"g, ?;p b nb 0-1 iz i a4l "I o Hilt i- 0;'N I 'i IT %11 4.4i T" all pi% "Vi- l: uj8jsD"3 lIPUD1,11, SI IDWO14 w1l 9, -41 c2i, ;l 11' V I17 aid K IIN 1, , I , "  ,J : I , , , - ) I 9 lij Oidd Na 'N 71 OJ L, 7. pA -1. -iM -k In t 'Al k d7 d'.) ''i - U "I 1. I .,5-;woa S;,, 51 no ,-'L60 !%V .14,  , .   I L L, 4 L I ' id Z `,KJid j I t G 1- I L j NO - NN- j ey, iq gj 7 JP jlN ir, V gr 1 14P, NE;uja 1 TI N t f9 Xtritixt Q Q, J-I 4lod I, Nn It 4 , Q "W Q Pura 'Y F, g"' C 0 A d K i LWwf S < 1," V n 4.. Z' IN jv 65i J 7 ol if T-m IN da M 4A 1 -J, t I I - -1, ll F lN R 11P 22 s jt i0i IN ;nVU Id -  -7' L J-` )  -  L . E  q?N id um A19 N UOP I, i-, Nt nre J I , "I ,, ,1 - Wd d wrw 'Y ZLZOZ 51 'I Table of Contents PART A. Confronting the Challenge I. Introduction ...........................................1 II. The Limits of Narrow Growth ..........................................1I 111. Poor Quality of Human Development Investments ..................................5 IV. High Risk and Vulnerability .....................................7 V. Accountability, Information and the Poor ....................................8 PART B. Priorities for Action VI. Outlining A Poverty Reduction Strategy .................................. . .9 Annexes References Map ACKNOWLEDGEMENTS This is a summary of the report "Swaziland: Reducing Poverty Through Shared Growth" (January 2000) prepared by a World Bank team in response to a request by the Govemment of the Kingdom of Swaziland. The team from the World Bank was led by Trina Haque and consisted of Francisco Pichon (on rural development), Lisa Garbus (on HIV/AIDS and overall), John Ngwafon (onl the poverty profile), James Sackey, Lemma Merid and David Bruns (on public expenditure incidence analysis), Elena Glinskaya (on orphans), Albert de Roo (on the issues paper) and Lori Geurts (task team assistance). Collaborative studies undertaken by technical agencies in Swaziland as background for the analysis included: analysis of rural poverty in the Southeastern Lowveld led by Steven Atkins and Samuel Dlamini, Ministry of Agriculture and Cooperatives; analysis of rural institutional structures by Patricia Musi, University of Swaziland; public expenditure incidence analysis with planners from Ministry of Finance, Economic Planning and Development, Health and Education, and UNDP staff; and analysis of the poverty profile with staff from the Central Statistical C)ffice. Mr. Lionel Demery is the Peer Reviewer for the report. The team thanks many colleagues in Swaziland, in particular Mmes./Messrs. Ephraim Hlophe, Principal Secretary (Ministry of Economic Planning and Development), Musa Fakudze, Principal Secretary (Ministry of Finance), Nomathemba Dlamini, Chief Economic Planning Officer and Coordinator of the National Development Strategy (MEPD), Philomene Makolo (UNDP Resiident Representative), Zandile Tshabalala (MEPD), Nomusa Tibane (MEPD), Ncane Dlamini (Microprojects Programme), Noma Maphalala (UNDP), Sara Dlaimini (CANGO), Jabu Dlamini (Deputy Prime Minister's Office), John Ngwisha (UNISWA) and Phumzile Magagula (UNISWA). The task team is also very grateful to the World Bank's Swaziland Country Team for their contributions and comments. The report was completed with the guidance of Sector Managers for the Human Development Group, Eastem and Southern Africa, Ms. Ruth Kagia and Mr. Dzingai Mutumbuka, and the Country Director for Swaziland, Ms. Pamela Cox. Swaziland Reducing Poverty Through Shared Growth Summary Report Prepared as an input into the development of Swaziland's Poverty Reduction Strategy SUHHALAY - 5w+&.iA: R4cW PWft'* , G, 5 q4OW CONFRONTING THE CHALLENGE 1. Introduction The people of Swaziland are its greatest resource. Yet social and economic indicators of household welfare converge to confirm fundamental inequalities in access to incomes and assets, and the existence of significant poverty and deprivation. Furthermore, as the regional economic and social climate is transformed, the fragile gains of the past are being fast eroded. At this historic juncture, the Swazi poor need to come to the fore of the public policy framework. There is an urgent need-and an opportunity-to catalyze a new pattern of pro-poor development in Swaziland where the poor participate and share fully in growth, human development and social protection. Policy framework: Many of the policy actions required to reduce Need prioritized poverty have already been identified and embodied in Swaziland's implementation of NDS National Development Strategy (NDS). A poverty reduction strategy in Swaziland means prioritized implementation of NDS policies. II. The Limits of Narrow Growth The Swazi economy performed strongly during the 1980s, but over High growth in past, the past seven years, average real GDP growth has hovered close to the population growth rate estimated at 2.7 percent by the 1997 currently stagnation Census. In other words, real per capita incomes have remained virtually stagnant in the mid-1990s (Figure 1). Moreover, it is estimated that Southern Africa Customs Union (SACU) receipts- which have supplied about 50 percent of Swaziland's total govern- Figure 1: Growth of Real Per Capita GDP, 1990-1998 6.00'yo, 5.00% ,- 4.00'!/. 4.00%1 3.00% - 2.00% \ 4.00%] -1.00% X< -2.00% -3.oo0/D 1990 1991 1992 1993 1994 1995 1996 1997 1998 Years Source: World Bank Live Database. 1999 1 S1ll4HARY - Xsv~4 J: R POVe4Y* W4oM a 4 ment revenue-will drop by as much as 8 percent of GDP beginning around 2000/01. This expected decline underscores the need for Government to allocate scarce fiscal resources in pro-poor ways. Fig. 2: Shares of Consumption by Different Groups 45 - -Growth has not 40 benefited everyone ~35 o 30 E 25i 20 10 15 Poorest lo Richest PO1. Decile (10%) Grou0s Note: Population ranked by per adult equivalent consumption. Source: SHIES 1995. Incomes are highly concentrated. Only a small share of the population have garnered most of the national income. In fact, analysis of the national income distribution from the 1995 Swaziland Household Income and Expenditure Survey (SHIES) shows that the richest 10 percent of the population control almost 40 percent of the total incomes in Swaziland (Figure 2). In sharp contrast, the poorest 40 percent of the population control only 14 percent of total incomes. Inequality is high in both rural and urban areas. Paralleling the concentration and inequality in the distribution of L.and and cattle incomes, key assets such as land and cattle are also found to be ownership is also very highly unequally distributed in Swaziland, even among smallholder unequal farmers on Swazi Nation Land (CSO Agriculture Surveys). Table 1: Poverty Indic es (people) ___________ Core poor Poor Region Prevalence Prevalence Depth Severity (%) (ew gap : Urban 17.2 29.7 12.4 6.8 Rural 20.8 42.8 15.5 8.0 National 20.0 40.0 15.1 7.8 Source: SHIES 1995. Approximately 43 percent of the rural population are poor, while Rural poverty is much about 30 percent of the urban population are poor (see Annex for worse than urban details on poverty lines). The depth and severity of poverty are also poverty- worse in rural Swaziland. Moreover, rural areas have a greater 2 5L1HtARy - SWA4.': Z44-e* Pov"I* gio4 9 ;4_1w; share of the Swazi poor (84 percent) than they have of the total Swazi population (79 percent). Thus, not only is a rural Swazi more likely to be poor-and in deeper poverty-than a Swazi living in town, but the majority of poor people reside in rural Swaziland. Within rural areas, the worst poverty levels (all indices) are found in -particularly in rural rural Shiselweni region. In rural Shiselweni, more than half the Shiselweni population live in poverty. Despite the strong rural dimension, the vulnerability of the urban -but also in the large poor