rural poverty. Combined with an increase in during this period. An increase in wage income led to a consumption in urban areas, which were already better 3.3 percentage point decline in rural poverty and a 2.2 off than the rural areas before the shock, the fall in percentage point decline in urban poverty (Figure 53). agricultural incomes induced a 1.8-point increase in the The increase in wages in rural areas could be explained Gini coefficient. by the presence of local governments and NGOs. Increased wage income also led to a 3.1-point decline in A demographic dividend associated with reductions the Gini coefficient. Social assistance drove down rural in average household size and dependency ratios poverty by 4.3 percentage points and urban poverty by supported poverty reduction in Lesotho. The 1.3 percentage points. The relative importance of social demographic dividend refers to the positive impact on assistance on poverty reduction was greater below economic growth and incomes per capita that results the poverty gap, suggesting that social assistance from changes in the population’s age structure brought had a larger impact on the very poorest. Indeed, social on by declining fertility rates. Demographic changes protection drove down theGini coefficient by 2.6 points. in Lesotho have a long-term economic impact on the economy (Box 11). Dependency ratios have fallen Remittances declined significantly between 2002 consistently since 1950, including a decline from 41.8 and 2017. Remittances and transfers have fallen nearly in 2002 to 40.3 in 2017. Over the same period, average 50 percent since 2002, when many Basotho households household size fell from 6.2 to 5.3 people, resulting in depended on money from abroad. Since remittances 5.7 percentage point decline in urban poverty and a 1.6 tend to be pro-poor, this decline, all else equal, has been percentage point decline in rural poverty (Figure 53a). associated with increases in the Gini coefficient and 22 To decompose changes in poverty, Azevedo et al. (2013) developed a method of linking changes in income aggregates to changes in poverty. Intuitively, their method tries to estimate what poverty would have looked like in 2017 if an income component had not changed from 2002 and 2017. The method then utilizes the relationship between a household’s income and consumption to estimate what poverty would be with this counterfactual income component. The difference between the counterfactual and actual poverty rates can be used to gauge each income component’s impact on poverty. The method also indicates whether the changes in poverty are driven by changes in household size. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 77 Figure 53: Decomposing changes in poverty and inequality into income components (a) Decomposition of poverty (b) Decomposition of inequality Percentage points change in poverty Percentage points change in the Gini coefficient -10 -5 0 5 10 -10 -5 0 5 10 Household size -1.6 -5.7 Household size-8 -3.3 Factors Factors Formal wages -2.2 that Formal wages -3.1 that reduced reduced Self-employment income -2.3 poverty -5.1 Self-employment income -0.6 Gini Social protection income -4.3 -1.3 Social protection income -2.6 Factors 4.1 Agricultural income that 0.8 Agricultural income Factors 1.8 increased that Remittances poverty 9.5 2.7 Remittances increased 5.4 Gini Other income 4.3 4.8 Other income 1.3 Rural Urban Source: Calculations based on the 2002/03 HBS and the 2017/18 CMS/HBS. Note: There is also a residual effect, not displayed in the figure. It is related to the relationship between the consumption aggregate and the income aggregate. 78 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Box 11: Demographic dividend opportunity for Lesotho Changes in mortality and fertility rates decrease dependency ratios, resulting in a demographic dividend. One of the main potential boons a country receives from a demographic transition is an increased supply of working-age people. In Lesotho and other SACU countries, mortality and fertility are decreasing, and dependency ratios are falling (Figure 54, Bruni et al. 2016). The surge in the working-age population is accompanied by a relative fall in the number of young dependents. Because of this transition, Lesotho enjoys a window of demographic opportunity in which the dependency ratio is historically low and the working-age population’s proportion is historically high. Having a larger share of the population at work can potentially lead to increases in output per capita, thus the demographic dividend. Figure 54: Total dependency ratios for SACU countries 120 100 Dependency ratios 80 60 40 20 0 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050 Botswana Eswatini Lesotho Namibia South Africa Source: UNFP (https://population.un.org/wpp/Download/Standard/Population/). Notes: Dependency ratios calculated as the population less than 15 or above 64, divided by population between 15 and 64. If employed in productive jobs, this large workforce will allow the country to reap the benefits of the demographic dividend. Between now and 2050, Lesotho’s working-age population will increase by 36 percent. Currently, people between the ages of 13 and 25 are the largest population cohort. To generate a demographic dividend as they move into working age, these people will need labor-market relevant education and skills and the private sector will need the capacity to absorb them. In Lesotho, however, key skills remain in short supply and entrepreneurialism is still nascent. Demographic dividends increase labor supply depending on the ability of the economy to absorb extra workers. Decreases in fertility rates also result in healthier women and fewer economic pressures at home. This also allows parents to invest more resources per child, leading to better health and educational outcomes. Improvements in the Rural Lowlands and and the latter the “intra-area effect.” The population-shift urbanization contributed to a large decline in the effect drove the poverty rate down by 2.6 percentage national poverty rate. The kind of decline in poverty points, mostly due to more households residing in Lesotho experienced between 2002 and 2017 could urban centers. The intra-area effect drove poverty down occur because of urbanization – households moving to by 3.7 percentage points – but it was behind a larger areas with lower poverty levels – or because certain part of the decline in food poverty. Breaking it down by areas decreased their poverty rates. Figure 55 presents geographical areas shows that the largest impact came the changes in poverty decomposed into these two from the Rural Lowlands. Although its poverty rate channels, the former called the “population-shift effect” did not decrease as fast as the urban areas, the Rural Lesotho Poverty Assessment I Progress and challenges in reducing poverty 79 Lowlands, larger population share meant a larger impact points. This suggests the urban-rural divide is a bit more on poverty reduction. In contrast, the Rural Mountains’ complicated and rural areas differ markedly from each poverty rate increased by more than 2 percentage other. Figure 55: Decomposing change in poverty into population-shift and intra-area effect, 2002–2017 Total change in poverty Decomposing intra−area effect 2 2 0 0 Percentage points Percentage points −3.7 −2 −2 −4 −0.7 −8.0 −4 −2.6 −6 −6 −8 −8 −0.0 −2.0 −10 −10 National poverty line Food poverty line Rural Other Maseru Rural Rural Senqu Rural Lowlands Urban Urban Foothills River Valley Mountains Intra−area effect Interaction effect Population−shift effect National poverty line Food poverty line Source: Calculations based on the 2002/03 HBS and the 2017/18 CMS/HBS. D. Increase in educational attainment drove changes in poverty by whether they are driven by down poverty changes in household characteristics and the return to these characteristics. Results are shown in Figure 56. To better understand the factors associated with increasing incomes and declining poverty, Changes in endowments drove down the poverty it is important to analyze whether the changes rate. Poverty fell by more than 5 percentage points were driven by gains in individuals’ endowments due to changes in endowments, while returns to (characteristics of the population) or improving endowments moderated the reduction in poverty returns to endowments. The former could occur (Figure 56a). Taking a closer look at the endowments, if, for example, individuals obtained more education, increasing education levels led to a 2.4 percentage while the latter could occur if the return to a secondary point decline in poverty. Both a reduction of individuals degree increased (in terms of reducing the probability of with no education and an increase of individuals with being poor). This aspect is analyzed using the Oaxaca- higher education played a role. Smaller households also Blinder decomposition method (Blinder 1973; Oaxaca contributed a significant decline in poverty. Changes in 1973). The decomposition shows whether changes in the composition of the labor force had nearly no impact poverty were driven by increases in education levels on poverty reduction because the unemployment (endowments) or by the returns to education (return rate remained very high. Changes in the location of to endowments). It similarly indicates whether poverty households – when accounting for all the other factors – was driven by changes in the labor force status or by also had little impact on poverty reduction. the returns to being employed. Finally, it breaks down 80 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Figure 56: Decomposing change in poverty into endowments and returns, 2002–2017 (a) Impact of changes in endowments and returns (b) Breaking down the effects from endowments 4% 4% Contribution to poverty reduction Contribution to poverty reduction 2% 2% 0.55% 0.3% 0% 0% - 0.2% -2% -2% - 2.4% -4% -4% - 3.1% -5.41% -6% Household size -6% Education Labor force Location Endowments Returns Source: Calculations based on the 2002/03 HBS and the 2017/18 CMS/HBS. E. Conclusion and policy discussion income and social protection were particularly beneficial for the poorest households, they had large positive This chapter used decomposition techniques to impacts on inequality. However, it is possible that the analyze the reduction in poverty between 2002 and impact of social protection on poverty reduction could 2017. It found that poverty reduction was driven by have been higher if poor rural households had been decreasing inequality, suggesting that understanding able to receive a greater share of benefits. The negative the drivers of poverty reduction requires understanding impact of environmental shocks in and the decline in changes in inequality. remittances slowed poverty and inequality reduction. Overall, the analyses suggest further reductions in Declining poverty and inequality can be explained Lesotho’s poverty and inequality are likely to come from by the expansion of social protection, improvements improvements in human capital and a well-functioning in formal wages associated with higher education labor market that creates decent jobs, especially for the levels and a demographic dividend. Since wage poor. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 81 Chapter II.2: The role of human capital on poverty reduction Compared to other countries at the same level of income per capita, Lesotho spends a relatively large share of its public resources on education and health. Even so, education and health outcomes are relatively poor and inequitable, and this has adverse impacts on labor productivity. Despite universal access to primary education, primary enrollments have been stagnant in recent decades and poor learning outcomes remain a challenge. Children from poor households face disproportionately high repetition and dropout rates at the primary level, and access to post-primary education remains low. Like education, health outcomes are poor. Many individuals, especially among those from poorer quintiles, do not seek care when needed, the affordability of medical care being one of the main reasons. Health expenditures pose a disproportionately large burden for poorer households. Improving access to quality health care and financial protection for all households, particularly the poor, is therefore important for poverty reduction because it will equalize opportunities and enable the poor to participate in and benefit from economic growth. A. Public spending on education and health Like education, spending on health is relatively high. is high but outcomes are relatively poor Government health expenditures as a percentage of the national budget has hovered around 11-12 percent, Lesotho allocates a relatively large amount of its below the Abuja Declaration commitment of 15 percent public resources on education and health. Around of the national budget. As a percentage of GDP, health 13.8 percent of the overall government budget in spending was around 6.2 percent in 2016 (Figure 58). 2018, equivalent to about 6.3 percent of GDP, went Every year, the health budget is mainly allocated to to education. As shown in Figure 57, this spending recurrent spending – from 77.7 percent in 2013/14 to level is high compared to other countries at the same 87.5 percent in 2017/18, with more than half going to level of income per capita. Lesotho has a higher public general administration. Development expenditures education expenditure-to-GDP ratio than South Africa. remain low, limiting the expansion of the country’s health In Lesotho, seven years of primary education are free infrastructure. This spending pattern fails to narrow the and compulsory. As a result, most of the education gap in health service delivery in remote areas, where budget goes to primary education – 56.4 percent in fiscal long distances and poor transportation options often year 2017/18. Next comes secondary education (28.5 represent additional barriers for the poorest. percent), followed by tertiary education (5.4 percent), technical and vocational education and training (2.1 percent) and early childhood education (0.3 percent). 82 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Figure 57: Government expenditures on education Figure 58: Government expenditures on health as percentage of GDP, 2018 as percentage of GDP, 2016 18 8 16 Government expenditure on Government expenditure 7 Lesotho 14 on health (% of GDP) education (% of GDP) 6 12 5 10 4 8 Lesotho 3 6 2 4 1 2 0 0 2.5 3 3.5 4 4.5 5 2.5 3 3.5 4 4.5 5 Log GDP/capita (2011 USD PPP) Log GDP/capita (2011 USD PPP)) Source: World Development Indicators (WDI). Despite progress, Lesotho faces high HIV/AIDS lower middle-income countries and 10 years lower than rates. The heavy burden of HIV/AIDS represents a the average for Sub-Saharan Africa (World Bank 2017). development challenge in terms of both morbidity and mortality. According to the 2014 Lesotho Demographic Lesotho has stagnant primary net enrollment and Health Survey (LDHS), the prevalence of HIV/ and poor education outcomes compared to other AIDS was 24.6 percent among people aged 15-49, and countries. As in the health domain, education outcomes it accounted for 41.4 percent of all deaths, despite a are poor relative to spending. Among the Southern and decline in the incidence. Owing partly to the prevalence Eastern Africa Consortium for Monitoring Educational and burdens of HIV and AIDS, infants, children, and Quality (SACMEQ) countries, Lesotho had the lowest women are disproportionately affected by poor health primary net enrollment rate in 2017 at 81 percent. outcomes, with their mortality rates among the highest Namibia, for example, was at 97 percent. In Lesotho, in the world. Stunting rates are among the highest in this net enrollment rate has remained almost stagnant the SACU region: Around 33 percent of children under at 80 percent since 2002, and the primary completion age 5 were stunted in 2014. Linked to high HIV and rate was only 65 percent in 2014.23 Lesotho had the third AIDS prevalence is high incidence of tuberculosis. The lowest score on reading and mathematics for Grade 6 2016 Global Tuberculosis Report reports that Lesotho’s students, just above Malawi and Zambia (Figure 59). tuberculosis incidence rate is the highest in the world at The country scored 468 on the reading tests, compared 788 cases per 100,000. In another link to the challenges to the average of 513 for the participating countries. In of HIV and AIDS, life expectancy at birth in Lesotho was mathematics, Lesotho scored 477, with the SACMEQ 50 years in 2014, 18 years lower than the average in average at 512. 23 UNESCO Institute of Statistics. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 83 Figure 59: SACMEQ III scores 700 600 SACMEQ score 500 400 300 200 100 0 Lesotho Botswana Mozambique Kenya South Africa Zambia Malawi Mauritius Tanzania Zimbabwe Namibia Swaziland Uganda Seychelles SACMEQ III (Reading) SACMEQ III (Mathematics) Source: UNESCO Institute of Statistics. A national assessment of junior secondary learners Access to primary education has stagnated, but carried out in 2018 also showed poor performance Lesotho has slowly increased secondary education for Grade 9 (Form B) students in mathematics coverage. The primary net enrollment increased from and science. The national mean scores were 76.7 59 percent in 1999 to 80 percent in 2000 – a result of for Sesotho, 51.6 for English, 32.4 for science and the introduction of Free Primary Education in 2000.24 26.3 for mathematics. There were considerable However, it has been stagnant around 80 percent in differences between subjects, indicating that students recent years. According to the 2016 Education Statistic face challenges with mathematics and science. Their Report, the primary gross enrollment rate is estimated performances in languages, especially Sesotho, were at 116 percent, driven by a large share of overage much higher. Teachers were also assessed and found pupils still in primary school because of high repetition to perform poorly in literacy and numeracy. rates. Based on the 2017/18 CMS/HBS, the proportion of overage pupils in Grade 1 was estimated at only 29 Participation in pre-primary education in Lesotho percent, and it increased by grade, reaching 59 percent remains very low. In 2017, the net enrollment rate in in Grade 6 and 55 percent in Grade 7. In contrast, lower pre-school was only 23 percent (Figure 60). Nonetheless, secondary net enrollment has slightly increased from 24 participation in pre-primary education increased with percent in 2010 to 30 percent in 2016 (Figure 61). age: 17 percent for boys and 18 percent for girls among children aged 3; for the children aged 5, the participation rate rose to 27 percent for girls and 26 percent for boys. In absolute terms, more girls than boys accessed school, but the difference was marginal. 24 UNESCO Institute of Statistics. 84 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Figure 60: Net enrollment rates by level, 2016 Figure 61: On-time, under-age and over-age in primary 100% Upper secondary 13% 29% 30% 80% 42% 40% 44% 59% 55% Lower secondary 30% 60% 29% 34% 25% 28% 40% 29% Primary 81% 21% 21% 20% 41% 36% 33% 31% 26% 20% 24% Pre-primary 23% 0% 1 2 3 4 5 6 7 Grade 0% 20% 40% 60% 80% 100% Under-age On time Over-age Source: UNESCO Institute of Statistics Source: Calculations based on the 2017/18 CMS/HBS. Poor education and health outcomes contribute that there is no way to determine the real staffing needs to Lesotho’s low score in the World Bank’s in the field. Unwarranted variations in student-teacher Human Capital Index (HCI). The index ranks ratios for schools with similar enrollments are also Lesotho 143rd among 157 countries (Figure 62). closely linked to lack of a coherent and consistent policy Children born in Lesotho today will be 37 percent as for appointing additional teachers or reducing teacher productive when they grow up as they could be if they numbers in schools when enrollment declines. The enjoyed complete education and full health. Lesotho is country could also benefit from more investments in performing poorer than the average for its region and its teacher training on the new curriculum, strong coaching income group.25 Children could expect to complete 8.7 and supervision mechanisms for teachers at the school years of pre-primary to secondary school by age 18 in level and continuous learning assessments. In general, 2017. However, when years of schooling are adjusted for low levels of human capital have adverse impacts on quality of learning, this was equivalent only to 5.5 years labor productivity. Investing in education and health is – a learning gap of 3.2 years. Poor planning and human therefore important for worker productivity (both in the resource management are some of the challenges in government and the private sector), economic growth the education sector (World Bank 2019g). The lack of an and poverty reduction in the long term. established/official list of teachers in schools has meant 25 The average HCI for Sub-Saharan Africa is 0.4. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 85 86 Human capital index 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 Nigeria Mauritania Côte d'Ivoire Angola Yemen, Rep. Les otho Sudan Papua New Guinea Pakistan Cameroon Zambia Eswatini Congo, Rep. Timor-Leste Ghana India Solomon Islands Lao PDR Guatemala Vanuatu Myanmar Bangladesh Kiribati Egypt, Arab Rep. Honduras Source: World Bank (2018b). Cambodia Morocco El Salvador Tunisia Kenya Nicaragua Tajikistan Indonesia Philippines West Bank and Gaza Kosovo Jordan Armenia Kyrgyz Republic Moldova Sri Lanka Georgia Figure 62: Human Capital Index, comparison to other lower middle-income countries Mongolia Ukraine Vietnam Lesotho Poverty Assessment I Progress and challenges in reducing poverty Box 12: Human Capital Index (HCI) for Lesotho The HCI measures the amount of human capital that children born today can expect to attain by age 18. HCI is made up of five indicators: the probability of survival to age 5, a child’s expected years of schooling, harmonized test scores as a measure of quality of learning, adult survival rate (the fraction of 15-year-olds that will survive to age 60) and the proportion of children who are not stunted. It conveys the productivity of the next generation of workers compared to a benchmark of complete education and full health. It is constructed for 157 countries. • Lesotho ranks 143 among 157 nations in the HCI index. Children born in Lesotho today will be 37 percent as productive when they grow up as they could be if they enjoyed complete education and full health. • In Lesotho, 91 of 100 children born in Lesotho survive to age 5. • Children who start school at age 4 can expect to complete 8.7 years of school by their 18th birthdays. • Students in Lesotho score 393 on a scale where 625 represents advanced attainment and 300 represents minimum attainment. • Factoring in what children actually learn, expected years of schooling is only 5.5 years. • Across Lesotho, 50 percent of 15-year-olds will survive until age 60. This statistic is a proxy for the range of fatal and non-fatal health outcomes that children born today would experience as adults under current conditions. • In Lesotho, 33 of 100 children are stunted and at risk of cognitive and physical limitations that can last a lifetime. Figure 63 presents ranking of countries on HCI components. It shows that Lesotho lags behind an average lower middle-income country as well an average Sub-Saharan African country with regard to most of the components. Figure 63: Human Capital Index components, cross-country comparison Probability Of Survival To Age 5 131 149 Expected Years Of Human Capital Index 131 130 School 143 72 129 74 76 74 32 78 156 100 83 115 Harmonized Test Adult Survival Rate 133 76 Scores Fraction Of Kids Under 5 Not Stunted Lesotho Lower middle income Sub-Saharan Africa Source: World Bank (2018b). Lesotho Poverty Assessment I Progress and challenges in reducing poverty 87 B. Poor education outcomes are associated Reaching secondary education significantly reduces with high poverty in Lesotho the likelihood of being poor. Households headed by individuals with no education have a higher poverty rate Incomplete primary education remains a common than others: 56 percent of them are poor. The poverty characteristic of the poor. In Lesotho, 77 percent incidence decreases but remains high for those headed of household heads in the poorest quintile had not by someone who has not completed (52 percent) or completed primary school in 2017, while 61 percent of attained primary education (45 percent). Significant those in the richest quintile had finished at least lower decreases in poverty rates are only observed starting secondary school (Figure 64). Completing primary from those who attained lower secondary and above: a education seems to not make a significant difference reduction of 15 percentage points (Figure 65). between wealth quintiles. Figure 64: Education of household head by Figure 65: Poverty rate by educational attainment of quintile, 2017 household head, 2017 100% 60% 56% 52% 50% 45% 80% 77% 40% 30% 60% 61% 30% 20% 15% 40% 10% 5% 5% 29% 0% 20% Lower secondary Upper secondary Higher education Vocational/Programs Incomplete primary Completed primary No education 11% 0% Q1 Q2 Q3 Q4 Q5 No education or incomplete primary Completed primary Lower secondary and above Source: Calculations based on the 2017/18 CMS/HBS. Access to primary education is relatively equal primary education, most children enter school on time across quintiles. Estimates from the 2017/18 CMS/ or under the official entry age, and disparities are almost HBS show that primary net attendance rates for poor marginal across different quintiles (Figure 66c). One- children are less than only 2 percentage points lower third of children enter late at school, partly due to the than for non-poor children (Figure 66a). Across wealth cultural practice of herding among boys. Conversely, the quintiles, rates vary slightly from 89 percent to 92 percentage of overage children from the poorest quintile percent (Figure 66b). This equitable access to primary increases throughout the cycle, reaching 68 percent in education is the result of strong government efforts in Grade 7, compared to only 38 percent for the richest expanding access to education in the past decade. This quintile (Figure 66d). Those figures suggest that children resulted in primary attendance rates from 81 percent from poor households are more likely than those from for the poorest quintile in 2004/05 (Demographic and wealthier households to repeat grades throughout the Health Survey 2004/05) to 89 percent in 2017/18. cycle. Administrative data at the national level clearly show that repetition rates increase throughout the This relatively equal access to primary education cycle, standing at 8 percent in Grade 1 and reaching 12 masks other difficulties for poor children, mainly percent at the highest grade of primary school (Figure related to repetition of grades. At the first grade of 67). 88 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Figure 66: Education outcomes by poverty status, 2017 (a) Net attendance rates by education level and (b) Net attendance rates by education level and poverty status, 2017 quintile, 2017 100% 90% 92% 100% 80% 80% 60% 51% 60% 40% 30% 40% 22% 20% 8% 20% 0% 0% Primary Lower Upper Primary Lower Upper secondary secondary secondary secondary Poor Non poor Q1 Q2 Q3 Q4 Q5 (c) On-time, under-age and over-age in the first (d) On-time, under-age and over-age in the last grade grade of primary, by quintile, 2017 of primary, by quintile, 2017 100% 100% 28 20 80% 30 30 36 80% 38 47 53 68 66 60% 48 60% 47 45 35 31 34 40% 40% 27 26 20% 20% 16 19 25 25 35 33 32 26 28 16 14 21 0% 0% Q1 Q2 Q3 Q4 Q5 Q1 Q2 Q3 Q4 Q5 Under-age On time Over-age Under-age On time Over-age Source: Calculations based on the 2017/18 CMS/HBS. Figure 67: Primary repetition rates by grade, In addition to more frequent repetition of grades, 2014 and 2015 children from poor households have lower test scores. According to the 2016 National Assessment of 14% Educational Progress Survey Report, the performance of Grade 6 learners was very poor at 32 percent for literacy 12% in English and 44 percent in numeracy. The results are 10% more worrisome for children from poor households: 8% the lowest quartile of Grade 6 learners averaged at 20 6% percent in literacy, which was 24 percentage points less than the richest quartile. The gap between the 4% two extreme groups stands at 22 percentage points in 2% numeracy. Analysis of the 2014 National Assessment 0% also shows a wide gap between poor children residing in G1 G2 G3 G4 G5 G6 G7 mountainous zone – for example, the district of Qacha’s Nek (averaging 92 on learning scores) – and children who 2014 2015 are relatively wealthy and residing in the Lowlands, with Source: Education Statistics Report (2016). Leribe, Berea or Maseru scoring above 106 (Lesotho Lesotho Poverty Assessment I Progress and challenges in reducing poverty 89 Diagnostic Study, 2016). Residing far from school is Access to post-primary education remains a likely detrimental to student performance. Altogether, challenge for children from poor families. The 21 percent of Grade 6 learners performed at such education gap is widest between rich and poor in low levels that they could be regarded as functionally secondary education. Secondary net attendance rates illiterate; another 42 percent can be classified as ranged from 18 percent in the poorest quintiles to 57 functionally innumerate (Spaull, 2012). Combining the percent in the richest (Figure 68a). The gap increases household survey and SACMEQ data, Taylor and Spaull in upper secondary, with a 30-percentage-point (2015) estimate functional literacy at 61 percent of the difference between the poorest and richest quintiles. age cohort; only 9 percent have higher-order reading The low enrollment rates in secondary schools reflects skills. Regarding numeracy, the situation is worse: 50 lack of secondary schools, especially in rural areas percent are functionally innumerate and only 2 percent (344 secondary schools but 3,031 primary schools), have acquired higher-order numeracy skills. This weak inadequate teaching and, finally, the high cost of performance came despite Lesotho having more secondary education (World Bank 2019g). educational resources than the SACMEQ average. Figure 68: Education outcomes by poverty status, gender and region, 2017 (a) Secondary net attendance rates by wealth (b) Net attendance rates by education level and quintile, areas and gender region 70% 70% 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% t t Ru t Ru t 2 3 Ri 4 2 3 Ri 4 es es es es l l Ur l Ur l Al Al n n ra ra Q Q Q Q Q Q ba ba or or ch ch Po Po Boys Girls Boys Girls Source: Calculations based on the 2017/18 CMS/HBS. Secondary education attendance is low, particularly Students in Rural Mountains and Rural Senqu River for boys living in the poorest families and in rural Valley still have challenges in accessing schools. areas. In primary education, girls and boys are at Though primary net attendance rates differ only slightly nearly the same levels of primary attendance by wealth across regions, significant disparities arise in secondary quintiles or residence areas. However, 36 percent of attendance (Figure 68b). Students in Rural Mountains boys attend secondary school, compared to 49 percent or in Rural Senqu River Valley have the lowest rates of girls. In rural areas, the gender gap is even higher of secondary attendance, and Maseru Urban or Other – a difference of 18 percentage points between boys Urban have the highest. An inadequate supply of and girls. This difference is partly due to the cultural secondary schools might explain the low attendance practice of herding among boys (World Bank 2019g). rates, particularly in rural areas. Children from Rural Furthermore, attendance for boys in the poorest wealth Mountains walk on average 43 minutes to the nearest quintile is considerably lower than it is for boys in the secondary school, 20 minutes more than children living richest, suggesting poverty remains correlated to low in Maseru Urban. Children from Rural Senqu River Valley demand of education. walk on average 10 minutes more than children in the capital. 90 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Boys, children from the Rural Mountains and poor richest quintile (Figure 69a). About 34 percent of boys households face the highest dropout rates. One- aged 13-17 years living in rural areas were no longer at fifth of 13-15 year-old children in the poorest quintile school in 2017, compared to 17 percent for girls (Figure dropped out of school, compared with only 4 percent 69b). By region, almost 37 percent of 13-17-year-olds in the richest quintile. Dropout rates reach almost 50 in Rural Mountains dropped out of school; the figure is percent among 16-17-year-olds in the poorest quintile, less than one-fifth in urban areas. 30 percentage points higher than households in the Figure 69: School dropout rates, 2017 (a) Dropout rates by age group and quintile (b) Dropout rates by gender and region, 13-17 years 60% 49% Rural Senqu river valley 31% 50% Rural Mountains 37% Rural Foothills 40% 25% Rural Lowlands 19% 30% Other Urban 11% 20% Urban Maseru 20% 18% Rural Girls 4% Rural Boys 17% Urban Girls 34% 10% 4% Urban Boys 8% 0% 9% 13-15 years 16-17 years 0% 5% 10% 15% 20% 25% 30% 35% 40% Q1 Q2 Q3 Q4 Q5 Source: Calculations based on the 2017/18 CMS/HBS. Poorer households are also less likely to receive did not receive medical care; in non-poor households, it healthcare. In rural areas, up to 35.2 percent of ill was 30.8 percent. The inter-quintile ratio for this variable people did not receive medical care, compared with is 0.79, highlighting a lower recourse to health care for 31.4 percent in urban areas. Young people aged 15- people from the poorest quintile compared to those from 19 years (44.0 percent) and 20-24 years (45.7 percent) the richest quintile.26 The negative and significant value receive medical care less frequently than other age of the concentration index confirms that people from groups, although there is not much difference between poorer households are disproportionately affected.27 men (34.1 percent) and women (33.7 percent). The These findings are concerning given that acute health socioeconomic gradient is more concerning when events are quite frequent in the population, with 24.5 looking at households’ living standards. While 38.1 percent of people in surveys reporting an acute illness percent of ill people from extremely poor households or accident during the past four weeks.28 The results 26 For a given variable, the inter-quintile ratio (IQR) is defined as the average value for the richest 20 percent of the population divided by the average value for the poorest 20 percent. By definition, the IQR ignores what happens in the three other quintiles and, therefore, does not allow an assessment of a variable entire distribution. 27 The concentration curve (CC) plots the cumulative percentage of the health variable (y-axis) against the cumulative percentage of the population ranked by living standards, beginning with the poorest and ending with the richest (x-axis) (O’Donnell et al., 2008). If the CC lies above (below) the 45-degree line running from the bottom left-hand corner to the top right-hand corner – known as the equality line – then the variable takes higher (lower) values among poorer people. The farther the curve is above (below) the equality line, the more concentrated the health variable is among the poor (rich). From the CC, it is possible to derive the concentration index (CI), a synthetic measure of the magnitude of the socioeconomic inequality in a variable. The CI is twice the area between the concentration curve and the line of equality. When there is no socioeconomic inequality, the CI is zero. The CI takes a negative value when the CC lies above the line of equality, indicating disproportionate concentration of the health variable among the poor, and a positive value when the CC lies below the line of equality (Kakwani 1977, 1980; Kakwani, Wagstaff, and van Doorslaer 1997; Wagstaff, van Doorslaer, and Paci 1989; O’Donnell et al., 2008). For both the CI and the CC, the interpretation of the results in terms of pro-poor or pro-rich distribution of the variable of interest will depend on whether the variable represents a “good” or a “bad” (health) outcome. 28 According to the 2017/18 CMS/HBS, the main health problems reported by these people are influenza/cold (42.7 percent), headache (10.2 percent), back/joint pain (8.9 percent), diarrhea/intestinal inflammation (6.3 percent) and mouth or dental problem (5.4 percent). Despite the negative effects of these health problems—overall, 65.2 percent of people who suffered from an illness or accident were unable to do their usual activities for at least one day—not all the affected population sought and received medical care. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 91 highlight the necessity for policies aimed at improving equivalent, received 30 percent of public spending on accessibility to health care to account for local/ primary education, while children in the richest quintile regional disparities as well as the differences among received 8 percent. The pattern, however, was reversed socioeconomic groups.29 for secondary and higher education (Figure 70). At the secondary level, the poorest quintile received only 14 C. Public spending on education and health is percent of total public spending, compared to 23 percent not fully pro-poor and affordability issues for the fourth quintile and 18 percent for the richest one. are large for poor households The difference widens with education levels. Public expenditures on primary education are Using concentration curves to break down inequality pro-poor, while outlays for secondary and higher in the provision of public resources by education education are biased toward the rich. Poor children level shows that spending on primary education is account for the bulk of enrollments in public and church pro-poor but spending on higher education is not primary schools, which suggests they benefit more (Figure 71).30 The government spends LSL 100 per from public subsidies for primary education. Based on student in primary education, LSL 165 per student in public expenditures on education in 2015/16 and data secondary education and LSL 326 per student in tertiary from the 2017/18 CMS/HBS, poorer children in the education (World Bank 2019g). lowest quintile, measured by consumption per adult Figure 70: Government spending going to each quintile, by level of education, 2015/16 61% 70% 60% 50% 30% 29% 27% 40% 26% 24% 23% 21% 20% 20% 18% 18% 17% 30% 16% 15% 14% 20% 9% 8% 2% 0% 10% 0% Primary education Secondary education Higher education All Lowest quintile Second quintile Third quintile Fourth quintile Richest quintile Source: Calculations based on the 2017/18 CMS/HBS and public expenditure for 2015/16. Girls benefit more than boys from government rates at each level of education. The trend differs only at spending on education. As illustrated in Figure 72, the lowest quintile, where the benefit for girls and boys girls from the second quintile to the richest quintile is almost the same. Overall, 56 percent of government receive greater benefit from government spending than expenditures on education benefits girls. boys. This can be attributed to girls’ higher enrollment 29 These findings are consistent with results from the Lesotho Demographic and Health Survey (LDHS) 2014, which showed that the poor tend to have worse health outcomes than the non-poor. 30 The concentration curve for primary education lies above the line of equality, meaning that the poor benefit more than the non-poor. Secondary and higher education show inequality. 92 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Figure 71: Concentration curves by education Figure 72: Government spending by gender level, 2015/16 and quintiles 100% 2395 80% 1701 1598 60% 1255 1298 1224 1188 1125 40% 1070 1062 20% 0% 0 Q1 Q2 Q3 Q4 Q5 Lowest Second Third Fouth Richest Line of equality quintile quintile quintile quintile Primary education Secondary education Higher education Subsidy to girls Subsidy to boys Source: Calculations based on the 2017/18 CMS/HBS and public expenditure for 2015/16. The cost of education is one of the main reasons equivalent); for a boarder, they rise to around LSL 9,200 for dropping out of school. As a consequence of to LSL 12,700 (US$649 to US$896) (World Bank 2019g). government support for secondary and higher education According to the 2017/18 CMS/HBS, annual education not being pro-poor, many children cite the cost of spending per child amounted to LSL 3,200 for poor education as the main reason for dropping out of school households (US$225), and LSL 8,300 (US$585) for non- (Figure 73a and Figure 73b). Data from the 2017/18 poor households. For poor families, secondary education CMS/HBS indicate that primary education accounts costs on average 33 percent of household expenditures, for 8.5 percent of total household expenditures for the while it represents 10 percent of non-poor families’ poorest quintile and 5.6 percent for the richest quintile. expenditures. These costs make it virtually impossible For most children from poor households, however, for most children from poor households to attend secondary school is prohibitively expensive and less secondary education without scholarships or without accessible. For a day scholar, annual costs range from mothers’ working in factories (World Bank 2019g). about LSL 7,400 to LSL 9,100 (US$522 to US$642 Lesotho Poverty Assessment I Progress and challenges in reducing poverty 93 Figure 73: School dropout rates, 2017 (a) Reason for school dropout for boys in rural (b) Reasons for school dropout for girls in rural areas, areas, 13-17 years 13-17 years Cultural/ Poor Other; Other; Performance 10.0% religious 16.3% / Truancy; Got beliefs; Too 2.1% married/ 4.9% expensive; pregnancy; 32.0% 10.7% Poor Performance/ No interest/ not Truancy; useful; 8.2% 10.3% Too expensive, No interest/ 69.0% not useful; 36.5% Source: Calculations based on the 2017/18 CMS/HBS. Costs are also a barrier for health care among poor are more than a third of all health care related costs. households. Table 10 illustrate some direct medical Using a monthly reference period, Table 11 shows that costs and direct non-medical costs incurred by people in total, health expenditures represented more than 5 who sought health care during the four weeks preceding percent of total household expenditures for 10.1 percent the 2017/18 CMS/HBS. The mean expenditure on health of the households surveyed. Health expenditures reach care visits and medicines is very high for people from at least 6.8 percent of total expenditures for 10 percent richer households (LSL 66.0 for the fourth quintile and of households, 5.3 percent for 15 percent and 4.0 LSL 78.5 for the fifth quintile respectively), compared percent for 20 percent. Whatever the threshold used, to poorer households (LSL 21.0 for the first quintile and the distribution of catastrophic health expenditures LSL 35.2 for the second quintile). Transport costs are a across socioeconomic group reveals a disproportionate key reason why poorer households are unable to afford incidence among poorer households. health care -- for the poorest quintile, transportation costs Table 10: Mean expenditures on health care and average cost of transportation to health care (LSL) Health care visits and   Transportation medicines Region Urban Maseru 54.1 13.4 Other Urban 75.6 20.2 Rural Lowlands 49.0 19.5 Rural Foothills 46.1 21.8 Rural Mountains 25.6 19.4 Rural Senqu River Valley 15.2 17.0 94 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Health care visits and   Transportation medicines Consumption per adult 1 21.0 12.3 equivalent quintiles 2 35.2 15.7 3 33.3 19.2 4 66.0 20.8 5 78.5 23.6 Poverty status Extreme poor 21.5 12.3 Poor 38.8 17.4 Non-poor 63.9 21.8 Overall 48.3 18.6 0.273 0.126 CI (p-value) (<0.0001) (<0.0001) IQR 3.74 1.92 Source: Calculations based on the 2017/18 CMS/HBS. Table 11: Share of health expenditure in total household expenditures during the last month (percentage of households) Threshold 5% 10% 15% 20% Region Urban Maseru 7.8 4.7 2.6 1.5 Other Urban 8.0 4.6 3.8 3.5 Rural Lowlands 11.2 7.1 6.0 4.2 Rural Foothills 13.2 11.4 8.8 7.9 Rural Mountains 12.4 9.45 7.2 5.6 Rural Senqu River Valley 9.2 7.3 5.4 2.9 Consumption per adult equivalent 1 15.3 11.3 9.2 7.1 quintiles 2 11.9 8.4 6.5 5.5 3 12.7 7.7 6.5 4.7 4 8.1 6.3 5.0 4.0 5 6.1 3.4 1.8 0.9 Poverty status Extreme poor 14.1 10.3 8.1 6.4 Poor 13.2 8.6 7.0 5.5 Non-poor 7.7 5.1 3.7 2.7 Overall 10.1 6.8 5.3 4.0 -0.207 -0.238 -0.281 -0.317 CI (p-value) (<0.0001) (<0.0001) (<0.0001) (<0.0001) IQR 0.40 0.30 0.20 0.13 Source: Calculations based on the 2017/18 CMS/HBS. Note: The numbers represent the share of the subpopulation that had health expenditures in the past month exceeding 5 percent, 10 percent, 15 percent or 20 percent of their total expenditures. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 95 Only 1.4 percent of surveyed people report affiliation D. Education levels matter greatly for labor with a health insurance scheme. This means that market outcomes financial protection is critical to preventing or mitigating the risk of a health catastrophe or falling into poverty Human capital levels are closely associated with due to disproportionately high health care expenditures. labor market status, and expanding secondary and Health insurance is one of the prepaid mechanisms that post-secondary education offers clear gains in terms can protect households from financial hardship related of wages. Those with no formal education had the lowest to illness episodes. Most people who are insured are unemployment rates but also the lowest wages (Table covered through a household member’s insurance (42.3 12). The poor cannot afford not working. Unemployment percent) or through a private employer or NGO (39.9 rates increases with levels of education, reaching nearly percent). Only 17.8 percent are self-insured. People from 22 percent for upper secondary. Wages also increase the poorest households mainly benefit from a household with levels of education for Basotho who are employed, member’s insurance, while the richest households are suggesting clear pay gains in expanding secondary and covered through other types of insurance. post-secondary education across Lesotho. However, a lack of jobs, as evident in high unemployment rates as well as challenges in educational quality, might reduce the attractiveness of education for many individuals. Improving access and quality of education is therefore critical because it increases the likelihood of finding employment (World Bank 2019g). Table 12: Employment, unemployment and yearly wage earnings, by educational attainment, 2017   Employment to working-age Unemployment rate Mean of yearly wage population ratio (% of labor force) (LSL, 2017 prices) No education 65.5 7.5 15,400 Incomplete primary 57.3 10.4 19,800 Completed primary 55.5 14.2 18,901 Lower secondary 50.3 14.5 22,675 Upper secondary 52.4 21.5 36,642 Vocational/programs 65.9 14.9 61,356 Higher education 69.3 12.2 80,320 Source: Calculations based on the 2017/18 CMS/HBS. Note: Working-age population is 15 years old and above. Unemployment rates are defined using the ILO definition. Access to quality education is central to accessing (Figure 74). By region, the rate of return above no higher-paying jobs. Each additional year of education education for primary education is highest in the Rural is associated with larger wage earnings, a return of 12 Senqu River Valley, while secondary education pays percent. A wage-worker with primary education earns off best in the Rural Foothills (Figure 75). In all regions, 24 percent more than one with no education. It rises investing in higher education more than doubles wages to 68 percent for secondary schooling, 66 percent for compared to the pay of those with no education. vocational training and 182 percent for higher education 96 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Figure 74: Rate of return by education level, 2017 Figure 75: Rate of return by education level and region, 2017 200% 182% 200% 175% 150% 150% 125% 100% 100% 68% 66% 75% 50% 50% 24% 25% 0% 0% Primary Secondary Higher Primary Secondary Vocational Higher education Programs education Lowlands Foothills Mountains Senqu River Valley Source: Calculations based on the 2017/18 CMS/HBS. Note: In both figures the reference category is no education. E. Conclusion and policy discussion primary level, is important in this regard. To achieve quality secondary education, it is imperative to raise Despite relatively high public spending on schooling, the quality of primary schools, particularly in the education outcomes are relatively poor and foundational skills of reading, writing and arithmetic. inequitable. Around 13.8 percent of the national budget This will require improving teaching quality and teacher was spent on education in 2017/18, or 6.3 percent of GDP. training. Improvements in primary and secondary Primary education accounts for most of the education education need to be complemented by increased budget, due primarily to the goal of achieving universal investments in Early Childhood Care and Development primary education. However, primary enrollments have (ECCD), a cost-effective strategy for reducing inequities been stagnant in recent decades and poor performance and substantially improving long-term adult health in learning outcomes remain a challenge. Furthermore, outcomes. Children who benefit from quality ECCD the system is inequitable in terms of educational perform better in primary school, repeat grades less outcomes. Despite equal access at primary education, often and drop out less frequently. Therefore, giving poor children, especially herd boys, face high repetition the poor the opportunity to access quality ECCD, will and dropout rates at the primary level. Access to post- improve not only the quality of education but will also primary education remains a challenge for children from improve the cost efficiency of the system. Investing poor families, boys and children in rural mountains in ECCD would ensure accelerated flows through and in Rural Senqu River Valley. Those who do remain primary school, so more children enter junior secondary until the end of primary school often face significant education with a solid knowledge base. impediments in transitioning to secondary education due to access issues and the high costs of secondary Expanding secondary education and making education. As the same time, public education spending secondary schools more accessible to the poor is in secondary and higher education favors the rich. Yet, important for supporting reductions in poverty and the educational level is closely associated with labor inequality. Expansion of the junior secondary school market status, and there are clear gains in terms of network would help make these schools more accessible wages in expanding secondary education. to the poor and enhance the equity of the educational system. In addition to putting in place the appropriate Given that universal and free primary education physical and soft infrastructure, increasing equity of has largely been attained, it is important to boost access to secondary education requires that the financial learning in primary education overall and increase cost of attending such schools be reduced as much as secondary enrollment and completion among the possible. Further details on policy recommendations poor. Expansion of access to secondary education, emerging from the Public Expenditure Review (World combined with attention to educational quality at the Bank 2019g) are summarized in Box 13. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 97 Box 13: Policy recommendations for the education sector The Public Expenditure Review (World Bank 2019g) identified poor planning and resource management as the main causes of inadequate educational outcomes. The report makes two sets of policy recommendations: Strengthen access to quality education in ECCD and primary schools: • Expansion of ECCD programs to prepare students for primary education; • Focus on improving quality of education in primary school by improving teacher management as well as increasing resources for learning materials; • Focus on teacher management through a teacher management database which should be linked to the Education Management Information System; • Ensure teacher lists are up-to date at the district and school level; • Revisit teachers’ salaries, retirement, and qualifications; • Invest in teacher training by focusing on coaching and improving supervision mechanisms; • Train teachers on new curriculum; • Prioritize systemic testing and collect information on cognitive development. This could be done through sample-based testing, as well as considering participation in other regional and international learning assessments, such as the Trends in International Mathematics and Science Study (TIMSS) Numeracy and the Progress in International Reading Literacy Study (PIRLS) Literacy Assessments. Prioritize secondary school expansion: • Junior secondary classes can be added to existing primary schools in mountainous and rural regions so that families do not have to pay for dormitories; • In the host primary schools, the student-teacher ratio could be increased, and the school’s facilities could be used in double shifts to diminish capital costs of construction; • The unit cost of school construction could be reduced by opting for science kits instead of laboratories, and an in-classroom library corner instead of school libraries; • Consider bursaries for students from poor households; • Subsidize secondary schools heavily by abolishing school fees and textbook rentals fees; • Expanding school feeding programs to secondary schools. Health expenditures pose a disproportionately large burdens on poorer households point to the importance burden for poorer households. Many individuals, of improving awareness and prevention as well as especially those from poorer quintiles, did not get care access to health care and financial protection for all when needed, the affordability of medical care being households – but particularly the poor. one of the main reasons. The disproportionately large 98 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Chapter II.3: Labor market challenges for poverty reduction in Lesotho Lesotho’s economy has been undergoing a structural transformation due to a growing services sector, improvements in the business climate and a more educated labor force. These changes, together with a demographic dividend, supported poverty reduction. Despite these improvements, labor market outcomes generally remain poor. The labor market is characterized by low employment rates and high levels of unemployment. Productivity is low owing to low skills, a large informal sector and reliance on subsistence agriculture. A high proportion of Basotho are employed outside of the country, mainly in South Africa. As a result, Lesotho remains dependent on remittances from South Africa, although these declined in the past 15 years. Creating an enabling environment for private sector job creation and improving agricultural productivity is critical for increasing incomes, reducing dependence on remittances and reducing poverty. Deepening regional integration is also important. On the supply side, targeted policies that boost entrepreneurship and promote development of skills will support improvements in labor productivity and poverty reduction. Adoption of Climate Smart Agriculture practices in Lesotho could increase productivity and incomes. It is essential to improve migration governance by developing a national strategy with a capacity to prevent irregular migration, promote legal migration and mobility and ultimately enhance synergies between migration and development. A. Characteristics and challenges of Lesotho’s Productivity is low in Lesotho, and growth has labor markets slowed substantially in recent years. Total factor productivity is low and has been stagnant for a long time Lesotho’s labor markets are characterized by (Figure 78). The marginal productivity of labor is relatively declining labor force participation and high low and negatively affected by HIV/AIDS. Weak total unemployment rates compared to the other factor productivity also appears to reflect inefficient Southern African countries. The challenges of poverty inter-sectoral allocations, with labor concentrated in the and inequality are partly due to the country’s inability economy’s least productive sectors. In recent years, to generate enough jobs, with government playing mining has become one of the economy’s fastest an important role in providing formal employment. growing sectors, but it has limited job-creation potential. Employment is concentrated in a few economic activities, including the apparel industry and wholesale- The public sector is one of the main growth-driving retail trade services. The manufacturing industry provides parts of the economy, but it does not employ many an important source of income for the low-skilled and of the poor. The public sector’s contribution to GDP is poor population (World Bank 2018a). Between 2002 highly dependent on the cycle of fiscal policies (World and 2018, labor force participation steadily declined Bank 2018a). In the past decade, Lesotho’s public sector (Figure 76). An increase in manufacturing and services has been the largest in the Southern African region, led to a growing demand for unskilled and low-paid making up 35 percent of GDP (Figure 79). Government workers; in turn, this led to a decline in unemployment spending is highly dependent on SACU revenues, and rates (Figure 77). However, a recent slowdown caused their volatility raises concerns about fiscal sustainability unemployment to rise and, according to International (World Bank 2018a). Most of the poor in Lesotho work in Labor Organization (ILO) forecasts, over a quarter of the private sector (Figure 80), which has had difficulties Lesotho’s working-age population was unemployed in creating jobs and attracting skilled labor (World Bank 2018 (Figure 77). 