cannot be overlooked. For example, 40.6 percent of the gazetted towns population of the large gazetted towns live in poverty, which is almost as high as the share of rural Swazis in poverty (42.8 percent). Given the inequality in ownership of physical assets observed above, Labor is the poor's the major asset available to poor households is the quantity and major asset quality of their labor power. The greater the imbalance of dependents to labor-force participants, the greater the chance of the household being in poverty. Yet adequate quantity of labor-power alone is not sufficient to prevent the household from being poor. There needs to be sufficient labor market demand to absorb the continuous new entry into the labor-force and households must be able to invest in the skills for which there is market demand. Unfortunately, in the mid- and late-1990s unemployment in Swaziland has been in excess of 20 percent and rising as job- creation fails to keep pace with high population and labor force growth. For the age group 15-24 years, the problem is acutely worse, with national unemployment levels above 40 percent. And it is estimated that over 40 percent of the core poor are unemployed As unemployment in South Africa rises, migrant workers from Migrant labor neighboring countries such as Swaziland find it increasingly difficult opportunities have to find employment and are returning to their home countries. In declined 1990 there were over 16,500 Swazi migrant workers employed in South African mines compared to about 13,000 in 1997. Labor income from Swazi mine workers fell from being equivalent to 13 percent of GDP in 1990 to only about 6 percent of GDP by 1995- 1997. Furthermore, the slowdown in foreign direct investment (has also led to stagnation in employment in the urban economy and the title deed land (TDL) sector. The duration of urban unemployment episodes appears to have increased, as has strain on the urban in- formal sector to absorb new entrants and lay-offs from the formal sector. Levels of growth need to be high and the pattern equitable to reduce poverty. Estimates show that a unit reduction in inequality would To reduce poverty have as great an impact in reducing poverty as a unit of inequitable faster, need more growth. equitable growth Additionally, an estimate was made of the minimum rate of national income growth that is needed to reduce the total number of the poor, 3 5UHHA-AY - 5WA&4.dk R PAe4". iA04 a Q4,OW if inequality remains unchanged. Under an assumption of 2.7 Under present percent per annum population growth rate, and inequality remaining inequality, need unchanged, Swaziland requires a minimum growth rate of 5 equm grwt in percent per annum in real GDP, or a growth rate of 2.3 percent minimum growth in per annum in real per capita GDP, in order to prevent the real per capita GDP of numbers of the poor from rising. This is clearly cause for 2.3% per year- concern. The actual real per capita GDP growth rate in Swaziland has stagnated around 0 to 1 percent in recent years. Furthermore, GDP growth projections for Swaziland in the short term are not very optimistic and are placed currently around 2 percent per annum by the World Bank. This is far below the minimum target growth rate required for reducing the numbers of the poor. These estimates tend to confirm, first, that the number of poor people in Swaziland have increased at least since the mid-1990s; -which is not viable second, to have a realistic chance of reducing poverty, high growth without fundamental rates are necessary at a time when Swaziland faces a crisis of reform economic prospects; and third, ensuring that the new sources of growth are sufficiently broad to be accessible to the poor-and indeed to focus on the poor-is essential. As historical sources of growth dry up due to permanent shifts in the regional context, poverty will continue to deteriorate unless New growth must be Swaziland takes measures to pursue a growth strategy that draws labor-intensive on its potential long-term comparative advantage. An emphasis on high labor-intensity is crucial if the poor are to share in the benefits of growth. One key option is commercialization of the hitherto neglected Swazi Nation Land (SNL). Labor-intensive SNL agriculture has significant growth and poverty- reduction potential. The small plots under present yields and methods of cultivation cannot provide much increased income to Potential engine of supplement declining wage earnings. In order to increase eamings growth-commer- from farming, households may find it advantageous to switch to cialization of higher valued crops, such as market vegetables or perennial smalholder farmin products, instead of cultivating predominantly maize, as at present. o Those who stay in maize can also achieve higher yields than at present. Research elsewhere in Africa has produced evidence that the growth potential from smallholder commercialization and exploitation of comparative advantage can also generate considerable multiplier effects on local incomes. As more small farmers commercialize they And its non-farm create demand for small farming implements, hired labor (backward Ilinkages linkages) and trading services, small transport, local consumer goods, and so on (forward linkages) which stimulate local labor demand. Both the initial and subsequent rounds of growth through demand linkages are labor-intensive and can be accessible to poorer households. 4 5UHHAtRY - W< Rx Povvay 5 4 Q4L SNL households have demonstrated dynamism in their modification Yet majorconstraints of survival strategies under numerous constraints. Yet, the persist diversification into high-value cash crops by SNL farmers continues to be very limited due to a number of factors that impede investment and growth in productivity: 6) Land rights are insecure. Under traditional land tenure arrangements, chiefs traditionally have the power to withdraw land rights. Insecurity of property rights over land can be important deterrents to productivity-enhancing investments. Swazi farmers may be inhibited from investing in their land for fear of losing it. 6) Under an open-access grazing system, cattle are allowed to graze in excess on SNL common pasture at no direct cost to the cattle-owning household, while Government dipping services and most other veterinary services are provided free. Ministry of Agriculture and Cooperatives (MOAC) predicts that these grazing systems will sooner-rather than later-lead to levels of soil erosion and land degradation that will undermine the sustainability of farming systems. O Access to capital is expected to be a severe constraintfor most SNL farms in moving into irrigated agriculture. Various features of the formal financing system tend to exclude poorer smallholders from accessing financing for irrigation. Most importantly perhaps, the disincentive to invest in irrigation is high without secure tenure to land. Ill. Poor Quality of Human Development Investments The ability of the poor to participate in labor markets-and the returns they earn from such participation-is closely linked to the The poor need the quantity and quality of their human capital. The success of human right skills and good development policy in reducing poverty depends largely on the health extent that public expenditures enable the poor to obtain and sustain good quality education and health services. Despite a substantial allocation of the public budget to these sectors, Swaziland's education and health indicators are worse than But public spending expected, and the disparities are greater for the poor, in rural areas, within education and and for women. The analysis indicates that the allocation of public health are not pro-poor expenditures within each sector is biased against the basic service levels utilized by the poor. The consequent under-funding of primary education and health services combines with an inefficient allocation between wages and other essential items to generate a poor quality of services available to the poor. Major re-orientation is needed. 