2018a). Lesotho Poverty Assessment I Progress and challenges in reducing poverty 99 Figure 76: Labor force participation rate in South Figure 77: Unemployment rate in Southern African countries, 2002–2018 African countries, 2002–2018 80 40 75 35 70 30 65 25 Percent Percent 60 20 55 15 50 10 45 5 40 0 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Lesotho South Africa Namibia Lesotho South Africa Namibia Eswatini Botswana Eswatini Botswana Figure 78: Total factor productivity in Southern Figure 79: Government spending as percent of African countries, 2002-2017 GDP, 2007-2016 1.2 45 40 1 35 0.8 Percent 30 25 0.6 20 0.4 15 10 0.2 5 0 0 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Lesotho South Africa Namibia Lesotho South Africa Namibia Eswatini Botswana Eswatini Botswana Source: ILO data; https://www.theglobaleconomy.com/download-data.php. For TFP: https://www.rug.nl/ggdc/productivity/pwt/. Lesotho went through a period of high economic enough to absorb the large future cohorts of working- growth, but with limited impact on employment. In age people, especially considering recent and expected recent years, Lesotho has mainly relied on public-sector declines in fiscal revenues. Mining is also one of the expansion as its key source of growth. The government fastest-growing sectors of Lesotho’s economy, but it is is already the formal economy’s largest employer. capital-intensive, with little impact on employment. It is unlikely that this sector will continue to expand 100 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Box 14: Government prioritizes business development The government’s National Strategic Development Plan II identifies four priority sectors: (1) manufacturing (measures include strengthening the competitiveness of the garment industry and attracting investment in other industries), (2) commercial agriculture (including livestock, deciduous fruit and other high-value crops), (3) technology, and (4) tourism and creative industries. In 2017, Lesotho elected a four-party coalition government that prioritized measures to increase investment and create jobs. The second (2018/19) and third budget speeches (2019/20) prioritized actions to reduce unemployment, increase foreign and domestic investment, expand access to finance for domestic businesses, attract private property developers to increase the availability of industrial infrastructure and strengthen vocational institutions to improve skills availability. The government is rolling out several initiatives to attract investment. The Ministry of Trade and Industry and the Lesotho National Development Corporation (the main parastatal in charge of the implementation of the country’s industrial development policies) are incubating several new business ventures. They include (i) eight new firms in clothing and electrical components, with an estimated 8,000 jobs to be created in two years; (ii) fresh produce activity and an out-grower scheme aiming at 500 jobs in the Maseru area; and (iii) expansion of the area under deciduous fruit production from the current 34 hectares to 150 hectares in the next two years and 500 hectares in the medium term, with a total impact of 1,000 jobs. Apart from the large investments outlined above, the government is implementing financial and business support to help start-ups and other small businesses. The apprenticeship strategy has been developed with the main objective to address the existing mismatches between skills provided by the training institutions and the needs of the industry. Through this strategy, an institutionalized program will be established to expose graduates of universities and higher learning institutions to industry and business practices. Figure 80: Private/public sector source of Figure 81: Labor force participation rate in Lesotho, household income by poverty status, 2017 2002–2017 80 100 60 Percent 80 40 54.9 67.7 Percent 20 60 1.5 0 40 -4.4 -11.7 -10.5 -6.0 -0.2 -8.4 -1.8 -20 Extreme poor Secondary Primary Non-poor None Poor Tertiary 20 45.1 32.3 0 Public Private Total Poverty Education 2002 2017 Change (2002-2017) Poor Non-poor Source: Calculations based on the 2017/18 CMS/HBS. Notes: The left figure refers to information about the main source of household income. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 101 Returns to education (especially post-secondary) firm operation by 35 percent of domestic firms (World are high, increasing the probability of being in the Bank 2016a), and bank credit to the private sector is labor force and finding a job. The higher the level of much lower than in comparable economies. education, the higher the probability of being in the labor The informal sector is large, and lack of access force. Both in 2002 and 2017, the labor force participation to credit and household savings are barriers to rate was highest among those with technical/vocational formality. Most of the informal sector in Lesotho training or with university or higher education (Figure involves activities related to the agricultural sector as well 81). It was lowest among youth below 25 years of age. as the textile and apparel industry (such as road freight In general, developing the skills demanded by industries and passenger transport, a small packaging industry, and offering entrepreneurship programs for growth- residential accommodation and catering services for oriented businesses could improve the competitiveness employees, water, telecommunications and utilities of Lesotho’s private sector and help address the skills services). Most farmers sell their produce in informal mismatch challenge. For example, training in sewing markets, although few supply institutional buyers (World machine repair as well as degrees in horticulture and Bank 2018a). As a result, a large share of workers holds ICT-related fields would help provide skills required on informal jobs, characterized by low productivity with labor’s demand side (World Bank 2018a). little access to technology and social protection benefits Youth unemployment is high and is driven by skills (World Bank 2019b). In both 2002 and 2017, more than mismatches and low wages. Even though multiple 80 percent of household businesses in Lesotho were educational institutions offer courses related to textiles not registered (Figure 82). Informal businesses are less and apparel, the skills taught do not match those required likely to require initial capital investments, suggesting by the industry. Universities and vocational schools offer that easier access to credit could encourage the start of courses in fashion and design; however, no institution formal activities (Figure 83). offers training in sewing machine repair, a skill in high Women tend to have the lowest paying jobs, demand at the factories (World Bank 2018a). Graduates participate less in the labor force and earn lower with tertiary qualifications will not accept the low salaries. While a relatively larger share of women than wages offered for machine operators; instead, the jobs men are educated, this advantage does not translate into go to low-skilled workers with incomplete secondary enhanced economic power. Females are less likely to education, no formal industry-specific training and a participate in the labor force and have a 22 percent higher rudimentary command of English (making it difficult to probability of being unemployed. In addition, women’s advance to managerial positions) (World Bank 2018a). wages are lower, mainly due to gender difference in The broad unemployment rate, which includes those returns to primary and secondary education.The ratio of in the labor force who were discouraged and no longer female-to-male labor force participation was estimated searching for jobs, was highest among the very poor at 0.81 in 2017, compared to 0.84 for Sub-Saharan and youth below 25 years old in both 2002 and 2017. Africa as a whole (selected years for other countries). Lesotho’s business climate has improved in recent In addition, the ratio of the total female unemployment years, but access to finance remains an issue. rate to the total male unemployment rate was 1.15 in Lesotho ranks 122 globally in the World Bank’s Doing 2017. Among employed women, the large majority is Business 2020 Report (World Bank 2020). Positive relegated to insecure and low-paying jobs in the informal developments took place in recent years, including the sector, including subsistence agriculture in rural areas opening of the One-Stop Business Facilitation Center and domestic work or street vending in urban areas. and establishment of the Credit Bureau. Furthermore, Females have a higher employment share than males in the process of obtaining construction permits was the government sector (10 percent vs 6 percent), self- streamlined by the introduction of electronic systems in employed (22 percent vs 14 percent) and household August 2017. However, weak access to finance remains workers. Females have a higher share in employment an obstacle for private sector development. The process in services and elementary occupations. On average, of receiving business licenses and permits remains males tend to have higher earnings than females, cumbersome and regulatory compliance costs are high. controlling for other factors. Access to finance is considered a major constraint to 102 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Figure 82: Share of registered and non-registered Figure 83: Source of funding for businesses by household businesses, 2002 and 2017 formal/informal, 2017 100 100 9.0 12.7 12 80 80 27 5 Percent Percent 60 60 10 40 91.0 87.3 40 79 58 20 20 0 0 2002 2017 Informal Formal Non-registered Registered Household Loan No funds Other Source: Calculations based on the 2002/03 HBS and the 2017/18 CMS/HBS. Subsistence agriculture is important for the poverty in rural areas. In addition, most of the horticultural livelihoods of most Basotho, but it is characterized output is owner-consumed or sold in informal domestic by low productivity. More than half of Lesotho’s markets, despite having a high potential of becoming population lives in rural areas (66 percent according to commercialized and a source of higher revenues (World the 2016 census), and the majority of them engage in Bank 2018a). In 2018, almost half of agricultural workers agriculture based on subsistence cultivation of cereal – and 16.2 percent of all workers – were in subsistence crops. This contributes to low incomes and widespread agriculture (Figure 85). Figure 84: Trends of the GDP share of the services Figure 85: Share of workers by type of and agricultural sector, 2002–2017 employment, 2017 35 30.5 70 30 Percentage of GDP 60 25 Percent 50 17.8 16.2 20 40 15 30 8.5 10 5.8 20 3.0 3.0 5 1.0 10 0 Casual employed Self-employment Employee Casual non- Casual agricultural Subsistence Agriculture for sale 0 agricultural Agricultural agriculture 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 Industry (including construction), value added (% of GDP) Services, value added (% of GDP) Agriculture, forestry, and fishing, value added (% of GDP) Source: ILO. Calculations based on the 2002/03 HBS and the 2017/18 CMS/HBS. Note: The reference period is the past seven days. Source: World Development Indicators. Source: ILO data Lesotho Poverty Assessment I Progress and challenges in reducing poverty 103 Figure 86: Minimum monthly wages for employees Figure 87: Bank credit to the private sector is with more than 12 months of service with the main lower in Lesotho than in comparable countries employer 160 148 2500 140 2049 2049 Percentage of GDP 1883 120 2000 99 1456 1530 100 1500 80 LSL 62 1000 596 60 47 40 31 500 21 18 20 0 Domestic worker 0 Hotels, motels and (clothing, textile and Restaurants, food General minimum supermarkets and furniture shops Machine operator Botswana Middle income Sub-Saharan Africa Lesotho South Africa Eswatini Namibia Wholesale, caterers lodges wage leather Source: Lesotho Government Gazette 2017. Source: IMF 2018a. Structural transformation is evident in Lesotho. The trade. Potentially, this can increase GDP and decrease process has been characterized by the relative decline poverty, but it could also mean more informal jobs of agriculture and gains in industry and services (Figure for less educated individuals. In 2017, most firms (89 84). Despite the dominance of subsistence agriculture, percent) were in services, and 71 percent were in retail there is evidence of agribusiness’ rising importance. and wholesale trade. Low entry barriers for capital and The evidence includes the value added from agro- skill requirements and the lack of opportunities in other related industries, agricultural trade and distribution sectors might be behind the concentration of firms in services as well as a shift from traditional exports commerce (World Bank 2018a). to high-value agricultural products in international Box 15: Labor laws and labor market institutions Lesotho has been a member of the International Labor Organization (ILO) since 1966 and has ratified 23 international labor conventions. Lesotho’s Labor Code Order of 1992 and its subsequent amendments are the principal laws governing the terms and conditions of employment, including worker health, safety and welfare. Statutory minimum wages are fixed annually by the Ministry of Labor and Employment, based on recommendations from a tripartite Wages Advisory Board representing the government, employers and employees. Separate minimum wages exist for main industries. Sectoral minimum wages exist for clothing, textiles and leather, manufacturing, construction, wholesale and retail trade, hospitality, services, transport, small business and domestic workers. The law permits union organization, freedom of association and the right to bargain collectively. Most unions focus on organizing apparel workers. The law provides for a limited right to strike. In the private sector, the law requires workers and employers to follow a series of procedures designed to resolve disputes before the Directorate of Dispute Prevention and Resolution, an independent government body, authorizes a strike. The law does not permit civil servants to strike, making all public sector strikes illegal. In practice, strikes are rare in Lesotho. The Labor Code allows firms to hire non-citizens with a work permit. Permits are issued based on a quota formula by the Labor Commissioner, who must be satisfied that no qualified Lesotho citizens are available for the position. 104 Lesotho Poverty Assessment I Progress and challenges in reducing poverty B. Poor labor market outcomes are associated with high poverty Poverty falls with employment, but work does not work does not guarantee lower poverty levels. In guarantee being out of poverty. Poverty among the fact, poverty among those employed in agriculture employed fell the most in Lesotho. A large portion basically stagnated between 2002 and 2017, leaving 56 of Lesotho’s population consists of working poor percent of agriculture workers poor in 2017 (Figure 89). who earn very low wages. In 2017, 36 percent of Basotho employed in services and industry have better employed were poor in Lesotho (Figure 88). This is employment outcomes and wages, with lower poverty associated with the type of employment. Agricultural rates and faster poverty reduction. Figure 88: Poverty rates by employment status, Figure 89: Poverty rates by employment 2002 and 2017 sector, 2002 and 2017 80 80 56 53 61 62 54 53 56 56 60 60 44 43 Poverty rates 36 39 Poverty rates 40 40 29 28 20 20 0 0 -3 -7 - 6 -20 - 20 -18 - 15 - 17 - 15 -40 Inactive Unemployed Employed - 40 Agriculture Industry Construction Services 2002/03 2017/18 Change 2002/03 2017/18 Change Source: Calculations based on the 2017/18 CMS/HBS. Note: The industry sector refers to the household business. Poor labor market outcomes contribute to Lesotho’s Employment composition changed drastically persistently high poverty levels. As discussed earlier, between 2002 and 2017, coinciding with increases the labor market is characterized by a low employment in education among the employed. The labor force ratio, a high level of unemployment, a high share of was more educated in 2017 than in 2002. As presented employment outside of the country and generally low in Figure 90, the share of people without education fell wages alongside a well-paid public sector. The portion drastically – from approximately 45 percent in 2002 to of wage-earning workers increased more rapidly among 8 percent in 2017. The share of the population with poor households than among non-poor households, secondary education increased from 13 percent to indicating that getting a wage-paying job might be 36 percent, while the portion with tertiary education instrumental in moving up the consumption ladder but increased from 2 percent to 6 percent. may not always be enough to rise out of poverty. This stresses the need to increase labor productivity. The level of education emerges as a crucial factor, correlated with the type of employment. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 105 Figure 90: Labor force composition by education level, 2002–2017 100 Percentage of labor force 13 80 36 60 34 40 46 20 45 0 8 2002 2017 None Primary Secondary Tech/Vocational Higher Source: Calculations based on the 2002/03 HBS and 2017/18 CMS/HBS. Education levels also play a role in securing educated people get more permanent or formal jobs employment outside agriculture and qualifying for (Figure 92), while temporary and seasonal jobs are permanent jobs. The employment sector and type of dominated by the less educated and lower-paying jobs. job held are closely related to education levels. Those The highest returns to education are for those who who have primary levels of education dominate jobs studied in technical schools or those who have post- in the agricultural sector (Figure 91). Clearly, more secondary education.31 Figure 91: Education level by wage employment Figure 92: Education level by type of job, 2017 type, 2017 100 100 20 22 80 28 40 35 34 Percent Percent 60 40 50 62 62 46 40 58 49 48 20 35 5 8 16 0 11 14 Non agric self- Wage employees Agric self- 0 5 7 employment employment Permanent Contract Temporary Seasonal None Primary None Primary Secondary Tech/ Vocational Secondary Tech/Vocational Tertiary Tertiary Source: Calculations based on the 2017/18 CMS/HBS. 31 The results are based on a Mincer regression (not included in the analysis but can be shared on request). 106 Lesotho Poverty Assessment I Progress and challenges in reducing poverty The size of the firms where individuals work also The duration of working contracts is strongly plays an important role in defining poverty levels. correlated with individuals’ poverty status. Many Smaller enterprises make up the largest share of firms, enterprises operate only seasonally, and this intermittent and they are more likely to be informal, less productive, pattern shows up more frequently in rural enterprises have little access to technology and not offer social than urban ones (Christiaensen et al., 2018). Workers protection (World Banks 2019b). The largest share of with seasonal contracts are the poorest, followed by individuals who are self-employed or work in micro firms those with casual contracts and temporary jobs. The are poor, while the share of poor is very low among share of poor is lowest among workers with permanent those working in large firms (Figure 93). jobs (Figure 94). Figure 93: Poverty status by firm size, 2017 Figure 94: Poverty status by type of job, 2017 100 Share of population (%) 100 Share of population (%) 80 46 51 80 39 66 70 77 54 48 60 71 69 60 40 54 49 40 20 34 30 52 61 23 20 46 0 29 31 Own account Medium 0 Large Small Micro Permanent Seasonal Temporary Contract Casual Poor Non-poor Poor Non-poor Source: Calculations based on the 2017/18 CMS/HBS. Note: The industry sector refers to the household business. Adoption of digital technology is lagging in Lesotho. and quality of Internet services should be an important This is demonstrated by the country’s low standing on priority for further development. More reliable internet the World Bank Digital Adoption Index, measuring the will also help businesses connect to markets and spread of technology among businesses, governments improve productivity. and citizens.32 Lesotho ranks among Africa’s lowest performers in the United Nations’ global e-Government C. Low productivity in subsistence Development Index (EDGI) – falling below the continental agriculture poses a challenge for poverty average and other Southern African Development reduction Community (SADC) countries. Weak competition in the broadband market remains a challenge and contributes The majority of Lesotho’s population lives in to low Internet use by businesses and consumers and rural areas and depends directly or indirectly on relatively high prices for service. Digital skills constitute agriculture as a means of livelihood. Nevertheless, a clear bottleneck for Lesotho, making it an area where agricultural productivity is low, and the value added is policy and institutional reforms are needed. Measures to limited (World Bank 2019f). Most of those engaged in improve digital skills need to be backed up by a strategic agriculture are smallholders with less than one hectare focus and relevant data. Upgrading the accessibility per family, and cereal yields are low. Low agricultural 32 The results presented here are based in the recent Lesotho Digital Economy Diagnostic report (World Bank 2019c). This report uses the Digital Economy for Africa (DE4A) methodology to conduct an examination of the five pillars of Lesotho’s digital economy, focusing on digital infrastructure, digital government platforms, digital financial services, digital entrepreneurship and digital skills. The report is based on extensive desk research and interviews of a wide array of government entities and other stakeholders. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 107 productivity and poverty are closely linked with food Low investment in agriculture has rendered the insecurity and malnutrition in Lesotho. About one-third sector vulnerable to weather impacts and poorly of all children under age 5 are stunted. In the lowest equipped to adapt to climate change. The sector’s income quintile, one-half of all children under age low productivity is largely a result of underinvestment 5 are stunted, compared to 10 percent of children in in infrastructure (including irrigation), low uptake of the highest. Growth in the agricultural sector will be new technologies and inputs, poor-quality extension central for poverty reduction in Lesotho. Recognizing and advisory services and limited access to credit. the sector’s importance for “job creation and inclusive Environmental challenges such as land and natural economic growth under a new growth path led by the resource degradation, erosion and low soil fertility private sector,” the government selected agriculture exacerbate the situation, leaving the sector vulnerable as one of four productive sectors central to its new to risks. Frequent droughts and high rainfall variability National Strategic Development Plan (NSDP) II for from one season or year to another, including the El 2018/19–2022/23.33 Niño Southern Oscillation (ENSO), have resulted in poor crop harvests and large livestock losses for rural Current agricultural production in Lesotho focuses farmers. Lesotho’s agricultural system faces a growing on extensive animal grazing and expansion number of climate-related vulnerabilities with droughts, of agricultural cropland to keep pace with the floods, pests and extreme temperatures occurring more population demand for food. This is largely frequently. In response, the government is integrating unsustainable and depletes the land resources needed climate change into the country’s agriculture policy for production over time. Adoption of Climate Smart agenda. Agriculture (CSA) principles in Lesotho should increase productivity and incomes, enhance food security and dietary diversity, reduce impacts of climate change on agricultural produce, and improve commercialization, employment opportunities and rural livelihoods (World Bank 2019e). 33 The three other productive sectors included in the draft NSDP II are manufacturing, tourism and creative industries, and technology and innovation. To strategically support agriculture’s role in climate change mitigation and adaptation and in nutrition and food security, the government is developing an Integrated Program for Agriculture and Food System Development (a National Agriculture Investment Plan, NAIP), a Multi-Sectoral Nutrition Strategy and, with the support of the World Bank, a Climate Smart Agriculture Investment Plan (CSAIP). All of which will be ready for adoption in 2019. 