5 SUHHARY - 5wS4~: k POVci9* d 4 GWWO Swaziland has one of the highest HIV/AIDS prevalence rates in the HIV/AIDS prevalence world. Over 18 percent of the total adult population are HIV positive, is extremely high AIDS is undermining many gains of the past. As a consequence of the epidemic, it is estimated that the average life expectancy in Swaziland has fallen from 57 years in 1996 to about 39 years by 2000, rolling back much of the gains in the last thirty years. The pressure to protect-and increase-social sector spending in the most equitable and efficient manner is thus critical in Swaziland. Public expenditures within the health sector are inefficient and Too little is allocated biased against the poor. There is high exposure for the poor to to preventive/ primary water-borne disease given that about 62-63 percent of the poor and health extremely poor people in rural areas must rely on rivers as their principal source of drinking water. Despite much of the country's disease burden being preventable, public health expenditures are skewed toward curative care. Only 20 to 30 percent is allocated to primary/preventive care, likely to be used mostly by the poor, whereas curative medical services tend to cater to the needs of the well-to-do in Swaziland. Strong urban bias is also evident in the distribution of health services and facilities. This results in rural clinics-to which the poor have greater access-being underserved. Country-wide, over 50 percent of the people who live in households ILack of education is headed by people with no education, live in poverty. In contrast, the -strongly related to poverty ratio is halved for people in households whose heads have obtained some secondary education (Figure 3). This suggests that poverty status higher incomes are related to more education and that ensuring affordability of primary and secondary education for the poor is thus an investment with high returns for society and the household. Figure 3: Poverty and Educational Level of Household Head 60- 7 -'50 40 30 20 0 No LowNer Upper Secondary High Higher education Primary Primary School Education Educational level of household head Source: SHIES 1995. 6 5ULIHAIY - A44 : Pt'c4q M4o QW Despite aggregate improvements in education indicators in The poor are less able Swaziland over the past two decades, education levels attained by to afford education the Swazi poor are much worse than that by the non-poor. For the poor in Swaziland, sustaining school attendance and obtaining quality learning may be a much larger issue than that of gaining initial access. It takes an average of 13 years to produce a primary school graduate in what should take 7 years, and repetition and Inefficiency is costly drop-out rates are thought to be higher for the poor than the non- for the poor and Govt poor. These high repetition rates are due to poor quality, inefficiency and high cost. By addressing these issues, Government could obtain a savings of 84 percent on primary education and also substantially reduce costs for poor households. Public education subsidies are biased to the tertiary level. The public cost (subsidies) to educate 1 university student in 1995/96 in Public spending for 1 Swaziland was sufficient to cover the cost of educating 47 primary or university student 13 secondary school students. Yet the poor often cannot keep their = 47 primary or children in school through the primary and secondary levels to reach 13 secondary the tertiary tier. Improved equity and efficiency in public education sendr spending will lead to better outcomes-at a lower cost-for both students poor households and the Government. IV. High Risk and Vulnerability To protect against community-wide and individual shocks, Swazi households and communities have developed informal insurance mechanisms on the basis of social norms and mutual exchange. Informal social These informal safety net mechanisms-whilst often being the only protection is not mechanisms accessible to the poor-do not necessarily provide enough effective protection. SNL farms, for example, are highly vulnerable to droughts, a time when the capacity of all households tends to be overwhelmed. Moreover, the poorest households can be marginalized from community institutions, such as women's groups (luholiswano). And the cumulative impact of the HIV/AIDS crisis is undoubtedly undermining the extended family support network. Drought is now a frequent phenomenon in Swaziland, and indeed, the whole of Southern Africa. During the historic drought of 1991/92, Recurrent drought is a SNL maize production fell dramatically from 145 thousand metric major source of risk tons in 1990/91 to 54 thousand metric tons in 1991/92. The poor, who have few assets, are hit even worse than aggregate figures may convey. And as droughts become more frequent, the assets of households have less time to recover between each episode. Box 1: Drought is Now Endemic Drought is now considered endemic in the lowveld and communities are coming to terms with the need to change agricultural practices to suit the average rainfall pattern. Source: Drought Consortium, 1996 7 SLIHHAI4 Y - 5wA.'k R PAewj 0Aom S44a dGow; Swaziland has one of the world's most severe HIV/AIDS epidemics. As is HIV/AIDS About 60 percent of HIV infections in Swaziland occur among those 20 to 39 years old, one of the most productive segments of society. Both the poor and non-poor are vulnerable to HIV but the poor are least able to cope with the illness and associated costs. Moreover, ,Swaziland is losing its AIDS can often result in the increased impoverishment of productive workers households through cost of treatment, lost adult income and assets, funeral expenses, etc. Multigenerational households that lack the middle (income-producing) generation are increasing. The gender disparity in legal rights exacerbates women's Women's minorstatus vulnerability. In both the indigenous customary law and the Roman- ire-inforces Dutch common law systems governing Swaziland, women are .. considered minors. This means that women-in law, and often in vulnerability practice-do not control or own major assets, particularly land. This fundamental imbalance in the property rights of men and women leads to gender differences in the pattern of poverty. De facto female headed households (i.e. where a male co-head is mostly nonresident) have the highest core poverty levels of any household type. V. Accountability, Information and the Poor A more coordinated approach to local development efforts-based on accountability of local government to the community and better monitoring of poverty information-is essential for improving economic management and poverty reduction efforts. The Tinkhundla system does not have a clear mandate, budgeting or Poor coordination of reporting structures. Line ministries do not necessarily maintain a development at common structure of personnel at the decentralized level. Even tinkhundla level when there are personnel at the level of an inkhundla, there is no structure that compels these departmental representatives to coordinate their activities at the level of the inkhundla. Reporting structures of the agencies continue to be vertical within the parent agency, with little emphasis on horizontal coordination at the local level. In many chiefdoms there are no women in the decision- making inner council and only one or two women in the development committees. Many people within and outside Government have also identified the Poverty information is lack of poverty data as a major constraint to developing and very limited sustaining a more pro-poor policy framework. There is also no clear institutional framework for coordinating data generated by different institutions and channeling it to policy-makers and community organizations. Both are necessary to build social consensus on the need for enacting pro-poor reform. 8 SLIHHARY - w Pwe*? o . Gw PRIORITIES FOR ACTION VI. Outlining A Poverty Reduction Strategy A new, equitable and dynamic pattern of growth and human Need to identify development is required for a serious poverty reduction agenda in critical actions Swaziland. Unless public action is undertaken to remove major structural constraints to labor-intensive development, however, livelihoods are expected to continue stagnation and poverty to worsen further. The analysis of poverty in this report can be used to identify priority areas for action. This agenda for reform will require courage, vision and sensitivity on the part of the country's leaders. Basic themes of a poverty reduction strategy for Swaziland are proposed below. Enabling the Growth of Smallholder Agriculture Smallholder SNL agricultural commercialization will need to be a central element of the poverty reduction and growth strategy of Bottom-up growth Swaziland. This emphasis is consistent with Swaziland's through dynamic SNL comparative advantage in labor-intensive sectors and location near a major port. There is scope for expansion, diversification and intensification to access export markets through commercialization of SNL agriculture into high value crops such as vegetables, perennials, etc. and through the related multiplier effects and demand linkages. The capacity of rural areas to augment incomes is expected to increase, reducing pressures on urban labor markets and services. Focusing public efforts on removing impediments to agricultural growth on SNL is thus a priority for reducing poverty in all of Swaziland. This bottom-up development approach is critically contingent, however, on catalyzing and protecting investments for small farmers on Swazi Nation Land. Yet SNL growth-however labor-intensive-is not a panacea to the unemployment and poverty problem. There is need to build the labor Not panacea but part absorption capacity of the economy as a whole by promoting choice of overall labor- of labor-intensive techniques in production (e.g. by increasing the intensive strategy cost of capital by removing implicit subsidies) and the growth of other labor intensive sectors for which there may be export demand. Promising areas could be identified in collaboration with regional investment partners. Rural land tenure reform is crucial and models already exist Land Need rural, SNL land holdings of households on SNL are small (on average 1.7 hectares), and two-thirds of households have less than one hectare. Earnings tenure reform from these small plots are low, because yields are low and the land 9 SUHHARY - _4AA: Ra PovftiF gim4 V4i*a( 4404 has little yield-enhancing investment, such as fertility improvements or irrigation. Increased returns from SNL farming thus requires productivity- enhancing investments on the land. Investment, in turn, is closely -to assure more linked to land tenure. Land tenure on Swazi National Land at secure property rights present is not conducive to investment. The skewed distribution of for small farmers land-holdings combined with the present system of land tenure provides little prospect for improved earnings from agriculture for households on SNL. Tenure is not secure, and land is subject to periodic redistribution. Women manage much of production, but do not have recognized tenure, and hence have limited decision authority. Furthermore, land is not traded through market transactions and hence values are not transparently visible. In contrast, experience from many countries-including low-income countries-has demonstrated that rural households have surprisingly Security and high savings rates. Investment in household farming is a high tradability of land priority for most rural families, but they will invest only if they are confident that their tenure is secure, and that they can sell or trade rights encourage the land for a fee if they so choose. Security and tradability of tenure investment allow the family to benefit from the investment as long as they hold the land, and to recover the investment in case the household leaves farming. Several changes in tenure of Swazi national land could stimulate investment and improve productivity of land. These issues are being discussed as part of the Land Policy debate. They bear emphasis as important pre-requisites for poverty-reducing growth: v Enhanced security of tenure: For example, ownership or 99-year leaseholds, such as those piloted in urban areas (Box 2), would stimulate investment. Box 2: Innovations in Land Tenure on SNL Innovations in land In November 1995, as part of the Urban Development Project tenure reform already (financed via the World Bank), the King agreed to transfer Swazi exist in Swaziland Nation Land to formally designated Local Authority Areas, in the form of a 99-yearleasehold. Under the project, this land is being serviced and leased primarily to low-income households. Beneficiaries thus acquire a valuable asset that may be leveraged to acquire further assets, such as shelter improvements and equipment for home-based enterprises. This is an innovative way to promote transition from traditional, collective ownership of land to a more market-based asset ownership. A monitoring and evaluation system is being developed and Government intends to use these evaluations in extending the new land tenure system in other urban areas. 10 SUHHARY - SwA4, I l Pt 9" 94 Q40W Greater gender equality in land rights: Women care for the land and investment in it, and can contribute most if they share in tenure rights. V Tradability of tenure rights: Active markets for owned land or for tradable leaseholds allow the value of land to increase as investments are made. Cattle-grazing reforms are necessary for sustainability of Swaziland's rural resource base. Not only is the public subsidy to Need more cattle-raising regressive-since the poor own very little cattle-but sustainable cattle the dynamic cost to society of this subsidy is high. It propagates grazing management uneconomic management of both cattle and land resources. To halt the deterioration of common resources (potential grazing and arable land), efforts are needed to move the management of cattle in more environmentally sustainable directions. This will involve increasing incentives for cattle off-take and sustainable grazing. Reductions in the public subsidy to the private cost of keeping cattle are critical in this regard. This could be achieved through dipping charges, veterinary fees and maintaining border parity prices. Institutional mechanisms could be piloted which provide community control and enforcement of fees associated with grazing cattle. Again, lessons from the pilot could later be scaled up nationally. There is need to overcome capital constraints to rural growth through reform of the rural financing environment. Innovations in rural Need regulatory finance could be introduced at the same time as land tenure reform, reform and to stimulate savings and provide better access to financial services. innovations in rural The rural poor have access to financial intermediation only at high finance cost, constraining capital investments such as small irrigation or working capital. Most savings are still undertaken in lumpy, non- financial forms such as cattle. In order to move towards a rural financial system with cheaper and more accessible savings and credit opportunities open to the poor, one key action is to address the underlying causes of the failure of the Swazi Development and Savings Bank. The Government should avoid mixing commercial Box 3: Rural financing via private partnerships The MOAC (1999) study found several SNL households which obtained credit as inputs from the local cotton ginnery. Very few of these households experienced problems in repaying their loans at the end of the season, reportedly as a consequence of substantial extension support from the ginnery company. In another instance, a loan was obtained by a household to develop sugarcane in an area next to the Usuthu River. This loan was under-written by the local sugar mill with repayment deductions taken directly from the sugar receipts by the miller. The sugar company also provided substantial technical assistance. These models of financing partnerships need to be evaluated for possible promotion. 