108 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Box 16: Lesotho Climate-Smart Agriculture Investment Plan (CSAIP) Lesotho’s CSAIP aims to identify climate-smart agriculture (CSA) investments that offer the greatest potential to transform the country’s agriculture into a more productive, climate-resilient and low-emission sector (World Bank 2018c and World Bank 2019e). The CSAIP offers two complementary pathways for scaling up CSA in Lesotho.34 Commercialization can be prioritized largely in lowlands and foothills, while the highlands would benefit from resilient landscape, or afforestation, and farmer-managed natural regeneration to restore and replenish less fertile land. a. Commercialization entails focusing on commodities for which Lesotho has a distinct comparative advantage, like horticulture, potatoes and aquaculture; developing the country’s irrigation to its full potential; and forging linkages that connect smallholders to export and domestic markets. Commercialization can be prioritized largely in lowlands and foothills—the fertile and most productive parts of Lesotho that are suitable for orchards, vegetables, wheat, potatoes and peas. Commercialization will improve farmers’ productivity, create more jobs and offer Lesotho the potential to export horticulture, potatoes and vegetables. Strong market-oriented agricultural policies should be developed, and Lesotho’s agricultural value chains should be strengthened. The proper functioning of land markets will also be important. b. Resilient landscape combines modern scientific practices, such as improved crop varieties, crop rotation, relay cropping and intercropping practices. Application of manure and plant ash should conserve soil moisture and replenish soil fertility. Resilient landscape focuses on investing in sustainable landscape and integrated catchment management while strengthening local institutions. A resilient landscape should control land degradation and be tailored toward locally adapted technologies that the average smallholder farmer can practice. Resilient landscape can be emphasized largely in the highlands, a region suitable for potatoes, wheat, peas and orchards. The highlands would also benefit from afforestation and farmer-managed natural regeneration to restore and replenish less fertile land. Effective scaling up of CSA in Lesotho will require addressing several adoption barriers, including limited implementation capacity, insufficient access to inputs and credits and insufficient agricultural research. Policy actions to support effective scaling up of CSA identified in the CSAIP include: • Strengthening agricultural research and extension. There is a need to strengthen research and establish partnerships with international research institutes to develop high-yielding, stress-tolerant, climate-ready varieties. Agricultural extension services should be upgraded to catalyze the agricultural innovation process, improve the CSA knowledge system, facilitate access to information, knowledge and expertise, and provide technical advice to farmers. • Building capacity to access climate finance. Lesotho faces a financing gap in the agriculture sector with low capacity to access climate finance. Critical areas that need capacity development include identifying funding gaps and needs, assessing public and private financing options, developing payment for ecosystem services programs, developing bankable investment plans, project pipeline and financing propositions, and developing financially viable opportunities for effective private sector engagement. 34 Generally, CSA comprises three pillars: increasing productivity, enhancing resilience and adaptation and reducing greenhouse gas emissions from the agriculture sector. The Government of Lesotho is collaborating with the World Bank to integrate climate change into the country’s agriculture policy agenda. The CSAIP builds on Lesotho’s Second National Strategy Development Plan (NSDP II), and Lesotho’s international climate commitments articulated in its Nationally Determined Contribution (NDC). Lesotho Poverty Assessment I Progress and challenges in reducing poverty 109 Smallholder agriculture plays a major role in the compared to 58 percent of the wealthiest quintile employment and livelihoods of Basotho. About (Figure 95). Although people with less education are 71 percent of the population is involved in some more likely to farm, over a third of university graduates form of agricultural activity. Although farming is more are involved in agriculture. This may indicate a lack of widespread in rural areas, where 81 percent of the economic opportunities in other sectors of the economy. population is engaged in agriculture, it is also common in Residents of urban Maseru – the main business center urban locations, where 50 percent of people grow crops of the country – are significantly less likely to engage in or tend some livestock. Farming is often a dominant agriculture than those living in Rural Mountains, Rural source of livelihood for the poor: 78 percent of those in Foothills and Rural Senqu River Valley. the lowest quintile participate in agricultural activities, Figure 95: Share of population engaged in farming activities, 2017 100% 90% 83.8% 86.0% 86.9% 85.4% 81.0% 80% 75.6% 78.3% 73.1% 73.2% 77.1% 75.5% 69.7% (Total 70.5%) 70% 65.4% % of total population 58.0% 60% 53.7% 54.0% 50.1% 50% 44.9% 39.5% 40% 30% 20% 10% 0% Richest 20% Other Urban Poorest 20% Rural Senqu river valley Urban Maseru Rural Lowlands Rural Mountains Rural University Urban Not poor Poor Sec/ high school Rural Foothills None Primary Q2 Q3 Q4 Poverty status Quintile Education Region Location Source: Calculations based on the 2017/18 CMS/HBS. Low productivity and low earnings in subsistence Crop farming in Lesotho is mainly based on agriculture have a strong association with poverty. subsistence cultivation of cereals, an activity The large differential in poverty rates between urban and characterized by the low incomes that drive poverty rural areas can be attributed to a larger share of the rural in rural areas. The largest share of households population being employed in subsistence agriculture produces only cereals (Figure 96); among them, poverty and/or low-productivity/low-paying agricultural jobs. In is widespread (Figure 97). Horticultural producers contrast, urban areas are characterized by a larger share are less likely to be poor, and they are most likely to of the employed population in wage-earning jobs outside invest in agricultural inputs. Currently, most of the agriculture, including the well-paid public sector. These horticultural output is consumed within the household government workers earn an average monthly salary or sold in informal domestic markets. Transitioning to about seven times higher than the minimum monthly commercial cultivation of fruits and vegetables can wage of a skilled worker in the textile sector. significantly increase farm incomes (World Bank 2018a). However, the horticulture industry is currently facing skills constraints, poorly functioning land markets, poor linkages within supply chains and limited access to finance (World Bank 2018a). 110 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Figure 96: Share of households by type of Figure 97: Poverty status by type of parcel parcel cultivation, 2017 cultivation, 2017 100 90 80 42 29.8 70 60 58 60 Percent 43.3 50 40 31 30 22 22 20 28 10 18 19 26.8 0 Cereal only Horticultural only Mix Extreme poor Poor Non-poor Cereal only Horticultural only Mix Source: Calculations based on the 2017/18 CMS/HBS. Most farming households engage in several consumption. Lesotho’s agriculture is subject to agricultural activities with very little specialization. significant weather shocks, and diversification is also part Among farming households, 55 percent are involved of the risk mitigation strategy. Furthermore, smallholder in more than one agricultural activity (Figure 98). farmers that live far from urban centers often encounter Diversification across a variety of activities is reflective difficulties finding good markets for their output, so they of the subsistence nature of agriculture in Lesotho: produce different products in hopes that at least some farmers produce the food they need for household could be sold. Figure 98: Number of agricultural activities Figure 99: Most common agricultural engaged in, 2017 activities, by quintile, 2017 100% 90% 90% 80% % of farming households 80% 70% % of farming households 70% 60% 60% 50% 45,4% 40% 50% 40,3% 30% 40% 20% 30% 10% 20% 11,8% 0% 10% 2,4% Vegetable Cereal Fruit cultivation Livestock 0% cultivation cultivation breeding 1 activity 2 activities 3 activities 4+ activities Poorest 20% 2 3 4 Richest 20% Source: Calculations based on the 2017/18 CMS/HBS. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 111 Livestock farming is the most prevalent agricultural Lesotho’s agriculture sector is dominated by activity in Lesotho, with over three-quarters of smallholder subsistence farmers. Average field size is households raising animals (Figure 99). Furthermore, 3.5 hectares (five hectares among non-poor households 53 percent of farming households are involved in and two hectares among the poor). Most farming cultivation of cereals, primarily maize. Lesotho does households produce for their own consumption and not have a competitive advantage in cultivating cereals only 3 percent produce primarily for sale. As might be because of its mountainous terrain, challenging expected, there is a higher share of market-oriented agroclimatic conditions, poor soils and barriers to farmers among the non-poor. Only 1 percent of farming attaining the large economies of scale and high levels of households in the poorest quintile produce only or mechanization required for commercial grain cultivation. mainly for sale/barter, compared to 5.3 percent in the Although horticulture crops are more suitable for wealthiest quintile. Urban farmers are almost twice Lesotho’s agroclimatic conditions and farm structure, as likely as those that live in rural areas to produce for most households, particularly the poor, continue to sale, indicating the importance of proximity to markets. grow maize – a staple food in local diets. Wealthier Furthermore, farmers with more education are more farming households are significantly more likely to grow likely than those with a low educational attainment to vegetables and fruit. In the top quintile, 49 percent of be market-oriented (Figure 100). households grow vegetables and 9 percent grow fruit, compared to 25 percent and 4 percent of households in the poorest quintile. By contrast, poor households are relatively more likely to raise livestock and grow cereals. Figure 100: Share of market-oriented farmers, 2017 10% 8.9% % of farming households 9% 8% 7% 6% 5.3% 4.2% 4.3% 4.5% 5% 4.0% 4.0% 4% 3.1% 2.9% 2.7% 2.5% 2.6% 3% 2.2% 1.8% 1.5% 1.6% 1.7% 2% 0.7% 1% 0% Lowlands No education Primary education Poorest 20% Senqu rivervalley University Mountains Richest 20% Sec/ high school Foothills Not poor Poor Q2 Q3 Q4 Rural Urban Total Poverty Quintile Education Zone Location Status Source: Ca lculations based on the 2017/18 CMS/HBS. Crop farmers in Lesotho tend to rely on rainfed fertilizer, and pesticide application is even lower (Figure low input/low output production methods with 101). Interviews with emerging commercial vegetable limited use of irrigation, improved seed, fertilizers farmers reveal that there are significant knowledge gaps and pesticides. Less than 2 percent of Lesotho’s in variety selection, plant spacing, disease management arable land is irrigated, and reliance on rain-fed crop and application of inputs. The extension services do not cultivation reduces the growing season and yields. have the knowledge and resources needed to provide In times of drought, an entire harvest could be lost. quality and timely advice to farmers. Few farmers do Furthermore, investment in equipment and inputs is soil testing. As a result, even farmers who spend more low. Few farmers can afford the use of tractors and other on inputs do not necessarily record higher yields (World machinery. About a third of farming households use Bank 2019h). 112 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Figure 101: Share of farming households that use inputs in production, by quintile, 2017 40% 35% % of farming households 30% 25% 20% 15% 10% 5% 0% Fertilizer Insecticide Fungicide Herbicide Poorest 20% Q2 Q3 Q4 Richest 20% Source: Calculations based on the 2017/18 CMS/HBS. Limited adoption of modern commercial agricultural tons per hectare. Cereal yields have declined since 1997 practices and low investment in machinery and and the yield gap between Lesotho and neighbouring inputs contribute to low yields. Household survey South Africa has widened. In 2017, Lesotho’s cereal data on yields is scarce, yet the available evidence yields per hectare constituted only about 18 percent of suggests that yields are significantly below those South Africa’s (Figure 102). Interviews with commercial expected for the varieties planted. Average cereal yields vegetable farmers reveal that yields are typically about are below one ton per hectare, significantly lower than half of what they are expected to be and even lower for the regional yields and the SADC’s target of at least two subsistence vegetable farmers (World Bank 2019h). Figure 102: Cereal yields in Lesotho and South Africa, 1997-2017 6000 Cereal yield (kg per ha) 5000 4000 3000 2000 1000 0 1997 1998 1999 2001 2003 2005 2006 2007 2008 2009 2011 2013 2015 2016 2017 2000 2002 2004 2010 2012 2014 Lesotho South Africa Source. World Development Indicators. Low productivity in agriculture contributes to low households than for female-headed households, incomes and high poverty of farming households. which may indicate lack of access to inputs, credit Average sales of agricultural output are LSL 2,671 while and extension support for female-run farms. Sales and average profits are LSL 1,806 per growing season profits are significantly higher among better-educated (Figure 103). As might be expected, poor households households. The average profit (per season) of a have much lower sales and profits than the better off household with primary education is LSL 111 vs. LSL ones. Profits are nine times higher for male-headed 5,448 for households with university degrees. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 113 Figure 103: Average farm sales and profits per season 8 000 6 000 Maloti (LSL) 4 000 2 000 0 Sec/ high school University Poorest 20% Richest 20% None Primary Not poor Other Urban Poor Q2 Q3 Q4 Total Female Urban Urban Maseru Male Rural Lowlands Rural Mountains Rural Rural Senqu river valley Rural Foothills -2 000 Poverty Total status Quintile Education Gender Region Location Average sales Average profits Source: Calculations based on the 2017/18 CMS/HBS. Average profits per season are over nine times Access to credit could increase investments in higher in urban areas than in rural locations, inputs and support commercialization of agriculture. indicating the importance of proximity to markets. It could also enable farm households to diversify into Agricultural activities are most profitable in Maseru non-agricultural activities. About 45 percent of farming but bring negative returns in Rural Mountains. Access households took out loans, with little difference to larger consumer markets and a greater number of between the poor and the non-poor. Most of the loans formal buyers allows farmers to realize higher profits. were for personal use (e.g. household consumption, Formal buyers, such as supermarkets and local grocery school fees, medical expenses, funerals or weddings) stores, tend to pay more for high quality produce than and only 5 percent of farming households took out loans informal traders operating in rural markets. Selling for business uses, including purchasing farm inputs to formal buyers requires meeting quantity, quality, or equipment. Compared to other households, non- packaging and delivery timing requirements that many poor, urban (particularly in Maseru) and male-headed poor farming households struggle to fulfill. farming households were more likely to take out loans for business use (Figure 104). This pattern indicates the type of population groups most likely to engage in entrepreneurial activities. 114 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Figure 104: Share of farming households that took loans for business use, 2017 12% 11.4% Percentage of farming households 10.0% 10% 7.7% 8% 6.3% 6.9% 5.1% 6% 3.7% 3.3% 3.8% 4.3% 3.6% 4% 2.3% 2.1% 2% 0% Rural Lowlands Female Rural Mountains Not poor Urban Maseru Other Urban Rural Foothills Male Urban Rural Senqu river valley Rural Poor Gender of the Total Poverty Status household Region Location head Source: Calculations based on the 2017/18 CMS/HBS. Most farming households who borrowed took out collateral are among the factors that reduce farmers’ loans from their neighbors, friends and relatives (58 access to formal financial institutions. Only 232 farmers percent) and only 3.2 percent received credit from have titles to their land, and most rural land transactions commercial banks (Figure 105). The average loan size (buying or renting) take place informally (World Bank provided by relatives was LSL 684, compared to over 2018a). Lack of land titles prevents farmers from using LSL 36,000 for loans granted by banks (Figure 106). High land as a collateral to obtain bank credit. risks associated with the agricultural sector and lack of Figure 105: Source of loan for farming households Figure 106: Average loan amount, by loan that received a loan source Insurance company 0.2 Religious institution 558 Religious institution 0.3 NGO 608 Micro-finance 0.9 Relative 684 Employer (salary advance) 0.9 Neighbour/friend 1 304 NGO 2.6 Co-operative 3.1 Co-operative 1 833 Commercial banks 3.2 Other (specify) 2 931 Other (specify) 10.1 Employer (salary advance) 4 167 Relative 10.8 Micro-finance… 6 361 Money lender 21.0 Money lender 6 361 Neighbour/friend 46.9 Insurance company 11 957 0 20 40 60 Commercial banks 36 242 Share of farming households with a loan 0 10 20 30 40 Average loan amount in Maloti (thousands) Source: Calculations based on the 2017/18 CMS/HBS. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 115 Falling international migration D. and Both push factors and pull factors characterize remittances slow poverty reduction Lesotho’s migration patterns. These factors include high poverty levels in rural areas, lack of employment Migration continues to be a dominant strategy for opportunities and basic services, intensified economic maintaining households’ livelihoods in Lesotho. inequality between the rich and the poor, better Migration of skilled workers to South Africa is prevalent, opportunities and services in South Africa and porous generating a large influx of remittances. Dependency borders between the two countries (IOM/Economist on remittances has decreased over time, but it still Intelligence Unit, 2018). According to the CMS/HBS 2018 makes Lesotho’s households dependent on the South survey, the main reasons cited for these movements African economy.35 Remittances in Lesotho represent include: for work (42 percent), to live with other relatives a coping strategy against poverty. Based on the 2017 (19 percent), for school/training (15 percent), to look for Population Census, more than 328,000 Basotho were paid jobs (4 percent) and other reasons (20 percent). migrants in other countries, representing around 14.8 Compared to households in the wealthiest quintile, a percent of the population. The top three destination higher percentage of households in the poorest quintile countries of migrants from Lesotho are neighboring cited former household members looking for a paid job Southern African Development Community (SADC) as a reason for migrating. Richer households have a countries: South Africa (95.3 percent), Mozambique (2.6 greater percentage of their former household members percent) and Botswana (1 percent). However, there are in countries other than South Africa. also Basotho in the United Kingdom, United States, the Netherlands, Malaysia, Australia and other countries. A Over the years, the context of labor migration to second generation of the Basotho diaspora can be found South Africa has changed dramatically. The changes in Australia, Europe and the United States. The 2017/18 have occurred in the terms of legislation and institutional CMS/HBS data show that 98.9 percent of respondents frameworks, demographic profiles (the feminization live in a household where a former household member of Basotho migrant workers), sectors of employment has migrated to South Africa. in South Africa, conditions of recruitment and social protection. While agreements between Lesotho and South Africa have been renegotiated over time, there seems to be a growing disconnect between objectives and outcomes, pointing to the inadequacy of current agreements and the need to rethink Lesotho’s approach to bilateral labor agreements (Box 17). 35 The decline is mainly due to a sharp decline in number of migrant mineworkers employed in South Africa – from around 90,000 in 1990 to 31,000 in 2012. The adoption of the Mining Charter in South Africa in 2003 confirmed the decision to phase out foreign labour in South African mines. 116 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Box 17: History and features of Lesotho international migration South Africa has historically been the top destination country for Lesotho migrants. This is attributed to geography, with Lesotho being surrounded by South Africa. For Lesotho’s households, migration has historically been entrenched in decisions since the 19th century, led by male migration to South Africa’s mining industry (UN-INSTRAW, 2006; Botea et al., 2018). This phenomenon continued during most of the 20th century; in the 1980s, up to 40 percent of Basotho men were employed in South Africa. But by the 1990s, Lesotho-South Africa migration patterns had dramatically changed. In 2002, only 12 percent of Lesotho’s households had relatives in South Africa’s mining industry, compared to 50 percent 20 years earlier (Boehm 2003). Since then, a declining trend has been observed in Basotho male migration to South Africa because of increasing retrenchments from the mining industry in South Africa, a shift to more capital-intensive production, stagnant gold prices and pressures from the South African government to hire more locals (Motelle 2012). The halt in new hiring of Basotho men in South Africa mines threatened remittances inflows to Lesotho as a major source of household income. Therefore, households have developed new livelihood strategies that still involve migration. The decline coincided with growing female migrants from rural to urban Lesotho, employed mainly in the export- oriented garment industry, the largest formal employer in the country with employment at 90 percent female. Since independence, Lesotho entered into the three labor agreements with South Africa – all different in nature and form.36 The 1973 agreement provided a comprehensive framework regulating the recruitment and movement of Basotho workers into South Africa. It included the establishment of a Lesotho office in South Africa, plus guidelines on the movement of South African employers into Lesotho for recruitment as well as provisions regarding ports of entry, documentation, employment contracts and social benefits. The 2006 and 2013 agreements, in contrast, addressed priority areas for cooperation between the two governments but no longer regulated the above-mentioned areas. This is a significant difference that has bearing upon the objectives and implementation scope of the agreements.37 The current Lesotho Special Permit expires December 31, 2019. There are more than 100,000 low-skilled labor migrants from Lesotho working in South Africa and an unknown number of irregular migrants are job-hunting or working in inhumane conditions in neighboring countries. An increasing number of women are the main income providers for their households (Botea et al., 2018). However, the characteristics of Basotho migrant women differ from their male counterparts. Basotho women are less likely than men to engage in cross-border migration. Although the share of female Basotho migrants in South Africa doubled over 10 years, it was still only 8 percent in 2014. Basotho migrants are involved in various occupations in their destination countries. Considering both genders, the highest percentage of Basotho migrants work in domestic service (29.65 percent), followed by construction (12.85 percent), mining (11.20 percent), the informal sector (7.08 percent) and other sectors including retail, agriculture, government (39.22 percent). Female migrants are more likely to be employed in informal activities, such as domestic service (50 percent), trade and commercial agriculture (Crush et al., 2010). The feminization of migration has posed a new challenge in poverty reduction due to the multiple vulnerabilities of female migrants – their employment status, lack of contracts, lack of documentation, immigration status and issues of gender-based violence. These vulnerabilities often worsen during emergency periods, including prolonged dry spells (IOM, 2019). 36 (i) Agreement between the Government of the Republic of South Africa and the Government of the Kingdom of Lesotho relating to the Establishment of an Office for a Lesotho government Labour Representative in the Republic of South Africa, Lesotho Citizens in the Republic of South Africa and the Movement of such Persons across the International border. Treaty Series No.1/1973 (1973); (ii) Memorandum of Understanding between the Government of the Kingdom of Lesotho through the Ministry of Employment and Labour and the Government of Republic of South Africa through Its Department of Labour on Cooperation in the field of Labour (2006); (iii) Memorandum of Understanding between the Government of the Kingdom of Lesotho through the Ministry of Employment and Labour and the Government of Republic of South Africa through Its Department of Labour on Cooperation in the field of Labour (2013). 37 IOM Lesotho, Bilateral Labour Agreement Guidelines and Model Agreements for Lesotho (2017), page 8. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 117 Migration to the mines and other areas in South Lesotho but also in South Africa (ACP-EU, 2012). The Africa increases the vulnerability of migrants, two main drawbacks associated with unofficial means their partners and other community members to of sending remittances are related to the risk of losing HIV and tuberculosis infection. With the decline of money and financial exclusion of households that 83,000 migrant mine workers employed in South Africa would otherwise reap the benefits of financial inclusion. between 1987 and 2013, many Basotho households, Although the South Africa-Lesotho corridor has been particularly in rural Lesotho, have become more the least costly for sending remittances to Lesotho, lack vulnerable (IOM, 2017). The availability of commercial of affordable remittance facilities in rural areas excludes sex at South African mines led to the rapid diffusion of Basotho migrants working on farms in Ceres, Eastern HIV among the workforce in the 1990s. The burden of Cape Province of South Africa, from access to sending diseases and the weakening of household structures remittances back home (IOM / Migrant Workers are two prominent social consequences of Basotho Association, 2019). labour migration. Academic research and several reports have documented the scope and impact of Lesotho remains dependent on remittances. In occupational diseases among the Basotho workforce, 2018, remittances inflows were estimated at about particularly tuberculosis, silicosis, HIV/AIDS and various US$415 million, accounting for 14.7 percent of the types of work-related disabilities (IOM, 2017). Stigma country’s GDP. Approximately 18.5 percent of the and non-disclosure of HIV status as well as failure to surveyed households acknowledged they had received test for tuberculosis and seek treatment in time are international remittances during the 12 months prior challenges that lead to the spread of disease. Family to the 2017/18 CMS/HBS survey.38 One percent of wellbeing is negatively affected when migrants return households received remittances from both household home with illnesses and inadequate financial resources and non-household members, 13 percent received (IOM, 2018). remittances from household members and 13 percent received remittances from non-household members. Guidelines regarding how mining companies compensate employees for occupational hazard Remittances to Lesotho have always been larger were not fully followed. According to former than Official Development Assistance (ODA) and mineworkers and other stakeholders, ex-mineworkers Foreign Direct Investment (FDI) (Figure 107). and their families in most cases were shortchanged by Remittances reached a peak of US$649 million in 2011 employers. The Government of Lesotho’s current role and then started declining due to a halt in employment of in social protection for families of mineworkers and new Basotho workers in South African mines. In 2011, ex-mineworkers needs strengthening. The institutional remittances were 2.5 times larger than the amount of arrangements for paying the benefits need simplification ODA received in Lesotho, 10.6 times larger the amount and rigorous monitoring (IOM, 2018). of FDI and 1.9 times larger than the combined amount of ODA and FDI. Households mostly receive remittances through unofficial channels because of the high costs of Rural households, particularly in the Senqu sending and receiving money (World Bank 2018f). River Valley, are the most reliant on remittances. A large proportion of remittances is channeled through Constituencies closer to the border with South Africa unregulated means, such as cash carried across borders are more reliant on remittances (Figure 108). In the by vehicle operators or delivered by the migrants southernmost areas, more than half of all individuals themselves (Sekantsi, 2018). About 42 percent of reside in households with a member working abroad. In remittance recipients surveyed indicated they had the central mountainous regions, where access to South received remittances through these unofficial channels. Africa is more difficult, less than 20 percent of individuals This is attributed to the high costs of sending remittances. reside in households dependent on remittances. The lack of national identification documents for some Households with low educational attainments are also migrants and their family members hinders access to more likely to be reliant on remittances. banking and remittances transfer services not only in 38 The survey significantly underestimates the magnitude of remittances in Lesotho. This number and the subsequent analysis should be taken as lower bounds for the impacts of remittances. 118 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Poorer constituencies are more reliant on data show that households received an average of about remittances, with the mountainous constituencies LSL 5,661 per month in remittances. Figure 109 indicates being an exception. In general, the poorer the the richest quintile received higher remittances than constituency, the more reliant households are on poorer ones. Among recipient households, 64 percent remittances (Figure 109). The exception is once more reside in rural areas and 36 percent live in urban areas. the mountainous constituencies, which do not rely Urban dwellers tend to receive higher remittances than much on remittances despite high poverty levels (Figure rural households. Remittance recipients mainly spend 110). This makes these households more dependent remittances on food and clothing followed by education.39 on agricultural income and more vulnerable to weather Among non-household remittance recipients, poor shocks. households in the bottom quintile spend the greatest share of their remittances on education, followed by Remittances are an important source of income, and food and clothing. The richest quintile spend the highest poor households use the money to buy necessities proportion of their remittances on other expenses. and invest in human capital. The 2017/18 CMS/HBS Figure 107: Trends in remittances, ODA and FDI Figure 108: Share of individuals in household with remittances from South Africa 700 600 Maseru 500 US$ million Share (% ) 400 <10 300 10−20 20−30 200 30−40 40−50 100 50−60 0 1990 1995 2000 2005 2010 2015 Remittances FDI ODA Source: World Development Indicators (WDI). Source: Calculations based on the 2016 Population and Housing Census. 39 Remittances may help improve economic growth, especially if used to finance children’s education and develop human capital. Even when used for consumption, remittances generate multiplier effects, especially in countries with high unemployment. Whether remittances are used for consumption or human capital investments, they generate positive effects on the economy by stimulating demand for other goods and services. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 119 Figure 109: Average remittances received in urban Figure 110: Share of individuals in household and rural areas, by quintile with remittances and poverty, by constituency 80 12000 10000 60 Poverty rate (%) Amount in LSL 8000 40 6000 4000 20 2000 0 0 0 10 20 30 40 50 Share of individuals in household with remittances (%) 1 2 3 4 5 Quintile Mountainous constituencies Urban Rural Rural non−mountainous constituencies Urban non−mountainous constituencies Source: Calculations based on the 2017/18 CMS/HBS. Source: The 2016 Population and Housing Census and the 2017/18 CMS/HBS. Remittances incomes are strongly progressive in remittances from households receiving them would Lesotho, benefiting the poor population the most. make them poor. The concentration curve indicates Ranking households with and without remittances that remittances are strongly progressive – located well changes the distribution’s shape (Figure 111). Taking above the line of equality (Figure 112). Figure 111: Share of population getting remittances Figure 112: Remittances concentration curve by quintiles of consumption 1 30 Cumulative % of benefits Share of population receving 25 remittances (%) 20 0.5 15 10 0 5 0 0.2 0.4 0.6 0.8 1 Cumulative % of population ranked from 0 poorest to richest Q1 Q2 Q3 Q4 Q5 Poorest Richest Line of equality Without remittances With remittances Consumption per capita Remittances Source: Calculations based on the 2017/18 CMS/HBS. Consumption aggregate used for the deciles calculation is net of remittances. Remittances contribute positively to poverty and was smaller – 2 percentage points – because only inequality reduction. Households that received 18 percent of the population received the transfers. remittances would have been 13 percentage points Remittances also reduce inequality in remittance- poorer had they not received any remittances (Figure recipient households. The Gini index would be higher by 113). The overall impact of remittances on poverty 2 points without remittances (Figure 114). 120 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Figure 113: Impact of remittances on poverty Figure 114: Impact of remittances on inequality 100% 0.50 0.47 0.45 80% Poverty rate 63% 0.40 60% 50% 52% 50% Gini coefficient 0.30 40% 20% 0.20 0% 0.10 Households Receiving All households remittances 0.00 Without remittances With remittances Without remittances With remittances Source: Calculations based on the 2017/18 CMS/HBS. Note: Counterfactual analysis, re-estimation of the poverty and inequality indicators with and without remittances from abroad. The impact of remittances on poverty would have • Access to finance is among the major constraints been much larger if remittances had not decreased to local businesses. Enterprise Survey data show in recent decades. The decomposition analysis showed that 35 percent of domestic firms identify access to that the national poverty rate would have been nearly 5 finance as a major or severe constraint to operations. percentage points lower had remittances remained at Access to finance is impeded by a lack of legislation the 2002 level. Likewise, food or extreme poverty would that could reduce the risk of lending to SMEs. At the have been down by 5 percentage points had the level same time, non-bank financial institutions are not of remittances not changed. The impact on inequality well developed in Lesotho (World Bank 2018a). would have been muted because households in all parts of the distribution were reliant on remittances in 2002. • Pursuing deeper integration with South Africa, the United States and the European Union (EU) E. Conclusion and policy discussion is important for growth of Lesotho’s private sector. Lesotho’s lower labor costs and more Strengthening the private sector through reforms stable labor relations have already helped the to increase competitiveness and foster private country attract South African investment in the investments is key to creating jobs, encouraging apparel industry. A targeted investment promotion formal businesses and reducing poverty. campaign could help Lesotho attract more South African FDI into apparel-related activities as well • Policies aimed at easing the process of receiving as other labor-intensive industries. Given the large business licenses, accessing finance and dependence on the U.S. market, negotiations with lowering the costs of regulatory compliance the U.S. government on the post-African Growth would improve the business environment. In and Opportunity Act (AGOA) trade regime could give addition, supplier development programs would confidence to current investors and help avoid job enhance the capacity of local small- and medium- losses. Furthermore, the Government of Lesotho sized enterprises (SMEs) to become subcontractors may consider organizing an investment promotion or suppliers to foreign direct investment (FDI) firms. campaign targeting the EU market to build on the The programs might help increase the size of firms Economic Partnership Agreement with the trade in Lesotho and their productivity (World Bank 2018a). bloc. It will also be important to sign international investment and double taxation agreements to give greater confidence to potential investors. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 121 • The services industry is growing but could be • Improvements in agricultural productivity can be expanded to create jobs. Most jobs are concentrated achieved through transition from subsistence to in wholesale and retail trade. Agribusiness services commercial agriculture. This will involve significant could be developed for the growth of commercial public intervention and private investment. An agriculture. The services would include financing, important priority is improving the functioning of the certification, packaging, storage, distribution and land market by promoting land titling and creating a branding; they could contribute more than primary land registry. This will encourage new investment production to GDP (World Bank 2018a). in commercial farming and enable farmers to use their land as collateral. Investment in irrigation • Policies aimed at creating jobs for the poor infrastructure and rural roads will also be important. and increasing their incomes would reduce Furthermore, it will be necessary to provide training dependency on remittances and exposure to to agri-entrepreneurs on business skills, record the South African economy’s fluctuations. keeping and marketing as well as on agronomic Remittances relieve poverty because they contribute practices and climate smart agriculture (World Bank a large share to household income among poor 2018a). households, but the downside is dependency on the South African economy. • Lesotho has considerable potential for increasing incomes and creating jobs through diversification • Targeted policies that boost entrepreneurship from cereals to higher-value crops, such as fruit and promote skills development will increase and vegetables. A favorable climate, strong local labor productivity and reduce poverty. High demand and opportunity for import substitution unemployment rates and high youth unemployment contribute to the good conditions for cultivation of arise from skills mismatches. Training should fruits and vegitables. Vegetables are scale-neutral concentrate on skills in high demand by firms, such and can be profitably cultivated on plots of less than as sewing machines repairs and ICTs. One way to do a hectare. Some commercial vegetable farmers that this is through introducing a Supplier Development use irrigation and have greenhouses or shade nets Program aimed at enhancing the capacity of local report per hectare profits of above 1 million maloti per SMEs to become subcontractors or suppliers for FDI season. In addition, fruit and vegetable production firms. It will also be important to partner with the is more labor-intensive than cereal farming and can private sector to address skills constraints as well create new jobs. Lastly, there is strong demand for as raise awareness about the existing tax incentives fresh produce among formal buyers in Lesotho, for training. and local farmers are cost competitive with imports (World Bank 2019h). Providing matching grants Adopting Climate Smart Agriculture (CSA) to agricultural entrepreneurs with strong potential principals in Lesotho should (i) increase productivity and support to vertical and horizontal alliances (as and incomes, (ii) enhance food security and envisaged by the second phase of the World Bank’s dietary diversity, (iii) reduce impacts of climate SADP project) can further encourage investments in change on agricultural produce and (iv) improve climate smart technologies and agricultural activities commercialization, employment opportunities with higher value added. and rural livelihoods. The Lesotho CSAIP aims to identify investments that offer the greatest potential • Commercialization of agriculture will also involve to transform agriculture into a more productive, building linkages between farmers and buyers. climate-resilient and low-emissions sector. The CSAIP This could be done through a Supplier Development offers two complementary pathways for scaling up Program with the goals of strengthening linkages with CSA in Lesotho. Commercialization can be prioritized buyers, providing technical assistance to farmers and in lowlands and foothills, while the highlands would supporting local agro-dealers/extension services. The benefit from resilient landscape, or afforestation, and program will involve continuous monitoring of buyer farmer-managed natural regeneration to restore and demand, including such considerations as types of replenish less fertile land. crops, volumes needed, quality (size, color, stage of ripening) and delivery timing requirements. Technical 122 Lesotho Poverty Assessment I Progress and challenges in reducing poverty assistance to farmers will include seed variety • To reduce over-reliance on one country, Lesotho selection, cultivation practices, soil nutrient testing should look for alternative destination countries and amelioration, pest and disease management, for Basotho migrant workers. New bilateral water management, product standards and financial agreements, such as the ongoing efforts to have management. Working with the local agro-dealers a Memorandum of Understanding (MoU) between and extension services to provide better support to Lesotho and Mauritius, should be encouraged. In farmers will also be useful. February 2018, the two countries identified the most appropriate technical instruments in a Declaration of It is essential to improve migration governance Intent and a MoU. by developing a national strategy with a capacity to prevent irregular migration, promote legal • Reduced costs would ease the sending and migration and mobility, and ultimately enhance receiving remittances.40 This can be achieved synergies between migration and development. through financial-sector development, led by encouraging more entrants into the remittances • It is important to manage irregular migration to market in Lesotho. Subsequent competition would South Africa by facilitating cross-border safety likely drive down the cost of sending and receiving and legal migration. Creation of an institution remittances. Strengthening awareness campaigns for the Basotho diaspora has the potential to give on remittances transfer services could also help. migrants an opportunity to coordinate and run joint Information on the range of money transfer products, developmental programs. A bilateral labor agreement costs and requirements could be disseminated between Lesotho and South Africa would create through public and private radio stations, national greater labor mobility and ease credit constraints. television, newspapers, magazines and the internet Negotiations between the two countries are taking in both Sesotho and English (ACP-EU, 2012). place. 40 Remittances are private not public funds and are not a panacea for the country’s development needs. They can only be a complement. Given the high relative importance of remittances, there is a high risk of overdependence on them. All measures taken to leverage remittances need to be accompanied by measures creating a favorable economic, legal and political environment to foster human development. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 123 Part III: Vulnerability to shocks and poverty reduction in Lesotho III Chapter III.1: Vulnerability to economic and environmental shocks Households in Lesotho face significant environmental and economic shocks. The most commonly reported shocks over the period 2012–2017 were drought and high food prices. The higher prevalence of shocks with large impacts on welfare in rural areas results in poorer households being exposed to greater risk. In many instances, households do not have good coping mechanisms to manage shocks. When hard times hit, the most common coping strategy is relying on help from family and friends. Consumption is often reduced in response to experiencing this type of shock, with potential long-run consequences. Rainfall shortfalls have a significant impact on poverty, particularly for rural households. Had rainfall in 2015-2016 been at the historical average, rural poverty would have been 6 percentage points lower. Regular pension support helps pensioners mitigate the impact of weather shocks on consumption, suggesting the transfers are adequate. Other safety nets targeted at rural households mitigate the impact of rainfall, but the effect is marginal, leaving these beneficiaries very susceptible to weather risk. Irrigation mitigates the impact of rainfall shortfalls on nutritional outcomes. A. Defining vulnerabilities in Lesotho vulnerability not adequately covered by these programs are identified, providing guidance for public policy to Households in Lesotho, like many others across strengthen households’ resilience. To the extent the data Sub-Saharan Africa, face significant uncertainty allow, an assessment is also made of how households with regards to future consumption. This uncertainty can reduce their exposure to important sources of risk. has many sources – for example, illness that affects the productive labor and income of a household, death of Vulnerability to weather risk has historically been a family member, drought that reduces crop yields and high among rural households in Lesotho. The incomes and price shocks that make basic food supplies predominant source of income for many households more expensive for those with few resources. is rain-fed agriculture; much production is from maize, which makes livelihoods very susceptible to rainfall This chapter develops an understanding of the shortfalls at specific periods during the production sources of households’ vulnerability in Lesotho and cycle. Changes in livelihoods may make households the strategies for coping with shocks. It examines the more resilient: irrigation use in farming is increasing, government programs, markets and informal networks the agricultural base is diversifying and households are households rely on to manage the consequences of transitioning to other income sources. unexpected shocks to consumption. The sources of Lesotho Poverty Assessment I Progress and challenges in reducing poverty 125 Box 18: Measuring vulnerability to shocks in Lesotho Vulnerability to poverty can be described as the threat of future poverty (Calvo and Dercon, 2013). There are two components to vulnerability analysis (Hoddinott and Quisumbing 2003): first the likelihood of adverse events occurring and, second, the impact on wellbeing, should the shock materialize. Because vulnerability is about wellbeing in the future, it is important to consider both households’ current welfare, which determines how well it can manage risks, and its exposure to risks in the future. The analysis in this paper incorporates (i) structural constraints that drive food insecurity and poverty and (ii) exposure to shocks that increase poverty and food insecurity and make currently food secure, non-poor households vulnerable to food insecurity in the future. There are different conceptual frameworks for measuring vulnerability. The Household Economy Approach is “a livelihoods-based framework for analyzing the way people obtain access to the things they need to survive and prosper” (Bodreau et al. 2008). This approach underpins the livelihood baselines used in the Lesotho Vulnerability Assessment and Analysis Report (LVAC) to generate estimates of the numbers of people needing emergency assistance each year. Here, the approach set out in Hill and Porter (2017) is followed, with vulnerability to poverty assessed by the impact of shocks on consumption and the probability of shocks occurring from cross-sectional and historical data. This approach follows the vulnerability measures used in Pritchett et al. (2000), Chaudhuri (2003) and Calvo and Dercon (2013) and is distinct from other approaches that have assessed downturns in the wellbeing indicator (Kamanou and Morduch, 2002; Dutta et al., 2011) or vulnerability as welfare loss due to increased variance (Ligon and Schechter, 2003). Between 2016 and 2018. households in Lesotho experienced both weather and food-price shocks. These types of events are often termed as covariate shocks and affect many households in one place at the same time. This chapter uses the variation in the severity of these shocks across time and in different parts of Lesotho to assess their impacts on welfare. The perfect dataset would include several years of observations for each household; even better, it would information on what could happen and how probable this is/was in differing states of the world. In the absence of panel data and such scenarios, information on the spatial and historic distribution of shocks is used to estimate how weather and price shocks affected consumption and nutrition outcomes during 2016-2018. Economists have recently incorporated rainfall data as exogenous sources of shocks in various empirical models to estimate ex-post impacts of weather (Dell, Jones, & Olken, 2014). Conditional on the probability of drought, the timing of the shock is exogenous. We control for the probability of drought by including measures of the mean and variance of the historic drought distribution in a given location and with district fixed effects. Such shocks as death and non-contagious illness are not typically experienced by multiple households in the same location at the same time. They affect one or two individual households rather than whole communities. The frequency of these shocks is also usually constant across years. Events like these are termed idiosyncratic shocks. B. Basotho face severe environmental and The high frequency of weather and price shocks economic shocks reflect abnormally bad weather conditions and high inflation during the survey period. Figure 116 In terms of frequency, the most important sources shows that rainfall conditions have been quite bad in of risk are environmental and macroeconomic the past five years, with two years at or below the 20th shocks: drought and high food prices. This is true percentile of rainfall during the peak December and when looking at the past year and when looking at the January maize pollination season, a time when yields past five years (2012-2017). Crop and livestock disease are most susceptible to rainfall losses. Since 2001, the were also reported frequently (Figure 115). In addition worst year for recorded rainfall was 2015-2016, when to the presence of specific disease pathogens, this can Lesotho experienced one of the strongest El Niño’s reflect more challenging weather conditions. in recent history. In fact, 2016 was classified as a 1 in 126 Lesotho Poverty Assessment I Progress and challenges in reducing poverty 30-year event. Rainfall was low, poorly distributed and difficult than normal for most households during the late to arrive. Crop production was estimated to be 62 period considered in the survey questions. percent lower than the previous year and 51 percent below average (LVAC Report, 2016). The CPI reported Data collected during the annual Lesotho very large price increases in the middle of 2016, partly vulnerability assessments confirm that 2016-2017 because of harvest losses experienced both in Lesotho was a particularly bad period for households in and regionally (Figure 117). Prices had come back down Lesotho. Data collected on food security and nutrition by the time households were interviewed, starting in outcomes confirms that a lot of households were January 2017, but the price spikes were within the predicted to be food insecure in 2016-2017, an expected reference period of one year for respondents. Together, result of the poor harvest in 2016 (Figure 118). this suggests that the economic environment was more Figure 115: Frequency of shocks experienced by Figure 116: Rainfall was much lower during peak household (last five years and in the last year) pollination in 2016 80 of households experiencing... 70 60 Total rainfall in peak maize pollination period (December-January) 50 400 350 40 300 Total rainfall (mm) 30 250 200 20 150 10 Percent 100 50 0 0 s ic ) r e en od sts ld e h b er old r be ag od 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 be ho ol gr fo e m w eh m flo p m a se t me es f e f p us so me on or m to ou ro er ho e ily n ng rc ts en ld c ( rw am i de re gh ho rki eo r ym np th o ilu se ou rf tsi o as pa ie d n w ei a dr i ou u he sf ise th ho or n ris kd ot no m es b ir pd ad of oc ge ro of in or he nt sf cro us t lar th es nt e ce b a f id liv o me de old an c h ac at loy itt eh de or em mp us ss o rr ee h e do illn ag ai fw re e, ve so nc se ta los ic/ s is ron as ch ar ul eg fr do en Last five years Last year Source: Calculations based on the 2017/18 CMS/HBS. Source: Calculations based on CHIRPS 2001-2017. Note: Total rainfall in peak maize pollination period (December- January). Lesotho Poverty Assessment I Progress and challenges in reducing poverty 127 Figure 117: Inflation was high in 2016 Figure 118: Food insecure population by year, rural Lesotho 350% 300% 800 000 250% 700 000 200% 600 000 150% Number of people 500 000 100% 400 000 50% 0% 300 000 -50% 200 000 -100% 100 000 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 0 2009/10 2013/14 2014/15 2015/16 2016/17 2017/8 Food Total Source: IMF. Source: Lesotho Vulnerability Assessment Report 2017/18. Note: Percent change in CPI (compared to same month in previous year). After drought and price shocks, ill-health and death differences in livelihoods. Owning land or engaging in were the most commonly reported shocks. Death of agriculture makes households much more vulnerable a household head or other working-age member was to climate shocks and livestock-related shocks. Not reported by 9 percent of households, and death of other surprisingly, rural households are more likely to report family members by 22 percent of households. In addition, droughts and crop and livestock diseases. In the past 21 percent of households were affected by a household five years, 81 percent of those who own land reported members’ chronic or severe illness or accident. being negatively affected by drought, compared to Collectively, this is a large number of households. In the 55 percent of those who do not. For livestock related past five years, 16.3 percent of households experienced shocks, the comparable figures are 36 percent and 10 loss of wage employment and 8.4 percent experienced percent. Urban households are more likely to report the failure of a non-agricultural business. Other shocks – loss of non-agricultural business or wage employment. fire, being jailed or a household breakup – were also In addition, urban households are more likely to report reported, but this category was less common. chronic or severe illness in the household, perhaps reflecting the fact that illnesses have a larger effect in The nature of shocks varies between rural and urban the urban areas’ more formalized labor markets. Loss Lesotho, reflecting different risks to livelihoods. The of household members is reported as frequently in risks vary considerably across rural and urban Africa urban and rural areas. Similarly, large increases in food (Christiaensen, Nikoloski and Hill, 2017). This is also the prices were reported as frequently by rural and urban case in Lesotho (Figure 119). The differences reflect households. 