11 S1IPHA.Y - 5WOIgK: Rd Poway 9 9 L"wOg with developmental activities in a restructured bank. It could instead explore revival of post-office savings functions, NGO group-based savings programs, and other vehicles for savings mobilization and small-scale lending in rural communities (Box 3). Changes in tenure, backed by appropriate changes in law, could be Could pilot and then introduced on a pilot basis in rural localities where residents and scale up nationally local authorities seek to increase opportunities for land-holders to invest in their land. Indeed, pilot projects incorporating the full set of recommendations (e.g. land tenure reform, new technology packages, small-scale irrigation, improved rural finance) and international participation from institutions such as the IFC in processing for export could be designed to show in more detail what can be accomplished. Lessons from the pilot could be used to scale up the program nationally. Ensuring Effective Human Development Investments The better health and strong basic education of its people are among the most productive assets in which a country can invest. To facilitate higher productivity of labor, lower fertility and greater mobility of labor in these changing labor markets, the quality, relevance and affordability of basic education and health needs to be improved. A structural shift in the public spending pattern is needed to focus on primary and secondary education levels rather than academic tertiary levels, and on preventive, core health services rather than curative services. There is a need to prioritize Swazi education spending towards quality and efficiency at primary and secondary levels. The current Need focus on quality system of almost total subsidization of tertiary education while and efficiency in primary education is still unaffordable to some households, cannot primary/secondary be supported within a poverty reduction framework. Greater education efficiency and equity of education spending in Swaziland can be simultaneously achieved through: a) cutbacks in the grants scheme to tertiary level students, and greater cost-recovery at the tertiary level; and b) re-allocation of these funds to enable funding of early childhood programs (ECD) for poor communities, reduced fees at the primary and secondary levels, reductions in class-size, improved teacher training and curriculum content, increased materials and supplies and a small, need-based grant scheme at the tertiary level. Institutions such as the World Bank and the African Development Can draw from Bank could be mobilized to draw on their international experience in designing reform of education financing and quality enhancement nternational programs for primary and secondary education investments. experience 12 SUHHhAY - 5w - &Rw Povftl. o 9 G44WU Evidence from many countries of the world shows that children's ability to learn is greatly influenced by his or her experiences in the first years of life. Early Childhood Development (ECD) programs which combine training of parents, child care in groups, training of community caregivers, integrated nutrition interventions, etc. can Need to expand early build strong foundations of cognitive and nutritional development for childhood the child. The effect of ECD programs is especially beneficial for development programs children from poor families that lack private resources-and often the . . knowledge-to provide a healthy and stimulating environment for forpoorcommunites children. The programs often generate savings that more than compensate the cost of the investment (Box 4). Box 4: Brazil's ECD program (PROAPE) pays for itself Cost of producing a first grade graduate is at least $41 less for the PROAPE children than for children without pre-schooling. This per child saving is higher than the PROAPE cost of $28. PROAPE not only pays for itself, but results in primary school cost savings in the first year over and above the costs of PROAPE. Such early, timely investment can enhance the effectiveness of investments in primary and secondary education in Swaziland by improving quality of learning and lowering the high repetition and drop-out rates currently experienced. Communities, NGOs and national government will need to build a consensus on the most cost-effective approaches to ECD. A final issue to note in the Swazi public education system is with regard to the relevance of the education offered. Communities Need to improve skills themselves want education to be more relevant to realistic work matching to labor opportunities and self employment, rather than educating children for market needs- civil service or white-collar jobs that do not exist (Box 5). Box 5: Relevance of course content "...Our schools are educating people for office jobs that don't exist... .We sell the last cow to educate our children, only to find that they can't get jobs when they leave school." Source: Swaziland PPA 1997. Some possibilities include planning for shifting labor market needs in -but solving the regional Southern African context (e.g. demand for miners may be declining but that for nurses may be increasing). In the end, unemployment is really however, experience from around the world indicates that higher and related to growth more equitable growth is the main instrument necessary for addressing a youth-or other large-scale-unemployment problem. 13 SUHHARY - 5wr4S: ZAe Povw4 dv V44t 4";a In the area of health, prevention not only of further HIV infection but Need to emphasize a also other diseases such as malaria, diarrhea, respiratory diseases, rimar and etc. can be a highly productive investment for Swaziland. These Primatv and diseases take a heavy toll in terms of lost lives, labor days and preventive health school attendance. Prevention is generally cheaper than treatment service after an infection has occurred, typically making preventive care more cost-effective than curative care. In Swaziland, a few key actions will generate a high pay-off, especially for the poor: v Significantly reallocate public health expenditure to primary/preventive care: An essential package of services may be considered for universal coverage via public financing. In order to be cost-effective, however, the basic package needs to be defined carefully to emphasize primary, preventive and ambulatory services rather than curative and in-patient practices. Furthermore, financing for the basic package should be protected during economic crises. V Improve access to health facilities for the poor: This has a physical as well as financial aspect. The 1997 PPA underscored that improvements in roads, bridges, and bus services were essential to improve access for the Swazi poor to clinics and hospitals. Also essential is a structure of public subsidies and user fees where exemptions are targeted directly to the poor or to services or facilities used primarily by the poor. V Improvements in rural water supply (and sanitation) are crucial not just to cutting back the burden of disease but also to reducing constraints on women's time. Insuring the Poor against Major Risks Swaziland's poor continue to be highly vulnerable to major shocks and to be locked in poverty traps because of excessive uninsured Hligh vulnerability of risk. Complementing households' own informal risk management poor requires a social strategies, a public social protection policy is needed to respond to t t the most severe sources of risk, reach the socially and economically protection strategy excluded and do so in the most cost-effective ways. Public action needs to be selective, preventive and targeted where possible to reduce the economic vulnerability of the poor to the major risks. One of the key elements of a social protection strategy is prevention Need cross-sectoral, and mitigation of HIV/AIDS. His Majesty the King of Swaziland has labeled the HIV/AIDS epidemic "a national disaster" (in Speech from multi-level the Throne, Official Opening of the First Session of the Ih response to AIDS Parliament, February 19, 1999). Nonetheless, in Swaziland-as in almost all African countries-the response has been purely sectoral. Swaziland's National AIDS/STDs Programme is housed in the Ministry of Health. Services for counseling, testing, condom promotion, and treatment of other sexually transmitted infections are 14 5L1HHARY - w R Poveq* Vd4, 9 q4ow inadequate. Key actions to develop a coordinated and multi- sectoral approach to HIV/AIDS should include: V Home-based care: linking prevention and care efforts currently operating in isolation, by further developing the draft of, and subsequently implementing, the National Community/Home- based Care and Prevention (CHBCP) program; v Care for orphans and children affected by AIDS: The most long lasting consequences of AIDS are felt by these children. There is both need and scope for a concerted Swazi response (Box 6); Box 6: Protecting Children in Need The scale of orphans The estimated number of AIDS Orphans in the year 2000 is in Swaziland demands 112,000 or 22% of children below 15 years. Extended family a national response networks, institutional solutions and charitable responses will not be sufficient, and a coherent vision for long term development of family and community based responses is needed. Two things are essential to promote more realistic views of future needs: 1. Planning and Costing Models. > Estimates for planning: Reliable orphan estimates, which clarify the size of the population needing care and support; > Costing of alternative models or scenarios for care. > Understanding the cost of not caring: scenarios that estimate the future social costs of not caring in terms of delinquency, correction or remand facilities, social unrest and the like might help build social commitment for mobilizing and strengthening local resources. 2. Strategy Development Building on Swaziland's considerable strengths in its families and communities, government, and private voluntary (NGO and church) and commercial sectors, development of a shared strategy to foster the long term welfare of orphans and other children affected by the epidemic. Source: UNICEF, February 1999. Behavior change: developing a strategy for an effective behavior change program by working with high-risk groups such as transport workers, migrant laborers, and collaborating with specialists, e.g., educational curriculum designers; and V Better monitoring and knowledge-building: forming a knowledge base on the epidemic, and especially the feedback between HIV/AIDS, poverty, access to land, job creation, etc., would aid in mobilizing opinion leaders and better targeting interventions. 15 SUM4HARY - w R Poway V4 S4L. Ce.I" Another major social protection priority must be preparedness for Need continuous drought. A permanent drought preparedness program is now called drought preparedness for in Swaziland. This will require agricultural extension policy to provide more appropriate extension messages in drought-prone areas; an early warning system capable of assessing and disseminating climatic information in a timely manner; Government partnership with the NGO Drought Consortium to promote diversified incomes through non-farm activities and environmental rehabilitation. The analysis clearly indicates that women must be guaranteed equal rights under law to break one of the elements of the vicious circle of poverty and vulnerability. The insecurity of property and other economic rights of women in Swaziland not only render them highly Need legal reform to vulnerable, but also constrain their ability to make investment Nee egal rights to decisions for the household. Legal reforms are necessary which provide equal rights to take account of these linkages and give women full legal rights as women adults. Although other neighboring countries with a similar legal legacy of customary and civil law have enacted legislation to give women the status of adults (for example, South Africa and Zimbabwe), this has not occurred in Swaziland. Box 7: Towards Gender-Equitable Development In several countries in Eastern and Southem Africa, a dialogue is under way among government, civil society, and academia on the laws that impact negatively on women. The main messages emanating from this series of consultations and follow-up activities-including creation of specific action plans-provide a useful framework for addressing these issues in Swaziland: I Empowering women though the law is an economic need. $ Countries must focus on practical, everyday manifestations of needed regulatory changes, addressing land tenure, inheritance, commercial rights, schooling, labor markets, and other topics of immediacy and impact. Countries themselves must define solutions. . The political will of policymakers at the highest levels is required to ensure that laws are revised or enacted, that there is a deliberate effort to build a social consensus for these laws, and that their implementation is carried out with vigor and impartiality. ,/Civil society plays a critical role in influencing policy. Source: World Bank Gender Web Site, 1999. Priority legal reforms involve improving women's right to property and eliminating their minority status. The Gender Perspective Group working as part of the Urban Policy development process has already identified some areas in need of immediate redress: $ amendment of the Deeds Registry Act to allow women to register 16 SLIHMAR' - SwA4d4: R P.4Vet V-4o 4444 Q4" title to land in their own names; / abolishing the requirement under Roman Dutch Common law for a woman to be assisted by her husband to enter into a valid contract; and strengthening of the Age of Majority Act to prevent removal of a married woman's capacity to enter into a contract. Given the strength of the current restrictions on women, legal reform and support to NGOs to provide women from poorer households legal assistance or arbitration services may both be regarded as a part of the core social protection policy of Swaziland. Finally, as an integral step in achieving an effective and affordable social protection program in Swaziland, it is important to improve the efficiency and equity of public safety nets. The fiscal constraints on the Swazi state will not allow it to provide a safety net for all Need public safety households for all types of shocks. A public safety net program should, thus, be invoked only for shocks where the expected impact nets which are well is very severe for a sizable group of the population. To enable targeted to the poor limited resources to go farthest in reaching the poorest of