128 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Figure 119: Percentage of households affected by Figure 120: Percentage that occurred in the risk by urban/rural past five years by poverty status end of regular assistance, de ath of other family death of working member death of household head death of household head birth in the household birth in the household chronic/severe illness or crop disease or crop pests large rise in prices of food household business failure end of regular assistance livestock died or were stolen loss of wage loss of wage employment death of other family household business chronic/severe illness or livestock died or were droughts or floods crop disease or crop pests large rise in prices of food droughts or floods 0 10 20 30 40 0 20 40 60 80 100 Rural Urban Poor Non poor Source: Calculations based on the 2017/18 CMS/HBS. A higher share of poor households suffers from The shocks that matter most for welfare are not droughts or floods. This may reflect the fact that rural necessarily the ones most frequently reported. households tend to be poorer. When considering shocks Households were asked whether the shock impacted that occurred over the past five years, 78 percent of income or assets or both. The most severe shocks could poor and 68 percent of non-poor reported suffering from be those that affect assets and income. Taking this as drought or flood (the share is comparable when looking the measure of the shock’s severity indicates that loss at shocks that occurred over the past year – 30 percent of the household head was the most severe shock, for poor and 22 percent for non-poor) (Figure 120). More closely followed by death of other working member of than 80 percent (83 percent) of households headed the household and other family members (Figure 121). by someone with no education report experiencing drought or flood, compared to just above half (54 percent) for households headed by someone with a tertiary education. For the shocks that occurred in the past 12 months, the corresponding proportions are 30 percent and 21 percent. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 129 Figure 121: Proportion of households that lost Figure 122: Price and weather risk are the most income and/or assets as a result of a shock covariate sources of risk households face death of household head death of household head death of working livestock died or were… death of working member… birth in the household death of other family… loss of wage household business… household business crop disease or crop pests end of regular assistance droughts or floods chronic/severe illness or loss of wage employment… death of other family chronic/severe illness or… livestock died or were end of regular assistance,… crop disease or crop pests large rise in prices of food droughts or floods birth in the household large rise in prices of food 0% 50% 100% Both assets and income Assets Income 0 0.5 Share of variance that can be explained by enumeration area fixed effects la st year last five years Source: Calculations based on the 2017/18 CMS/HBS. Note: The right figure shows the results of a regression of the reported shock on village fixed effects. The higher the covariate the nature of the shock, the larger the degree to which shocks are explained by these fixed effects. C. Households have few coping strategies at these types of shocks particularly difficult to deal with. their disposal to manage shocks Death of other family members and the loss of wage employment were also bad. The most severe shocks may not be the ones that cause the largest income or asset losses; the biggest Households often report reducing food consumption blows can be the shocks households find harder in the face of a shock. The share of households stating to cope with. A death in the household, drought and they reduced their food intake (or substituted preferred loss of wage employment were consistently ranked by food for cheaper food) is fairly large. Looking at shocks households as the hardest shocks to deal with. Nearly that occurred in the past five years, for example, 19 all households that experienced the death of their head percent of households stated they cut food intake and ranked it as one of the worst two shocks they had 14 percent substituted cheaper but less preferred food experienced. For those who experienced it, the death (the share of households is also large when looking at of a working household member was also ranked as shocks that occurred in the past year). Interestingly, very severe. This was followed by drought. The shocks there is no stark difference in the share of urban and that are most likely to affect multiple households in the rural households who report reducing food intake in same community at the same time are drought and response to a shock, perhaps because this was a period price risk. Figure 122 shows that drought and food price with both crop income losses in rural areas and high increases are the most covariate shocks. This makes food price inflation in rural and urban areas alike. 130 Lesotho Poverty Assessment I Progress and challenges in reducing poverty The high numbers of households reporting It is not the poorest or the most well-off that a reduction in food intake is concerning. For employ consumption-based coping strategies, but households with lower levels of consumption prior the households in between, who are both affected to the shock, reductions in consumption can lead to and able to reduce food consumption intake. Figure undernutrition, with poor health consequences in the 123 reports data collected through the LVAC survey short-run and important consequences in the long-term. in 2016 and shows overall worsening consumption These include a high risk of stunting, impaired cognitive because of poor harvests and high inflation. However, development, lower school attendance rates, reduced the very poorest or the better off households indicated human capital attainment and higher risks of chronic little change in the frequency of eating less preferred disease and health problems in adulthood (Black et al., food or in the frequency of reducing adult consumption. 2013; Hoddinott et al., 2013). Those who are stunted The figures also highlight that the variation in the use of average 1.6 fewer years of education, and have 0.625 consumption-based coping strategies across households standard deviation lower outcomes on cognitive tests. is largely driven by the wealth of the household. As a result, Galasso and Wagstaff (2017) estimate the cost of childhood stunting to be 9-10 percent of GDP per capita for countries in Africa and Asia. Figure 123: Change in consumption behavior as result of shock, by consumption level 3.0 3.0 Reduced adult consumption Ate less preferred food 2.5 2.5 2.0 2.0 1.5 1.5 1.0 1.0 0.5 0.5 0.0 0.0 Very poor Poor Middle Better off Very poor Poor Middle Better off Source: Calculations based on LVAC 2016. Those who are educated, who have access to agricultural conditions (Hill and Mejia 2017). Households irrigation or who have a source of remittance with migrants in the year prior to the survey were income are less susceptible to risk. Consumption also better able to withstand shocks (Figure 124). For of households with no education drops by 38 percent these households, the absolute loss from a shock the in the face of a bad shock, compared to 11 percent size of 2016 would be LSL 60, less than 10 percent of for households with a secondary education or above. consumption. The nutritional status of children living In the LVAC, women with higher levels of education in households that use irrigation is less susceptible to are less susceptible to rainfall risk. Having education rainfall risk. This reflects the fact that access to coping makes households more resilient because they can strategies varies across households, making them less diversify their sources of income in the face of poor vulnerable. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 131 Figure 124: Households with education, migrants and irrigation are less susceptible to rainfall losses addition mm of rainfall in December and 40% 0.05% Consumption per adult equivalent 35% Percent increase in MUAC for each 30% 0.04% 25% 0.03% 20% 15% 0.02% lost (percent) 10% 0.01% 5% January 0% 0.00% secondary education primary education irrigate rainfall Secondary plus Five Second Primary No Fourth Third None None Yes First Education Land quintile Migrant Child Woman Source: Calculations based on the 2017/18 CMS/HBS (left) and LVAC 2016-18 (right). Note: The size of rainfall shock used in the left graph was a loss of 68mm, representing the difference between the much lower rainfall in 2016 and median rainfall. The loss was calculated using coefficients from a regression of rainfall on consumption, interacting with different household characteristics. MUAC is mid-upper arm circumference. Shocks have a large impact on welfare because much impact. The shocks that households were more households have few coping strategies at their likely to act on were deaths in the household or shocks disposal to manage shocks. Most households do not that resulted in loss of non-agricultural income – loss take specific actions to manage the shocks they report of wage employment, the loss of household business experiencing, and poor households are less likely than income or the end of regular assistance. These are also other households to act to manage shocks. Figure the most idiosyncratic risks, easier to manage through 125 shows that nearly three-quarters (72 percent) of local markets and networks. A larger share of the poor households facing a shock in the past five years reported did not do anything in the wake of the shock, which they had done nothing to manage it, and this proportion could suggest that they do not have any means of was only slightly lower for shocks experience in the coping with the shock. For shocks in the last five years, past year (62 percent). Compared to urban households, 76 percent of poor and 67 percent of non-poor reported a higher share of rural households responded that doing nothing. The gap is nearly as large—67 percent they did nothing in the wake of a shock. This could compared to 58 percent—for shock episodes that be because households had few actions available to occurred in the past year. manage the shocks or because the shocks did not have 132 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Figure 125: How households manage shocks in Lesotho 90 Percentage of households 80 70 60 50 40 30 20 10 0 s s ey sh n s O t r ks ng nd ie en e es io G th on ca g oc pt i ie m N n th te O si m m st /fr ve rn no m ra bu su ily ve ed or sa st fro id m n w go w ts n to D co lp fa ne io rro se he m d pt d m e/ te as Bo o fro m or fro d fo ar su of ve m St lp d lp on i s ce he ce d he le ke rc du Re Sa ed or ed e Re iv th W iv ce O ce Re Re Non poor Poor Total Source: Calculations based on the 2017/18 CMS/HBS. However, households are sometimes able to use of households headed by someone with no education networks and markets to manage the impact of relied on help from family and friends, compared to income shocks. Receiving help from family and friends 15 percent of households whose head has a tertiary is the most commonly reported coping mechanism in education. the 2017/18 CMS/HBS. About a quarter (24 percent) of households that experienced a shock over the past None of these coping mechanisms—informal five years – and 17 percent over the past year – report networks, credit, asset sales or increased labor relying on informal networks of friends and family. market activity—were used much in the face of Compared to urban households, a larger share of rural covariate shocks that affected many households households relies on these informal networks. For in the same area at the same time. This highlights shocks experienced in the past five years, 26 percent the challenge of covariate shocks. The initial shock of rural households and 20 percent of urban households is compounded by the fact that fewer labor market have received help from friends and family. This pattern opportunities are available in local markets, fewer was also present when considering the most recent willing local buyers of stocks or non-productive assets shock episode: 19 percent of rural households received and fewer lenders available in localized credit markets. help from friends and family, compared to 13 percent of urban households. Urban households and better off households have more coping mechanisms available. A higher share Informal networks are used more by the non-poor, of urban households relies on working longer hours, the less educated and female-headed households. starting a new business or saving cash – all coping When considering shock episodes in the past year, 19 mechanisms that are arguably easier to apply in an percent of the non-poor and 15 percent of the poor relied urban setting. A higher share of non-poor household on informal help to cope with the effects of the shock. adopts saving cash or borrowing money as a coping However, it is less educated households that are more mechanism. likely to rely on help from family and friends: 21 percent Lesotho Poverty Assessment I Progress and challenges in reducing poverty 133 D. The impact of climate-related shocks is were negatively impacted by livestock death or theft. significant, especially among agricultural Adverse weather events have often been associated households with the reduced yields and increases in food prices reported by 69 percent of households. These shocks Climate change, with its prospects for frequent affected a similar share of poor and non-poor farmers. droughts and heavy seasonal floods, is a key Given the large role of agriculture in the livelihoods of challenge to the agriculture sector. The country also Basotho, weather shocks have a major negative impact experiences hail and frost. Lack of irrigation and weak on production, incomes and food security in both rural farmer knowledge of climate-smart agricultural practices and urban areas. Over half a million people are currently exacerbate the negative impacts of climate change. in need of food assistance due to poor rainfalls in 2018- Over 80 percent of households reported that their 2019. In addition to the adverse weather events, many financial situation was severely impacted by droughts farming households have also reported experiencing or floods during the past five years, while 30 percent other shocks, such as death or illness of a family suffered from crop diseases or pests (Figure 126). member, loss of wage employment and increases in Furthermore, about 37 percent of farming households agricultural input prices. Figure 126: Share of farming households exposed to shocks, by type of shock and poverty status, 2017 90% Percentage of farming households 80% 70% 60% 50% 40% 30% 20% 10% 0% Drought or Large rise in Livestock died Crop diseas e or Death of other Chronic/severe Loss of wage Large rise in flood price of food or stolen crop pests family member illness of a employment agr. input family member prices Climate-related shocks All Not poor Poor Source: Calculations based on the 2017/18 CMS/HBS. Rain shortfalls have a large impact on consumption households with secondary education or above (Figure in rural households, particularly those with little 127). Education makes households more resilient education and no migrants. For example, the 2016 because they can diversify their sources of income in rainfall shock reduced consumption by 23 percent the face of poor agricultural conditions (Hill and Mejia, for the median rural household.41 Consumption of 2017). Households with migrants in the year prior to the households with no education dropped by 38 percent survey were also able to better withstand shocks. in the face of a bad shock, compared to 11 percent for 41 These estimates come from a regression of consumption on satellite rainfall estimates for the main growing season prior to the month for which consumption data was collected. Rainfall totals from December and January, the period when maize is most susceptible to rainfall shortfalls, were used as a measure of the severity of the shock. Data collection spanned 13 months, so the very poor rainfall conditions in 2016-2017 were captured for some households. Although the timing of a rainfall shock is exogenous, the exposure to rainfall shocks in general is not exogenous. To control for this, the first and second moments of the historical rainfall distribution for each location were included. This allows the impact of rainfall shortfalls on consumption to be considered causal (see Dell et al. (2012) for a discussion of this approach and Hill and Porter 2017 for an application to Ethiopia). The full method and results are detailed in Hill et al. (2019). 134 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Figure 127: Impact of rainfall shocks across types of rural household, 2017 40% Consumption per adult equivalent 35% 30% 25% 20% 15% lost (percent) 10% 5% 0% Yes Secondary plus First Fourth Five Third Second None None No Primary Education Land quintile Migrant Source: Calculations based on the 2017/18 CMS/HBS. Rainfall risk appears to impact those without land and their consumption is more susceptible to shocks. and large farmers equally. Those without land may be Farmers are more susceptible to rainfall risk compared poorer but more likely to have income sources outside to non-farmers, but there is no significant difference in of agriculture. Those with more land are better off, the impact of rainfall shocks on consumption across but their income is more concentrated in agriculture farmer types. Figure 128: Poverty rates had rainfall been normal National Rural/urban 61.3 60.7 60 60 56.6 54.6 49.7 Poverty rate (%) Poverty rate (%) 50 50 44.6 41.5 40 40 28.5 30 30 25.3 20 20 2002 2017 2002 2017 National, actual Rural, actual Urban, actual National, with normal rainfall Rural, w. normal rainfall Urban, w. normal rainfall Source: Calculations based on the 2017/18 CMS/HBS. Rural poverty would have declined by 6 percentage then rural poverty would have declined by 6 percentage points if rainfall had been normal. The urban-rural points. Urban poverty would also have declined, but divide increased from 2002 to 2017, with urban poverty more modestly at 3 percentage points (Figure 128). decreasing while rural poverty stagnated. This is partly Yet, even with normal rainfall, a large urban-rural divide due to rural areas being more vulnerable to the low would remain in Lesotho. rainfalls experienced in 2016. Had rainfall been normal, Lesotho Poverty Assessment I Progress and challenges in reducing poverty 135 E. Existing social protection programs are shocks occur. A national and international response limited in the degree to which they help was initiated in response to the El Niño drought of households mitigate the impact of shocks 2015-2016. The Government of Lesotho mobilized US$21 million through budget reallocation, from which The social protection system is well set-up to US$10 million were used for sectoral response and an help households manage life-cycle sources of additional US$11 million to provide a food price subsidy. vulnerability because there is a high coverage More than US$40 million was raised from humanitarian of children and elderly. However, few households partners. In addition, the World Bank mobilized US$20 report receiving help from the government or non- million through the Crisis Response Window (CRW) governmental organizations (NGOs) in the face of and US$1.4 million through the Contingent Emergency shocks. Households that experienced drought, crop Response Component (CERC) under the Smallholder disease or chronic illnesses were more likely to report Agriculture Development Project (SADP). Cash transfers receiving help from government or NGO resources. and emergency funds were used to support the affected These are situations in which informal networks, labor population groups. The resources were delivered to markets and financial markets struggle to provide the the affected people through social protection schemes required support to affected households. But too few as well as emergency response systems. The food households have benefited from public support. subsidy was found to have limited impact for vulnerable populations. The resources would have been better The underutilization of public support persists spent on cash transfers (World Bank, FAO and WFP despite the significant government and humanitarian 2017). response often put in place when large covariate Figure 129: The impact of a rainfall shock on households receiving different types of public transfers Loss of consumption per adult equivalent 200 150 Maloti 100 50 0 No Yes No Yes No Yes Social protection Pension NGO help Source: Calculations based on the 2017/18 CMS/HBS. Note: The size of rainfall shock used was a loss of 68mm which represents how much lower rainfall in 2016 was compared to median rainfall. The loss was calculated using coefficients from a regression of rainfall on consumption, interacting with the receipt of different transfer programs. 136 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Analysis shows that pensions help households The 2015-2016 El Nino was particularly bad with mitigate the impact of shocks, but existing reductions in consumption, increasing poverty rates social protection programs are limited as coping by 5 percentage points. But it was not a one-off mechanisms. Regression results are used to estimate event. Climatic shocks will continue to impact Lesotho, the impact of the 2016 shock on rural households that likely with increasing frequency and severity as a did or did not receive three types of transfers: social result of climate change. Preparing for these events is protection, pensions and NGO support.42 Figure 129 essential, and some policy options are discussed below. graphs the results from the regressions of rainfall The key is for investments to be made now, before a shocks interacted with whether an individual received crisis, so that resource mobilization and planning do not transfers from the government or NGOs. Social start after the crisis, delaying an effective response. protection transfers reduced the impact of rainfall Hill et al. (2019) estimate that the cost of failing to get shocks on beneficiaries – but only a bit and did not a response in place in time to meet the consumption eliminate the impact. This was also true for help from needs of those suffering from drought is 3.9 percent NGOs. By contrast, a pension eliminated a household’s lower income (GDP) per capita in the long run. The gain vulnerability to rainfall risk. Rural households receiving from an emergency response that is one month quicker pension support are richer than other rural households is 0.8 percent of income per capita in the long run. – so those without pensions experience a larger relative loss than those with pensions. This could be part of The most cost-effective way to reduce vulnerability the reason pension recipients fare better in shocks. In is to reduce exposure to shocks (Hill 2019). The addition, pension income was constant and sizeable, analysis has shown that investments in irrigation and reducing any impact of variable agricultural income on education both have a potential to do that. Irrigation consumption. reduces the impact of low rainfall on agricultural output, and education increases the ability of individuals to F. Conclusion and policy discussion earn income in other, less-rainfall dependent sectors, either before or in response to the shock. Increasing This chapter has shown that life is risky for investments in education and irrigation are thus an many Basotho, and the period 2016-2017 was a important part of a plan to make Lesotho more resilient in challenging one for the many households that an increasingly uncertain world. The impact of the 2015- experienced severe drought and high food prices 2016 drought contributed to, and was compounded in addition to the health shocks typical present. by, the rise in food prices. This points to a role for The higher prevalence of shocks with large impacts lowering barriers to agricultural trade and facilitating on welfare in rural areas results in poorer households private investments in storage as steps toward reducing being exposed to greater risk. When shocks occur, exposure to price risk. households have few available coping mechanisms. This is particularly true for shocks that affect many Better financial market development can spread risk households in the same community or market at once. beyond the immediate social network and social As a result, consumption is often reduced in response protection, increasing the ability of households to to this type of shock, increasing poverty and producing manage the remaining risks. Agricultural insurance long-run consequences on human capital attainment for could protect farmers against disasters by transferring children. In 2016. the drought resulted in a 23 percent risks to credit markets. It could also help increase reduction of consumption for rural households. farmers’ agricultural productivity and access to financial services. The Government of Lesotho could look at policy options to support the expansion of agricultural insurance. 42 Social protection includes school feeding, cash for work assistance and food aid given as part of public assistance. These programs benefit 14.1 percent, 3.8 percent and 2.8 percent of the population respectively (Boko et al. 2019). Ideally, this category would have included the child grants received by 8.8 percent of the population, but data on this program was not collected in the household survey. Pensions capture all those receiving social pensions (comprising 4 percent of the population) or other forms of pensions (military, civil service, etc.) as well those receiving state support through disability grants. Those receiving secondary school and university bursaries were not included, and other programs not captured by the household survey were also not included. From the data presented in Boko et al. (2019), the main programs are included in the analysis. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 137 Social protection policies are another means for transparent rules for scaling-up support to households. governments to help households manage these It also requires pre-financing arrangements to finance risks. These programs provide a dependable source of the added benefits. The Disaster Risk Financing household income that is not subject to risk (for example, Diagnostic (World Bank 2019d) shows that the existing pension support) and can be scaled up to provide more policy framework for disaster financing in Lesotho does support in hard times. The government and international not have a systematic strategy to mobilize financing for community did increase support in response to the disaster response. The most recent El Niño drought in most recent crisis. The analysis in this paper shows the 2015-2016 was addressed through humanitarian aid response did help to ameliorate the shock a bit, but the and budget reallocations, which took considerable time impact was limited. Further reforms to social protection and diverted resources from high-yielding development systems can help, such as improving the coverage and projects. Scaling up would also requires having all being ready to scale up in the time of a shock. Box 19 potential beneficiaries in a national database. Scaling details some policy recommendations coming out of an up is easier for existing beneficiaries already in social OPM review of the latest response. Being ready to scale protection systems, but this analysis shows that non- up requires having an early warning system that provides beneficiaries were also exposed to shock, pointing to accurate information on when and where scale-up is the need to expand the list of potential beneficiaries. required and triggers that can be used to develop clear 138 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Box 19: Strengthening the ability of social protection to help households manage shocks A case study produced by researchers at The Oxford Policy Management (OPM) in cooperation with the Overseas Development Institute (ODI), the Cash Learning Partnership (CaLP) and INASP examined how the social protection system could become more shock-responsive. The study’s main recommendation were: • Support to routine social protection. Improvements to the regular functioning of the core social protection programs and delivery systems, including the Child Grant Programs (CGP) and Old Age Pension (OAP) (expanding coverage, updating beneficiary lists, improving payment systems, etc.), will make a considerable difference in their efficiency and effectiveness when they are used in a crisis. • Support to the emergency response mechanism in place. Strengthening the Disaster Management Authority (DMA) and its ability to deliver on its mandate will enhance its value in a crisis. Today, resources do not match the system that is meant to be in place. Steps forward could include identifying a mechanism to assure the emergency fund has money available to distribute in times of need. • Planning and preparedness. The crisis is not the time to design the mechanism. Once the immediate emergency is over, preparatory work can begin on the ways (if any) social protection programs and delivery systems might make useful contributions in response to future shocks. This should take into account not only what was done by the government and its partners during the response to El Niño, but also what was not done (use of the National School Feeding Programme, OAP, wider NISSA database). Integrate consideration of non-contributory social assistance programs into contingency plans, including an indication of how to provide resources for the expanded programs. • Coordination. Spell out more clearly the ways in which the social protection sector should feed into the DMA’s structures and processes for coordinating emergency response. • Anticipation and analysis of financing needs for crises. Strengthen the process for anticipating financing requirements on a routine basis through better preparedness and contingency planning processes, supported by available data and/or by an improved early warning system. • Development of the NISSA. While NISSA is undergoing reform, reach a decision on four major aspects determining its relevance in a shock: (i) Comprehensiveness: how will households not on the NISSA be reached? (ii) Accuracy: how can households report changes to their material circumstances that affect their categorization in the database? (iii) Accessibility: how can the District Disaster Management Teams easily use the data? This may require investment in infrastructure and training. • M&E of interventions and the use of delivery systems. Conduct reviews of the emergency interventions implemented during the El Niño crisis, including cost-efficiency analyses if possible. Prepare a framework of measurable criteria to gauge the efficiency and effectiveness of future interventions (extending to measures of outcomes, i.e. their benefit to households, not only indicators of the efficiency of outputs, such as timeliness of disbursement). Lesotho Poverty Assessment I Progress and challenges in reducing poverty 139 Chapter III.2: Social protection and poverty and inequality in Lesotho The social protection system plays an important role in mitigating poverty and reducing inequality in Lesotho. The country spends about 4.5 percent of GDP on social protection, higher than most countries in Sub-Saharan Africa. The social protection system has contributed significantly to decreasing the poverty gap and to reducing inequality, with cash transfers playing the most important role. The impact of social protection could further be enhanced through efforts to continuously improve harmonization and synchronization of programs, strengthen inter-agency collaboration and continue efforts to improve programs delivery. A. Social protection programs in Lesotho most of the public programs, implemented by various government agencies, whose objectives also include Lesotho has a comprehensive social protection direct or indirect income/consumption support to system, and the country has continuously pursued their beneficiaries. As an example, the school feeding actions to improve its efficiency, efficacy and program’s main objective is to improve school attendance equity. The right to social protection is enshrined in (school feeding is normally universal, i.e. it does not the constitution, which establishes the government’s exclude any child.). At the same time, it is an indirect responsibility to actively promote and maintain the income support to households whose children benefit welfare of its citizens. The National Social Protection from school meals. Similarly, the Tertiary Education Strategy (NSPS 2014/15-2018/19) for Lesotho, Bursary Program is an investment in tertiary education developed by the Ministry of Social Development is a key and may not be specifically targeted at poor students strategic document to guide the sector’s development. only. At the same time, it does provide income support The goal of the Strategy is to provide comprehensive to its beneficiaries. The programs were grouped into the and inclusive social protection that reduces poverty, following five categories: (1) in-kind programs (School vulnerability and inequality, increases resilience to Feeding Program); (2) public work programs (cash-for- risks and shocks, promotes access to services and to work or public works program – Fato Fato); (3) cash the labor market and stimulates economic growth and transfers (Old Age Pension, Child Grant Program, Public social stability. Assistance Program); (4) social insurance programs (Civil Service Pension, Public Officers Defined Contribution This chapter of the Poverty Assessment focuses Pension Fund43 and Workmen’s Compensation Fund44); on programs captured in the 2017/18 CMS/HBS to and (5) education subsidies (Orphans and Vulnerable better understand their performance in terms of Children Bursary and Tertiary Bursary). Box 20 provides coverage of the poor, targeting accuracy and impact an overview of the programs used in the analysis. on poverty and inequality. The programs represent 43 Public Officers’ Defined Contribution Pension Fund. The Public Officers’ Defined Contribution Pension Fund has been established under The Public Officers’ Defined Contribution Pension Fund Act of 2008. 44 This program is part of Lesotho’s labor programs, but it could also be classified as a social insurance program. 140 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Box 20: Lesotho’s Social Protection Programs • The Child Grant Program (CGP) was launched in 2009 with the technical and financial support of the European Union and UNICEF. Households qualify for the cash grant if they have children below the age of 18 and they are classified as ultra-poor or poor based on community targeting and proxy means testing administered under the National Information System for Social Assistance (NISSA). The program provides a monthly benefit of LSL 120 for households with one to two children, LSL 200 for households with three to four children and LSL 250 for households with five children or more. Benefits are provided on a quarterly basis, following a predetermined timetable. The program covered 37,000 households across all 10 districts in 2014 and expanded to close to 50,000 in 2018. The total budget of the program is LSL 58 million. • The Public Assistance Program (PAP) is the country’s oldest social assistance, with the objective of providing destitute people cash and in-kind support on a temporary (up to six months of benefits) or permanent basis (for example, cases of severe disability). PAP beneficiaries are self-targeted, but decisions on eligibility and the benefit amount rests mostly with the Ministry of Social Development district managers in the country’s 10 districts, based on the ministry’s guidelines. The PAP delivers a quarterly benefit between LSL 250 and LSL 500. Benefit delivery is done at the district level by social welfare officers. In fiscal year 2018-2019, the program reached a total of approximately 12,000 beneficiaries, or 0.5 percent of the population, for a total budget of about LSL 41 million. • The Orphans and Vulnerable Children (OVC) Bursary program was established in 2000 to support access to education (especially secondary education) for orphans and vulnerable children. The program is mainly self-targeted, but it follows established guidelines, with beneficiaries being children under 18 years old who have lost one or both parents or whose parents are sick, disabled or incarcerated. The NISSA has been used to target OVC bursary beneficiaries. Applications are received across the country and processed through the district OVC Bursary officers working under the Ministry of Social Development. Decision on the allocation of benefits is made based on targeting guidelines and a predetermined quota by district until available resources are exhausted. Until 2017, the OVC was implemented in parallel with a similar program, the Post Primary Bursary managed under the Ministry of Development Planning. Since the beginning of 2018, the two programs have been merged into a single one under the administration of the Ministry of Social Development. As of 2018, the program covers approximately 23,000 beneficiaries. • The Old Age Pension (OAP) was established in 2004 to prevent Basotho elderly from falling into destitution. The OAP is a universal, non-contributory social pension, targeted to every Basotho aged 70 years old and above not receiving a civil service pension. The OAP is the country’s largest safety net program, and it is managed by the Ministry of Finance’s Department of Pensions. OAP applications are received at the district level all over the country and submitted to the Pensions Office in Maseru for processing. OAP beneficiaries receive a monthly cash grant of LSL 750, disbursed by Ministry of Finance payment officers through 293 pay points. The benefit amounts have rapidly increased since the establishment of the program – from LSL 150 in 2004, to LSL 540 in 2015, LSL 700 in 2017, and LSL 750 in April 2019. • The School Feeding Program is overseen by the Ministry of Education and Training in cooperation with the WFP. It is one of the oldest programs in Lesotho to support attendance and nutrition of children. It covers all children in public primary schools and early childhood care and development (ECCD) centers across Lesotho by providing one or two meals at school. The program supports two main goals: better nutrition for children in primary schools and ECCD centers and retention and improved school outcomes for program beneficiaries. School feeding in Lesotho is divided between WFP (covering 200,000 pupils) and the government (which caters for another 200,000). Since 2011 the government has been paying WFP to manage approximately 55 percent of the program in the hard to reach parts of the country. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 141 • The Watershed Management Public Works, also known as the Fato Fato program, is administered by the Ministry of Forestry, Range and Soil Conservation. The Fato Fato serves the dual objectives of environmental conservation and income support through the cash-for-work approach. Through Fato Fato, households are offered a maximum one month of work per year to participate in conservation activities. The program employs a total of 69,000 people across Lesotho – 3.2 percent of the population – at a monthly wage of LSL 1,200 (or LSL 60 per day for a period of 20 days per month), one month per year. Public works activities include a range of land management and soil conservation activities, such as planting trees for fuel and wood, planting fruit trees, land rehabilitation (rehabilitation of rangeland and grass cover, removal of invasive species, construction of infiltration ditches, terracing to reduce erosion) and water harvesting (installation of roof tanks, water storage and dams). Farmers are provided with free seeds to protect eroding soils. The program is self-targeted and not restricted to the poor. The monthly wage of LSL 1,200 is high compared to the national upper-bound poverty of LSL 648.88 per adult equivalent per month in 2017-2018 prices.45 • The Tertiary Bursary Program is one of the biggest government programs in Lesotho. It was launched in 1978 to support skills development and employability of Basotho citizens by providing partial scholarships to qualified students who intend to pursue higher education either in Lesotho or abroad. The program is managed by the National Manpower Development Secretariat (NMDS), under the Ministry of Development Planning. It supports approximately 23,500 students. The program covers tuition, the cost of research, text books allowance, accommodation, a food allowance and travel cost to and from the location of training. The program was allocated LSL 641 million in FY18/19, the second highest allocation for a single transfer program, behind only the OAP. • The Public Officers’ Defined Contribution Pension Fund is established under the Public Officers’ Defined Contribution Pension Fund Act of 2008. The Fund provides government employees with pension benefits upon leaving the government services. The fund works like a saving account with joint government’s and employees’ contributions. Investment managers invest and manage the funds. The following public service employees are eligible to benefit from the Fund: servants at disciplinary services, local government, public service and teaching services. 45 WFP Lesotho supported the development of new guidelines on effective implementation of public works programs, in line with integrated catchment management principles, and is supporting the implementation of the other proposed reforms through pilot initiatives in Maseru, Botha-Bothe, and Leribe. An evaluation of the pilot is underway, and lessons learnt are intended to support wider reforms of the program. WFP also supported, in collaboration with the World Bank, knowledge exchange with the Ethiopia Productive Safety Nets Public Works from which lessons could also be learnt to improve the Fato Fato (World Bank 2019f). 142 Lesotho Poverty Assessment I Progress and challenges in reducing poverty The performance of the above listed programs that countries in Sub-Saharan Africa. There are 10 main are provided to the population by the Government social protection programs in Lesotho,48 compared to an of Lesotho is measured by targeting efficiency, average of 15 programs per country in the Africa region adequacy of the programs and their impact on and 20 programs in the lower middle income countries poverty and inequality.46 One of the key elements is (LMICs) (Figure 130). Despite a relatively small number to look at whether and to what extent the poor benefit of programs, Lesotho spends an average of 4.5 percent from the public support programs. The performance of GDP on social protection programs. This is higher is determined not only by the programs’ design, but than the averages of 1.9 percent in the Africa region also by resources invested in their implementation and 1.8 percent in LMICs (Figure 131).49 The spending and very importantly by the efficiency and efficacy of is dominated by two programs. The Old Age Pension the program’s implementation. In addition to looking (OAP) is the largest spending category, accounting at the performance of all programs classified as social for 1.9 percent of GDP and 42 percent of total social protection programs for the purpose of this analysis, protection spending (Figure 132).50 It is followed closely this chapter looks separately at the performance of the by the Tertiary Bursary Program, which accounts for 1.7 social assistance programs.47 percent of GDP and 38 percent of total social protection spending. The two programs account for 80 percent of B. Public spending on social protection total government spending on social protection. The third largest spending category is the School Feeding Compared with other African countries, Lesotho Program, accounting for 0.5 percent of GDP and 11 has fewer social protection programs; however, percent of the total social protection spending. resources allocated to them are higher than in other 46 It is important to empasize that reduction of poverty is not nececerly the target of every program. Some programs are universal, while thers are not targeted to address poverty. In this analysis we do not distinguish whether the program raches its objective, but measure to what extent poor benefit from them. 47 The World Bank classifies social protection programs into three main clusters of programs: (i) social insurance or contributory social protection programs (e.g. old age, survivors’ and disability insurance, unemployment insurance; work accident insurance). The function of these programs is to mitigate the risk of income loss due to old age, disability, unemployment or loss of a breadwinner; (ii) social assistance/ social safety net programs or non-contributory programs whose function is help people cope with poverty, risks and vulnerability to prevent them from falling into destitution and engaging into negative risk coping strategies; and (iii) labor market programs whose function is to build resilience and strengthen people’s capacity engage in gainful economic activities. 48 The following Lesotho social protection programs were included in this comparison: School Feeding Program, cash-for-work or public works program (Fato Fato), Workmen’s Compensation Fund, Old Age Pension, Child Grant Program, Public Assistance Program, Civil Service Pension, Public Officers Pension, Orphans and Vulnerable Children Bursary, and Tertiary Bursary. 49 Note that global social safety net average spending is 1.5 percent of GDP, according to The State of Social Safety Nets 2018 report: https://openknowledge.worldbank.org/handle/10986/29115. 50 Civil service pensions are considered part of social insurance, and they are not part of the social safety net analysis. However, to put it in the overall SP (social insurance + social safety net) context, expenditures on the civil service pensions are about 0.95 percent of GDP. Hence, total SP spending in the country is about 5.5 percent of GDP. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 143 Figure 130: Number of social protection Figure 131: Spending on social protection programs programs, percentage of GDP 25 5.0 4.5 Percentage of GDP 20 4.5 20 4.0 3.5 15 15 3.0 2.5 1.9 10 1.8 10 2.0 1.5 1.0 5 0.5 0.0 0 Lesotho Africa Lower middle- Lesotho Africa Lower middle- income countries income countries Source. For Lesotho: Calculations based on the 2017/18 CMS/HBS. ASPIRE database for other countries: http://datatopics.worldbank.org/ ASPIRE/ Figure 132: Top 5 social protection programs by total expenditure, percentage of GDP 2.0% 1.9% 1.8% 1.7% 1.6% Percentage of GDP 1.4% 1.2% 1.0% 0.8% 0.6% 0.5% 0.4% 0.3% 0.2% 0.2% 0% School Old age Integrated OVC Tertiary feeding pension water burcaries bursary program management schemes public work program Source: Calculations based on the 2017/18 CMS/HBS. Note: The spending on in-kind food assistance and postsecondary TVET bursaries, the two remaining SP programs account for only 0.04 percent of GDP. C. Program coverage and efficiency The overall coverage of social protection programs the majority of Basotho families having children who is high in Lesotho, mostly due to the School Feeding benefit from it. Social pension, a universal benefit for all Program. Close to 71 percent of the population lives elderly persons, also shows relatively higher coverage in a household with at least one member who benefits (15.5 percent). Fato-Fato cash for work program also from a social protection program, leading to high overall has a relatively large coverage (13.5 percent). The coverage (Table 13). The School Feeding Program, a Child Grant Program is benefiting 9.1 percent of the universal program covering children in public primary population. Coverage of social insurance (contributory and secondary schools has the largest coverage, with pensions and Workmen’s Compensation Fund) is 144 Lesotho Poverty Assessment I Progress and challenges in reducing poverty low, with only 1.7 percent of the population living in a (85.4 in the second quintile) reside in households where household receiving this benefit. The coverage of the at least one member benefits from social protection bottom two quintiles by social protection is excellent: programs. 92.1 percent of the population in the bottom quintile Table 13: Coverage of social protection programs, by quintile and poverty status Quintiles of per capita consumption, net of all     Poverty Status cash transfers Non-   Total Q1 Q2 Q3 Q4 Q5 Poor poor All social protection 70.5 92.1 85.4 74.4 61.5 39.1 86.8 52.3 In-kind subsidy 59.8 75.8 76.6 64.0 51.6 31.3 74.8 43.8 School Feeding Program 59.8 75.8 76.6 64.0 51.6 31.3 74.8 43.8 All public work programs 13.5 16.8 18.9 15.1 10.9 5.9 17.8 8.8 Cash-for-work or public works 13.5 16.8 18.9 15.1 10.9 5.9 17.8 8.8 program (Fato Fato) Cash transfers 22.6 45.2 25.9 19.8 15.7 6.5 32.0 12.6 Old Age Pension 15.5 35.4 16.0 11.7 9.8 4.5 22.4 8.0 Child Grant Program 9.1 17.4 11.4 8.4 5.8 2.8 13.1 4.9 The Public Assistance Program 0.8 1.4 0.8 0.8 0.7 0.2 1.1 0.5 All social insurance 1.7 0.4 0.4 1.5 2.5 3.5 0.6 2.9 Civil Service Pension, and Public 1.2 0.4 0.3 0.4 1.9 2.8 0.3 2.2 Officers Pension Workmen’s Compensation Fund 0.6 0.3 0.2 1.1 0.6 0.7 0.4 0.7 Education subsidies 6.0 1.2 2.3 5.4 9.1 11.8 2.8 9.5 Orphans and vulnerable children 1.5 0.9 1.2 2.5 2.3 0.6 1.4 1.6 bursary Tertiary Bursary 4.5 0.3 1.2 2.9 6.8 11.2 1.4 7.9 Source: Calculations based on the 2017/18 CMS/HBS. Note: Program coverage is the portion of the population in each group living in households that receives the transfer. While many programs operate at the individual level, they contribute to overall household consumption, indirectly benefiting all members of a household. Quintiles are based on the consumption aggregate net of all cash transfers (pre-transfers income). It should be noted that coverage of the programs also depends on the program design. For example, the coverage of the Old Age Pension is determined by the number of elderly 70 plus years of age; Tertiary Bursary is determined by the number of tertiary education students. Same for orphans, etc. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 145 The social protection coverage of the poorest are very low: among individuals of postsecondary age is very high: 86.8 percent of those living in poor (ages 19–22), 3.3 percent receive university bursaries households receive some form of social protection. and 1.3 percent receive postsecondary TVET bursaries. The survey-based estimates indicate that the school Coverage strongly favors the non-poor. While 9.5 feeding program has the highest coverage among poor percent of the non-poor live in the households receiving households, reaching close to 74.8 percent. Close to 32 education subsidies, only 2.8 percent of the poor get percent of the poor have access to the cash transfers, this payment. Very few of the poor children go into and 18 percent of the poor households have access to tertiary education. Low coverage reflects low number the Fato Fato program. of OVC in secondary and very few children attending universities, mostly from better off households. Transfers to support education have relatively small coverage and benefit mostly the non-poor due to the The coverage of cash transfers is comparable to adverse selection. Coverage rates for school bursaries that for Sub-Saharan Africa but lower than in other targeted to orphans and vulnerable children show LMICs. The cash transfers reach 23 percent of the that very few secondary school students receive the population, compared to an average of 20 percent for benefit. On average, 6 percent of the population lives Sub-Saharan Africa and 31 percent for LMICs (Figure in households with access to scholarships (Table 13). 133). Cash transfers reach 46 percent of the poorest The program’s overall coverage rate among secondary quintile, significantly higher than the Sub-Saharan Africa school students is 4 percent. For university and average (25 percent) but comparable to the LMICs postsecondary TVET bursaries, overall coverage rates average (44 percent). Figure 133: Coverage of cash transfers, Figure 134: Transfer amount received by a percent, 2017 group as a share of the total consumption, 2017 50 50 46 44 45 40 40 35 31 Percent 30 Percent 30 25 23 25 20 20 20 15 10 10 5 0 0 Total Q1 Q2 Q3 Q4 Q5 Total population Poorest quintile In-kind subsidy All labor market programs Lesotho Sub-Saharan Africa LMICs Cash transfers All social insurance Education subsidies Source: Calculations based on the 2017/18 CMS/HBS for Source: Calculations based on the 2017/18 CMS/HBS. Lesotho, ASPIRE database for other countries. 