the poor, some form of pro-poor targeting mechanism must be utilized. The experience that Swaziland has of moving from an untargeted approach to a work-based public safety net during the droughts of the 1990s is a useful step in this regard. Nonetheless, certain issues remain to be considered in future rounds of public works as safety nets: I Below-market wage rate: International experience has shown that in a public works safety net, the key to achieving self- Below-market wages targeting by the poor is the wage rate used to compensate are an important tool workers. If this is set at or above the prevailing wage rate for to target the poor agricultural labor in the local area, evaluations show that non-poor individuals will compete with the poor to take up employment under the safety nets public works program, in effect rationing the poor out of the safety net. Future Swazi public works-based safety nets will be better targeted to the poor if this issue is addressed as part of the program design. v Elasticity of safety nets: Especially for public works programs, it is often difficult to start from scratch and design/implement a Preparing a "shelf" of large program once a crisis has already hit. It is thus suggested projects allows that during "normal' years (e.g. non-drought in most of speedy response Swaziland), Government explore the possibility of operating a core, small program sharply targeted to communities which are in the deepest chronic poverty. The institutional experience from this permanent program could be built on to facilitate scaling up during national emergencies. 17 SUHMARY - 5 Re4 pe d V ~ q g Emphasis on work-based programs, but not exclusively so: IBuild an integrated While maintaining a primary emphasis on public works as safety program of nets, some form of free food or cash transfer to individuals who prventio cannot or should not work such as malnourished children, the iintelventions disabled or elderly poor, or labor-constrained female-headed households, is indicated. As much as possible, these programs should use some form of targeting other than work-norms, e.g. community-based targeting, or nutritional screening, etc. to reach the critically vulnerable in Swaziland. Strengthening Institutions to Increase the Poverty Impact of Policies To enable the poor to improve livelihoods and reduce vulnerability, local governance structures must be more responsive to the needs of the poor. Further, policymakers will need ongoing and reliable information to assess and improve the poverty reducing impact of policies. Coordinated and prioritized action-i.e., poverty planning-is e required at the central level to achieve sustainable reduction in poverty. Currently in Swaziland, there are several policy initiatives, poverty planning at such as ESRA, the NDS, the PSMP, the Land Policy, and the central level National Environmental Action Plan which have elements of core relevance to the poverty reduction strategy. A focal point in Government needs to be identified to a) carry forward the formulation of a poverty reduction strategy, b) to liaise with the responsible agencies to monitor implementation of the policy elements deemed to be priorities for the poverty reduction framework and c) to use and disseminate poverty information. The institutional Need a poverty focal home for this responsibility could be the MEPD or other central ne planning unit. Additionally, to improve effectiveness, planning staff point could be trained in pro-poor public planning and expenditure issues. The need for coordination and accountability at the local level is Need locallevel clear. Rural Swazi communities are dynamic and have shown that they can rally around a project that is a priority problem. Yet coodnation coordination among sectors is badly lacking at local levels and traditional or modern governance structures are not consistently accountable to the poor. It is difficult to assess the level of net support each community is receiving at any time, or indeed, has received over time. It is also difficult to assess the poverty reduction impact of different public or NGO initiatives. And communities themselves face confusing messages regarding resource mobilization. Important steps can be taken to build the capacity of existing institutional structures: / The Tinkhundla system could be given responsibility for cross- sectoral coordination at the local level. For this to happen, the The roles of Chiefs inkhundla committee would have to be expanded to include and Tinkhundla is development agencies (Government, NGOs) working in that area. important- 18 SULHHAIY - 5wVIK RW PAvt4*, 4 A44d4 LW; / The accountability and community responsiveness of chiefdoms -but must be would be improved if all chiefs were required (or encouraged) to accountable to have development committees. The development committees would send representatives to the inkhundla. communities Women must be integral to the Tinkhundla system of decision- making. Development committees could be asked to have a basic share of women members. Poverty monitoring and analysis needs broad capacity-building. The Need to build and use Government has initiated discussions to build a stronger poverty apovertymonitoring information and analysis system. and analysis system V Different instruments have different comparative advantages in addressing selected data needs. A well-designed poverty monitoring and analysis system will require capacity building of several different institutions. For example, the capacity of CSO, public expenditure units in sector and central ministries, UNISWA researchers, MOAC early warning units, etc. will need to be strengthened to improve the collection, integration, analysis and use of the different data sources in which they have a comparative advantage. V An institutional framework is also required for integrating these different data sources and channeling the information for use by Poverty monitoring policy makers (central and at inkhundla level) and others. The needs an integrated focal point of poverty policy discussed above could also be the system focal point for the poverty monitoring and analysis system. C Critical to the success of the overall poverty reduction program is the setting of goals and targets which can be measured and monitored. Swaziland will need to identify two or three key indicators for each element of its poverty reduction strategy to assess how well it is doing and where efforts need to be and indicators, goals strengthened further. Moreover, how m,uch reduction in poverty levels does it wish to achieve by the year 2005, 2010 and so on? to monitor success Looking Forward Poverty in Swaziland can indeed be reduced. But to do so will require fundamental change across sectors and leadership to carry forward the change. 19 SUHHAIV - w W PMwc4P 9AM4 A44 4"W ANNEX Building on What is Already Known The poverty analysis incorporates insights, ideas, and information from a large number of people engaged in different aspects of development activity in Swaziland. Utilizing the Participatory Poverty Assessment and other tools, it also reflects the voices of the poor. Preliminary analyses have been discussed in various fora, including workshops organized jointly by MEPD, UNDP and the World Bank. The report draws on existing studies and documents on socioeconomic dimensions of Swaziland. A preliminary literature review of journal articles, books, Government and NGO reports was carried out and major surveys catalogued to identify the scope of the existing analysis of the welfare of Swazi households. This exercise also assisted in identifying key gaps in knowledge on poverty and its determinants in Swaziland. Studies Commissioned as Background for rhe Report To address some key information gaps, the poverty report commissioned a few strategic studies. Given the time and resource constraints, extensive primary data collection was not feasible. All commissioned studies were carried out through collaborative partnerships between technical agencies within and outside the Government in Swaziland and the team from the World Bank . The studies include: - analysis by the Swaziland Central Statistical Office and World Bank staff of the 1995 Swaziland Household Income and Expenditure Survey (SHIES) to build a profile of poverty . I a small supplementary household survey in the semiarid Southeastern Lowveld region of Lubombo undertaken and analyzed by the Ministry of Agriculture and Cooperatives (MOAC) during May - November 1998 (MOAC 1999), to draw out dimensions of rural poverty. • a detailed public expenditure incidence analysis of health and education spending conducted by World Bank staff in cooperation with the Ministries of Economic Planning and Development, Finance, Health and Education, and the UNDP. V a participatory analysis of institutional structures in rural Swaziland conducted by members of the University of Swaziland in early 1999 (UNISWA 1999). 20 SUHHAIV - 5wbi4v: R PAovw4 94o4 94-a GLawUL Poverty Lines Analysis suggests that the SHIES data do not provide reliable estimates of the absolute numbers of the poor in Swaziland. The data are, however, robust to an analysis of the relative levels of incomes and other socio-economic characteristics of households. Accordingly, this report selects two relative poverty lines to focus on the Swazi poor and applies them to the SHIES: V Upper poverty line (UPL): People in the lowest 40 percent of the per adult equivalent national consumption distribution are regarded as poor. X Core poverty line (CPL): The core poor are people ranked in the lowest 20 percent of the distribution. 21 SLtiHARY - S u Rc POVWe -4M4 9-Ma L40a Statistical Annex Annex Figure 1 Shares of Income by Consumption Decile: Rural and Urban 45 40 -4 N 35 E Rural 30 ' UrbanJ 25 a 20 a..' 10 0. 1 2 3 4 5 6 7 8 9 10 _____________________ Consumption Deciles Source: SHIES 1995. Annex Figures 2a and 2b Urban Education Levels Rural Education Levels Not stated Not stated 10% t No 10% No Tertiaryuato ~~~education 3%;w~F educto Tertiary- 3%edcaio ******~* 28%42 7%/ ^ *+++*+++i 28% Secondary : ::-, 6% Ii Secondar I II 12%/3t Prim ary 3% Primary 43% Source: SHIES 1995. 22 SUHHARY - SV4w: PWWR PAWIy4 6404 S4A4 1q2O4 Annex Figure 3 Volatility in Terms of Trade (OlT) Index 20. - o 15 A10 5, 9 -20 -30 1990 1991 1992 1993 1994 1995 1996 1997 1998 Years Source: Africa Live Database, World Bank. Annex Figure 4 Sectoral Growth Rats (per cent) 30Q00% 25.0%/ 20.00D/O U -5*M/o \ X i \ / -10.000 C \ -5.0/o 0 -20.000/0 -25.ODI/o 1990 1991 19S2 1993 1994 1995 1995 1997 1968 Yemrs Source: Africa Live Database, World Bank. 23 SLHAI/ - 5weA R P-t 54d L"4g Summary Annex Table 1. Informational Needs and Potential Sources for a Swazi Poverty Monitoring and Analysis System Issue to Address Information Needed Frequency Disaggregation Potential Information Source Input Monitoring Are resources allocated to Financial and budget Annual Regionlinkhundla Budget and financial appropriate programs in statistics records accordance with plans? output (Program) otoig Are poverty reduction Records of services Annual Regionlinkhundla Administrative activities implemented in provided, facilities records accordance with plans? constructed, etc. Beneficiary Do the poor have Distance to facilities by Annual Regionlinkhundla/ PS or CWIQ improved access to SEG (at household or SEG services/ programs? community level) Are they making use of Utilization rates (e.g., Annual Regionlinkhundlal PS or CWIQ them? enrollment rates) SEG Are poor satisfied with Measures of Annual Regionlinkhundlal CWIQ and PPA services? Do services satisfaction SEG meet their needs? Impact Evaluation Are people healthier, GDP Annual National National accounts: wealthier, better educated, better employed? % below poverty line Every 5 years Region/ HH budget or IS (P-alpha measures) (possibly inkhundla/ annual on SEG Price statistics HH data on income, small sample) consumption, employment, health, education Dynamics of poverty (who As above Annual Selected Panel study moves in and out of communities/ poverty) SEGs What are characteristics of Time use, Every 3-5 Regionlinkhundla/ Integrated surveys, intrahousehold resource intrahousehold years SEG PPA, and/or Panel allocation and use? decisionmaking study authority Source: World Bank Staff (based on a prototype). Notes: SEG = Socioeconomic Group; IS = Integrated Survey; PS = Prionty Survey; CWIQ = Core Welfare Indicators Questionnaire; PPA = Participatory Poverty Assessment; HH = household. 24 SUHMARY - Sw,LJ : R4ce vt " 5 d q4w KEY REFERENCES Central Statistical Office, Mbabane, Swaziland 1995. Swaziland Household Income and Expenditure Survey, 1995: Main Report Delgado, C. L., Hazell, P., Hopkins, J., and Kelly, V. 1994. "Promoting Intersectoral Linkages in Rural Africa Through Agricultural Technology and Policy Reform." American Journal of Agricultural Economics, Vol. 76, No. 5, December. Drought Consortium of NGOs, 1996. Proposal for Implementation of a Drought Preparedness Program (mimeo). Government of the Kingdom of Swaziland, 1997. Swaziland: Poverty Assessment by the Poor, Report on Participatory Poverty Assessment. ___._ (various years) Development Plans, 1997/98 - 1999/OD, Mbabane: Ministry of Economic Planning and Development. International Monetary Fund. 1997. Recent Economic Developments, Washington, D.C. (mimeo) MOAC (Ministry of Agriculture and Co-operatives) 1999. People, Employment and Poverty: A Socioeconomic Study of Rural Communities in the Southern Lowveld of Swaziland. Mbabane: Ministry of Agriculture and Cooperatives. MTEC (Ministry of Tourism, Environment and Communications) 1997. Swaziland Environment Action Plan. Mbabane: Ministry of Tourism, Environment and Communications. Statistics Sweden (Gum Alm Stenflo). 1997. Poverty Analysis of Swaziland. (mimeo) UNDP. 1998. Swaziland National Human Development Report 1997. Swaziland: UNDP. UNICEF. February, 1999, "Stimulating Orphan Programming in Swaziland: Government and Community Response in a Multisectoral Context", UNICEF Report of an Assessment of Programming in Swaziland for Families and Children Affected by HIV/AIDS. University of Swaziland (UNISWA), 1999, by Patricia Musi. Institutional Framework for Poverty Reduction Programs. (mimeo) World Bank 1998a, African Development Indicators, Washington, D.C. . 1998b. Swaziland: Key Issues in the Agricultural Sector. Washington, D.C. . 1990. World Development Report, Washington, D.C. 25 IBRD 19802 310 320 S W A Z I L A N D -M MAIN ROADS, TARRED MAIN ROADS, UNTARRED I RAILROADS FOREST RESERVES + INTERNATIONAL AIRPORT Hectorspruit * MINES - DAMS - ~~~~~~~~~~~~~~~~~Ntfonjenir RIVERS : OTHERS To BTOWS \ MOZAMBIQUE O DISTRICT CAPITALS Kmtpot- 0 NATIONAL CAPITAL 5h ~ DISTRICT BOUNDARIES 260 -~ - -INTERNATIONAL BOUNDARIES neMhlume / (M~AGUGCA DdAM a 9)- ^ 5 . j Iron~~~~~~~Popsd Cafngrn Mle/s HOM J s X -~ v M To Catuane / 1 x~one; / To Johannesburg / X t%OAHMOhzu~~~~~wini S / ~~~~~~~~~~~~~~Matsapha 7 V S LImTEKI\ _~~~~~~~~~~~~~ay I/~ LUB BOB To Amsterdam U M NI2 HE RICKi ° B ^ 470->ig Ben C~~~~T vV ef cu ua >itobela = Jj y IeGege g < / tlge 4 To Piet Ret,e| 27 _ 27r. ,;P \),g_&)No / '\ >s w 5 ( ~~~~~~~~~~Lukhalweni , - T§ f ~To Piet Retief q hlnno \ 'Z ° / i \~~~~~Mohn ISELE W NI NAMIBIA e J.G STRIJDO HBOTSWANA> 8< DA ' SWAZILAND-kJ To Nongoma To Durban rO==. 't, Zr -* 7 S ,1 K l L O ME TE RS SRa ~ hbhKILOMETERS \-J LSTO0 10 20 30 40 50 SOUTH AFRICA ' 0 S 10 15 20 25 30 L=--; W tND§AN OCEAN MILES 320 o m e. JULY 1986 rA -ji :0 NZ lo Kl Kr 7V qj