146 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Social protection transfers contribute significantly • The public works program is slightly progressive. to the living standards of the poor. Government Although not targeted at the poor (see program programs account for almost 13.9 percent of households’ description above), the cash-for-work program total consumption (Figure 134). The figure increases to benefits the poor more (Figure 135). Among the 47.9 percent for the poorest quintile, in other words, workers in the program, 16.8 percent come from almost half of the consumption in the poorest 20 households in the poorest quintile, compared to 5.9 percent of the population comes from social protection percent from the top quintile. programs. Here, cash and in-kind transfers play the most important role, while scholarships and the social • Education bursaries are regressive in Lesotho. pension (OAP) are more important for richer quintiles. Both the university and Orphans and Vulnerable As a comparison, cash benefits cover 21 percent of the Children bursaries are regressive (Figure 135). consumption of the average poor person across LMICs, University bursaries are the most regressive compared to almost 37 percent in upper middle-income program, as students from richer households are countries. more likely to attend colleges: affluent children fare better in high school, graduate more frequently Social protection as a whole is progressive in and usually have parents who themselves pursued Lesotho, mainly driven by social assistance and higher education. in-kind transfers. Concentration curves grouped by types of programs and each individual program are • Social insurance programs have very small shown in Figure 135. Figure 135 shows that the social coverage, and they are generally regressive. As protection system as a whole is above the equality line, earlier discussed, only 1.7 percent of the population suggesting progressivity. However, not all programs report access to the Civil Service Pension, Public are strongly progressive. Public work programs are, Officers Defined Contribution Pension Fund and on average, neutral, while social insurance (mostly civil Workmen’s Compensation Fund. The coverage of service pensions) and stipends for secondary education these programs is regressive, benefitting mostly of orphans and vulnerable children and for tertiary richer segments of the population. As with education education are regressive. subsidies, the likelihood of having access to social insurance coverage is higher for better-off formal • Cash and in-kind transfers are strongly workers. progressive in Lesotho. This includes the Public Assistance Program (highly progressive), the Child Grant Program and Old Age Pension (Figure 135). Similarly, the school feeding program is progressive, providing more benefit to the poor population (Figure 135). Lesotho Poverty Assessment I Progress and challenges in reducing poverty 147 Figure 135: Concentration curves for social protection programs (Based on pre-transfer income, direct and indirect beneficiaries) All social protection Cash transfers Cumulative percentage of benefits 1 0.9 1 Cumulative percentage of benefits 0.8 0.9 0.7 0.8 0.6 0.7 0.5 0.6 0.4 0.5 0.3 0.4 0.2 0.3 0.1 0 0.2 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0.1 Cumulative percentage of population, ranked from poorest to 0 richest 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Line of equality Cumulative percentage of population, ranked from poorest to richest All social protection In-kind transfers Line of equality Old Age Pension All social assistance All social insurance All Education programs Child Grant Program The Public Assistance Program Line of equality Cash-for-work or public works program (Fato Fato) Public work programs In-kind transfers 1 Cumulative percentage of benefits 1 Cumulative percentage of benefits 0.8 0.9 0.8 0.6 0.7 0.6 0.4 0.5 0.4 0.2 0.3 0.2 0 0.1 0 0.2 0.4 0.6 0.8 1 0 Cumulative percentage of population, ranked from 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 poorest to richest Cumulative percentage of population, ranked from poorest to richest Line of equality Line of equality School feeding program Cash-for-work or public works program (Fato Fato) Social insurance: Civil Service Pension and Public Education subsidies Officers Pension 1 1 Cumulative percentage of benefits Cumulative percentage of benefits 0.9 0.9 0.8 0.8 0.7 0.7 0.6 0.6 0.5 0.5 0.4 0.4 0.3 0.3 0.2 0.1 0.2 0 0.1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 Cumulative percentage of population, ranked from poorest to 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 richest Cumulative percentage of population, ranked from poorest to richest Line of equality Line of equality All social insurance Line of equality Tertiary Bursary Program Workmen’s Compensation Fund Social assistancee Orphans and vulnerable childen bursary program Source: Calculations based on the 2017/18 CMS/HBS. Notes: Ranks are defined based on the consumption aggregate net of all cash transfers. 148 Lesotho Poverty Assessment I Progress and challenges in reducing poverty The social protection beneficiaries’ distribution once they reach secondary education. Making sure that across quintiles suggests that progressivity and all children attain secondary education will increase the targeting efficiency could be improved. This would program’s progressiveness. The OAP is universal for require a multi-sectoral approach. For example, a very the elderly 70 plus who have no civil service pension. few orphans and vulnerable children receive education It is a progressive program. It could be made more grants and those that receive them come from better progressive if limited to low income elderly, but this is a off households. This likely reflects their low progression sensitive political economy decision. to secondary education in general. Increasing the number of OVC coverage would require a multi layered Households frequently benefit from several approach, including tutoring and mentoring and better programs and the overlap is not monitored. Figure social oversight (services from social workers in 136 present the transfer frequency in the total population collaboration with teachers, etc.). Having more OVC and by poverty status. The multitude of programs is in the program would improve beneficiary and benefit a good feature of Lesotho’s social protection system, incidence and make the program less regressive and as different programs have different objectives. Also, eventually progressive. Similarly, the school feeding some programs, such as school feeding, or OAP are program is universal – yet the coverage is at about 75 universal as they pertain to all children and all elderly. percent. One of the most likely determinants is the fact The issue that deserves attention is that 13.5% of the that poor children, in particular boys, drop out of school poor report not receiving any transfers. Figure 136: Transfer frequency in each population group, percent 100 2.3 5.2 7.9 10.0 Share of total population (%) 90 18.7 80 26.8 70 39.7 60 50 46.0 40 51.8 30 47.9 20 30.1 10 13.5 0 Total Poor Non-poor 0 1 2 3 more Source: Calculations based on the 2017/18 CMS/HBS. Social protection implementation arrangements information management. Most are manual and paper- can be improved: Multiple ministries administer safety based, leading to significant inefficiencies. It can take net programs with limited coordination among them months for approval of applications. This results in on both the policy and administrative levels. Paper- administrative duplication and inefficiencies. A possible based application processes for some programs (e.g. solution is to establish a unified web-based platform the Public Assistance Program and Old Age Pension) for application and delivery of benefits, which would are lengthy and result in unnecessary costs and delays. simplify the application and enable cross sharing of Each ministry involved has its own application process information. However, for that, Lesotho needs much for eligibility determination, intake/registration and better access to electricity and broadband. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 149 D. Impact of social protection on poverty and The inequality-reducing impact of social assistance inequality is significant. Social transfers reduced Lesotho’s Gini coefficient by an estimated 7.8 percent (Figure 138). In The impact of social protection programs on other words, the Gini would be 3.8 percentage points poverty is strong. In 2017, social assistance transfers higher without the transfers. This is a significant impact reduced the poverty headcount rate by an estimated on inequality, greater than in many other counties. In 6.5 percent (a 3.1- percentage point reduction) and the other countries contained in the ASPIRE database51, poverty gap by 21.9 percent (a 5.3- percentage point the Gini coefficient is reduced by 1.7 percent by social reduction) (Figure 137). In reducing the poverty gap by assistance transfers, and the reduction is, on average, 21.9 percent, the poverty-reducing impact of Lesotho’s 0.7 percent in Sub-Saharan Africa. Cash transfers social assistance system ranks ahead of the average for play the most significant role in reducing poverty and Sub-Saharan African countries (15 percent). inequality. Figure 137: Simulated poverty reduction associated Figure 138: Simulated reduction in inequality with social protection programs, percent (Gini) associated with social protection programs, percent 1. All social protection 6.5 7.8 1. All social protection 2. In-kind subsidy 1.9 1.8 2. In-kind subsidy School feeding program 1.9 1.8 School feeding program 3. All labor market programs 0.8 0.8 3. All labor market programs Cash-for-work or public works… 0.5 0.5 Cash-for-work or public works… 4. Cash transfers 3.9 5.1 4. Cash transfers Old Age Pension 3.5 4.4 Old Age Pension Child Grant Program 0.2 0.5 Child Grant Program The Public Assistance Program 0.1 0.1 The Public Assistance Program 5. All social insurance 1.3 0.4 5. All social insurance Workmen’s Compensation Fund 0.3 0.3 Workmen’s Compensation Fund Civil Service Pension, and… 1.1 0.2 Civil Service Pension, and… 6. Education subsidies 3.7 1.5 6. Education subsidies Orphans and vulnerable children… 0.1 0.1 Orphans and vulnerable… Tertiary Bursary 3.6 1.4 Tertiary Bursary Source: Calculations based on the 2017/18 CMS/HBS. 51 ASPIRE: The Atlas of Social Protection Indicators of Resilience and Equity http://datatopics.worldbank.org/aspire/ 150 Lesotho Poverty Assessment I Progress and challenges in reducing poverty E. Conclusion and policy discussion etc.). Having more OVC in the program would improve beneficiary and benefit incidence and make the program Lesotho’s social protection system plays an less regressive and eventually progressive. One of the important role in reducing poverty and inequality. most likely determinants is the fact that poor children, The country spends about 4.5 percent of GDP on in particular boys, drop out of school once they reach social assistance, with significant impacts on poverty secondary education. Making sure that all children and inequality reduction. As a whole, the system is attain secondary education will increase the program’s progressive. However, not all programs are strongly progressiveness. The OAP is universal for the elderly 70 progressive. Public work programs are, on average, plus who have no civil service pension. It is a progressive neutral, while social insurance and education stipends/ program. It could be made more progressive if limited grants are regressive (although tertiary bursaries have to low income elderly, but this is a sensitive political a strong impact on poverty reduction). The gains in economy decision. poverty reduction could be increased, in particular by improving the programs’ administration. Lesotho needs to continue to improve the social protection delivery system. Putting in place a better Understanding the factors behind some of the integrated targeting system to identify the poor and programs’ weak progressivity and some programs’ grant them benefits, would improve the coverage of the regressive distribution is the first step to improve poor and beneficiary and benefit incidence. For example, performance. Solutions are likely to require multiple means testing is done only for the Child Grant Program. interventions by various agencies. For example, about For other programs, eligibility criteria and information 13 percent of the poor are not covered by any program. are difficult to verify. As a result, eligibility is essentially Understanding why that is the case would enable subjective. An important improvement would involve the government to take action to possibly reach 100 a good targeting mechanism, based on a means test percent of the poor. Increasing the OVC coverage would and linked across programs. An expanded National require a multi layered approach, including tutoring Information System for Social Assistance (NISSA) could and mentoring and better social oversight (services serve as the main targeting tool for social protection from social workers in collaboration with teachers, programs. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 151 References ACP-EU Migration Facility. 2010. The remittances Bedi, Tara, Aline Coudouel, and Kenneth Simler. 2007. framework in Lesotho: Assessment of policies and “More than a pretty picture: using poverty maps to programmes promoting the multiplier effect. design better policies and interventions.” Washington DC: The World Bank Group. Adams Jr., Richard H., and John Page. 2005. “Do international migration and remittances reduce poverty Beegle, Kathleen; Christiaensen, Luc; Dabalen, in developing countries?” World Development, 33(10): Andrew; Gaddis, Isis. 2016. Poverty in a Rising Africa. 1645-1669. Washington, DC: World Bank. African Union Commission (AUC) and the NEPAD Blinder, Alan S. 1973. “Wage Discrimination: Reduced Planning and Coordinating Agency (NPCA). 2016. The Form and Structural Estimates.” Journal of Human Cost of Hunger: The Social and Economic Impact of Resources, 8 (4): 436–455. Child Undernutrition on Lesotho Vision 2020. Bertoli, Simone, and Francesca Marchetta, 2014. Akobeng, Eric. 2016. “Out of inequality and poverty: “Migration, remittances and poverty in Ecuador.” The Evidence for the effectiveness of remittances in Sub- Journal of Development Studies, 50(8): 1067-1089. Saharan Africa.” The Quarterly Review of Economics and Finance, 60: 207-223. Black, Robert E., Cesar G. Victora, Susan P. Walker, Zulfiqar A. Bhutta, Parul Christian, Mercedes De Onis, Alkire, Sabine, and James Foster. 2011. “Counting and Majid Ezzati. 2013. “Maternal and child undernutrition Multidimensional Poverty Measurement.” Journal of and overweight in low-income and middle-income Public Economics, 95(7–8): 476–487. countries.” The Lancet, 382(9890): 427-451. Alkire, Sabine and Maria Emma Santos. 2014. Boudreau, T., M. Lawrence, P. Holzmann, M. O’Donnell, “Measuring Acute Poverty in the Developing World: L. Adams, J. Holt, L. Hammond, and A. Duffield. 2008. Robustness and Scope of the Multidimensional Poverty “The practitioners’ guide to the household economy Index.” World Development, 59: 251–274. approach.” London: The Food Economy Group, the Regional Hunger and Vulnerability Programme, Save the Antwi, Samuel, Ebenezer Fiifi Emire Atta Mills, and Children. Xicang Zhao. 2013. “Remittances and the poverty nexus: Evidence from Ghana.” International Journal of Boehm, Christian. 2003. The Social Life of Fields: Labour Academic Research in Economics and Management Markets and Agrarian Change in Lesotho. Journal for Sciences, 2(1): 206. Interdisciplinary and Cross‐Cultural Studies, 3: 1‐18. Azam, Muhammad, Muhammad Haseeb, and Bose, Indira, Giulia Baldi, Lynnda Kiess, and Saskia de Shamzaeffa Samsudin. 2016. “The Impact of Foreign Pee. 2019. “The “Fill the Nutrient Gap” analysis: An Remittances on Poverty Alleviation: Global Evidence.” approach to strengthen nutrition situation analysis and Economics & Sociology, 9(1): 264. decision making towards multisectoral policies and systems change.” Maternal & child nutrition. Azevedo, Joao Pedro, Gabriela Inchauste, Sergio Olivieri, Jaime Saavedra, and Hernan Winkler. 2013. Botea, Ioana Alexandra; Compernolle, Nell; Chakravarty, “Is Labor Income Responsible for Poverty Reduction? Shubha. 2018. “Female Migration in Lesotho: A Decomposition Approach,” Policy Research Working Determinants and Opportunities.” Policy Research Paper Series, 6414, The World Bank. Working Paper, 8307. Washington, D.C.: World Bank Group. 152 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Brown, Richard P.C., and Eliana Jimenez. 2008. Dell, Melissa, Benjamin F. Jones, and Benjamin A. Olken. “Estimating the Net Effects of Migration and 2012. “Temperature Shocks and Economic Growth: Remittances on Poverty and Inequality: Comparison of Evidence from the Last Half Century.” American Fiji and Tonga.” Journal of International Development, Economic Journal: Macroeconomics 4 (3): 66–95. 20(4): 547-571. Dutta, Indranil, James Foster, and Ajit Mishra. 2011. Bruni, Lucilla Maria, Jamele Rigolini, and Sara Troiano. “On measuring vulnerability to poverty.” Social Choice 2016.  Forever Young? Social Policies for a Changing and Welfare, 37(4). Population in Southern Africa. World Bank. Elbers, Chris, Peter Lanjouw, and Phillippe G. Leite. Brunori, Paolo, Paul Hufe, and Daniel Gerszon Mahler. 2008. “Brazil within Brazil: testing the poverty map 2018. The Roots of Inequality: Estimating Inequality of methodology in Minas Gerais.” World Bank Policy Opportunity from Regression Trees. Policy Research Research Working Paper, 4513. Working Paper, 8349. Washington, D.C.: World Bank Group. Fields, Gary S. 2003. “Accounting for Income Inequality and Uts Change: A New Method, with Application to the Calvo, Cesar, and Stefan Dercon. 2013. “Vulnerability Distribution of Earnings in the United States.” In Worker to individual and aggregate poverty.” Social Choice and Well-being and Public Policy. Emerald Group Publishing Welfare 41(4): 721-740. Limited. Chaudhuri, Shubham. 2003. “Assessing vulnerability to Firpo, Sergio, Nicole M. Fortin, and Thomas Lemieux. poverty: concepts, empirical methods and illustrative 2009. “Unconditional Quantile Regressions.” examples.”  Department of Economics, Columbia Econometrica, 77(3): 953-973. University, New York. Galasso, Emmanuela, and Adam Wagstaff. 2017. “The Chaudhuri, Shubham, Jyotsna Jalan, and Asep Economic Costs of Stunting and How to Reduce Them.” Suryahadi. 2002. “Assessing household vulnerability to World Bank Policy Research Note. poverty from cross-sectional data: A methodology and estimates from Indonesia.” Discussion paper no. 0102- Günther, Isabel, and Kenneth Harttgen. 2009. 52. New York: Columbia University. “Estimating households’ Vulnerability to Idiosyncratic and Covariate Shocks: A Novel Method Applied in Christiaensen, Luc, and Lionel Demery, eds. 2018. Madagascar.” World Development, 37(7): 1222-1234. Agriculture in Africa: Telling Myths from Facts. Directions in Development. Washington, DC: World Bank. Gustafsson, Björn, and Negatu Makonnen. 1993. “Poverty Remittances in Lesotho.” Crush, Jonathan, Belinda Dodson, John Gay, Thuso Journal of African Economies, 2(1): 49-73. Green, and Clement Leduka, eds. 2010. “Migration, Remittances and ‘Development’ in Lesotho.” Southern Hill, Ruth V., and Carolina Mejia-Mantilla. 2017. With a African Migration Programme. Little Help: Shocks, Agricultural Income, and Welfare in Uganda. The World Bank. Datt, Gaurav, and Martin Ravallion. 1992. “Growth and Redistribution Components of Changes in Poverty Hill, Ruth V., and Catherine Porter. 2017. “Vulnerability Measures: A Decomposition with Applications to to Drought and Food Price Shocks: Evidence from Brazil and India in the 1980s.” Journal of Development Ethiopia.” World Development, 96: 65-77. Economics, 38(2): 275-296. Hill, Ruth V., Zlatko Nikoloski, and Tao Tao. 2019. Deaton, Angus, and Salman Zaidi. 2002. Guidelines “Assessing sources of vulnerability to poverty in for constructing consumption aggregates for welfare Lesotho.” Background paper for the Lesotho Poverty analysis. Vol. 135. World Bank Publications. Assessment. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 153 Hoddinott, John, and Maria Agnes R. Quisumbing. 2013. Ligon, Ethan, and Laura Schechter. 2003. “Measuring Data Sources for Microeconometric Risk and Vulnerability vulnerability.”  The Economic Journal 113(486): Assessments. World Bank, Social Protection. C95-C102. Hoddinott, John, Jere R. Behrman, John A. Maluccio, López-Calva, Luis F., and Eduardo Ortiz-Juarez. 2014. Paul Melgar, Agnes R. Quisumbing, Manuel Ramirez- “A vulnerability approach to the definition of the middle Zea, Aryeh D. Stein, Kathryn M. Yount, and Reynaldo class.” The Journal of Economic Inequality 12(1): 23-47. Martorell. 2013. “Adult Consequences of Growth Failure in Early Childhood.” The American Journal of International Monetary Fund. 2018. Kingdom of Clinical Nutrition 98(5): 1170-1178. Lesotho: 2017 Article IV Consultation-Press Release; Staff Report; and Statement by the Executive Director Huppi, Monika, and Martin Ravallion. 1991. “The for the Kingdom of Lesotho. Sectoral Structure of Poverty during an Adjustment Period: Evidence for Indonesia in the Mid-1980s.” World Kakwani, Nanak C. 1977. “Measurement of Tax Development 19(12): 1653–78. Progressivity: An International Comparison.” Economic Journal, 87(345): 71–80. International Organization for Migration 2018. The HIV and TB Integrated Biological and Behavioral Study for Kakwani, Nanak C. 1980. Income Inequality and Poverty: Migrant Min Worker Sending Communities in Lesotho Methods of Estimation and Policy Applications. New (2016-2017). York: Oxford University Press. International Organization for Migration. 2019. Report Kakwani, Nanak C., Adam Wagstaff, and Eddy van on Provision of Life Saving Information to the Drought Doorslaer. 1997. “Socioeconomic Inequalities in Health: Affected Populations in Lesotho. Measurement, Computation and Statistical Inference.” Journal of Econometrics 77(1): 87–104. International Organization for Migration / Economist Intelligence Unit. 2018. Migration Governance Snap Ministry of Health of Lesotho and ICF International. Shot – The Kingdom of Lesotho. 2016. Lesotho Demographic and Health Survey 2014. Maseru, Lesotho. International Organization for Migration / Migrant Workers Association. 2019. Interview with Basotho Motelle, Sephooko I. 2012. “Labour Migration from migrants working in farms in Ceres. Lesotho to South Africa: Changing Patterns and Policy Implications.” Prabandhan: Indian Journal of Kamanou, Gisele, and Jonathan Morduch. 2002. Management, 5(1): 26-33. “Measuring vulnerability to poverty.” WIDER Discussion Papers//World Institute for Development Economics. Narayan, Ambar; Roy Van der Weide, Alexandru Cojocaru, Christoph Lakner, Silvia Redaelli, Daniel Lakner, Christoph; Mahler, Daniel Gerszon; Negre, Gerszon Mahler, Rakesh Gupta N. Ramasubbaiah, Stefan Mario; Prydz, Espen Beer. 2019. “How Much Does Thewissen, 2018. Fair Progress? Economic Mobility Reducing Inequality Matter for Global Poverty?” Policy Across Generations Around the World. Washington, Research Working Paper; 8869. Washington, D.C.: D.C.: World Bank. World Bank Group. O’Donnell, Owen, Eddy van Doorslaer, Adam Wagstaff, Lesotho Bureau of Statistics. Forthcoming. “Lesotho and Magnus Lindelow. 2008. Analyzing health equity Poverty Trends and Profile Report: 2002/2003 – using household survey data: a guide to techniques 2017/2018.” and their implementation. Washington DC, USA: World Bank Institute learning resources series. Lesotho Food and Nutrition Coordinating Office, WFP, IFAD, UNICEF, FAO. 2019. “Fill the Nutrient Gap Oaxaca, Ronald.1973. “Male-Female Wage Differentials Lesotho. Summary Report.” https://www.wfp.org/ in Urban Labor Markets.” International Economic fillthenutrientgap. Review. 14(3): 693–709. 154 Lesotho Poverty Assessment I Progress and challenges in reducing poverty Pritchett, Lant, Asep Suryahadi, and Sudarno Sumarto. World Bank. 2016b. Lesotho - Country Partnership 2000. Quantifying vulnerability to poverty: A proposed Framework for the Period FY16-20. Washington, D.C.: measure, applied to Indonesia. No. 2437. World Bank World Bank Group. Publications. World Bank. 2017. Lesotho: Public Health Sector Ravallion, Martin. 1998. Poverty lines in theory and Expenditure Review. Washington, D.C.: World Bank practice. The World Bank. Group. Roemer, John E. 1998. Theories of Distributive Justice. World Bank. 2018a. Unlocking the Potential of Lesotho’s Cambridge, MA: Harvard University Press. Private Sector: A Focus on Apparel, Horticulture, and ICT. Washington, D.C.: World Bank Group. Sekantsi, Lira P. 2018. “International Remittance Markets in Lesotho: Implications for Economic and World Bank. 2018b. The Human Capital Project. World Social Development.” Journal of Payments Strategy & Bank, Washington, DC. Systems, 12(2), 166-184. World Bank. 2018c. Climate-smart agriculture in Spaull, Nicholas. 2012. “SACMEQ at a Glance for 10 Lesotho. Washington, D.C.: World Bank. African countries.” 2-page research note per country. World Bank. 2018d. The State of Social Safety Nets Taylor, Stephen, and Nicholas Spaull. 2015. “Measuring 2018. Washington, DC: World Bank. access to learning over a period of increased access to schooling: The case of Southern and Eastern Africa World Bank. 2018e. Kingdom of Lesotho - Public since 2000.” International Journal of Educational expenditure review: improving expenditure efficiency Development, 41:47-59. for inclusive development and growth. Washington, D.C.: World Bank. UN-INSTRAW Website, 2006. https://trainingcentre. unwomen.org/instraw-library/2006-I-MIG-LSO-MIG-EN. World Bank. 2018f. The Market for Remittance Services pdf in Southern Africa. FCI Insight. Washington, D.C.: World Bank. United Nations. 2013. Realizing Women’s Rights to Land and Other Productive Resources. Office of the World Bank. 2019a. Global Shared Prosperity Database. High Commissioner for Human Rights and Women. https://datacatalog.worldbank.org/dataset/global- database-shared-prosperity von Grebmer, K., J. Bernstein, L. Hammond, F. Patterson, A. Sonntag, L. Klaus, J. Fahlbusch, O. Towey, C. Foley, World Bank. 2019b. World Development Report 2019: S. Gitter, K. Ekstrom, and H. Fritschel. 2018. 2018 Global The Changing Nature of Work. Washington, DC: World Hunger Index: Forced Migration and Hunger. Bonn and Bank. Dublin: Welthungerhilfe and Concern Worldwide. World Bank. 2019c (forthcoming). Lesotho Digital Wagle, Udaya R., and Satis Devkota. 2018. “The Impact Economy Diagnostic. The World Bank. Washington, DC. of Foreign Remittances on Poverty in Nepal: A Panel Study of Household Survey Data, 1996–2011.” World World Bank. 2019d (forthcoming). Lesotho Disaster Risk Development, 110, 38-50. Financing Diagnostic. Washington, DC: World Bank. Wagstaff, Adam, Eddy van Doorslaer, and Pierella Paci. World Bank. 2019e (forthcoming). Lesotho Climate- 1989. “Equity in the Finance and Delivery of Health Smart Agriculture Investment Plan. Opportunities for Care: Some Tentative Cross-Country Comparisons.” transitioning to more productive, climate-resilient, and Oxford Review of Economic Policy 5(1): 89–112. low carbon agriculture. Washington, DC: World Bank. World Bank. 2016a. World Development Report 2016: World Bank. 2019f (forthcoming). Lesotho Agriculture Digital Dividends. Washington, DC: World Bank. Public Expenditure Review. Washington, DC: World Bank. Lesotho Poverty Assessment I Progress and challenges in reducing poverty 155 World Bank. 2019g. Lesotho - Education Public World Bank. 2020. Doing Business 2019: Comparing Expenditure Review. Washington, D.C.: World Bank. Business Regulation in 190 Economies. Washington, D.C.: World Bank. World Bank. 2019h. Linking Smallholders to Markets: A Supplier Development Program for Vegetable Farmers World Bank, Food and Agriculture Organization and in Lesotho. Washington, D.C.: World Bank. World Food Programme. 2017. Assessment of the Government of Lesotho’s Food Price Subsidy Program. 156 Lesotho Poverty Assessment I Progress and challenges in reducing poverty This report examines progress and challenges in reducing Despite the growing urban-rural poverty divide, inequality fell poverty in Lesotho. Lesotho’s poverty rate is lower today as a result of expansion of social protection and an increase than it was 15 years ago. However, with a poverty rate of in wage incomes among the poor. In spite of this, Lesotho 49.7 percent in 2017, poverty remains widespread. Economic remains one of the 20 percent most unequal countries in the vulnerability is high, with more than 75 percent of the world. population either poor or vulnerable to poverty. This suggests that most of the population lack economic opportunities and A combination of policies that improve human capital, are deprived on multiple fronts. Urban areas experienced promote job creation and address high unemployment, greater poverty reduction due to improvements in education increase agricultural productivity, together with those that and increases in incomes from well-paying jobs, largely in build resilience against economic and environmental shocks, the services sector. In rural areas, poverty stagnated due would boost shared prosperity and accelerate poverty to slow growth in agricultural incomes, a fall in remittances reduction in Lesotho. and vulnerability of the rural population to weather shocks.