70113 Fiscal Dimensions of HIV/AIDS in SADC Member Countries Background Paper Prepared for Technical Conference on “Sustaining HIV and AIDS Responses in the Context of Shrinking Resources in the SADC Region� August 24–26, 2010 Conference Edition © August 2010 The International Bank for Reconstruction and Development World Bank, ACTafrica—AIDS Campaign Team for Africa The World Bank 1818 H Street, NW Washington, DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org E-mail: feedback@worldbank.org All rights reserved. The findings, interpretations, and conclusions expressed in this volume do not necessarily reflect the views of the Executive Directors of the World Bank or the governments that 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 nor the endorsement or acceptance of such boundaries. Rights and Permissions The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The International Bank for Reconstruction and Development/The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center Inc., 222 Rosewood Drive, Danvers, MA 01923, USA; telephone: 978-750-8400; fax: 978-750-4470; Internet: www.copyright. com. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street, NW, Washington, DC 20433, USA; fax: 202-522-2422; e-mail: pubrights@worldbank.org. | III Contents Acknowledgments IV I Introduction 1 II HIV/AIDS Spending and Financing in Across SADC Member Countries 2 III The Global Environment and the Financing of HIV/AIDS Programs 5 IV Introduction to Country Studies 7 V Fiscal Dimension of HIV/AIDS in Botswana 7 VI Fiscal Dimension of HIV/AIDS in South Africa 11 VII Fiscal Dimension of HIV/AIDS in Swaziland 15 VIII Concluding Remarks 18 References 19 IV | Fiscal Dimension of HIV/AIDS in SADC Member Countries Acknowledgments This background paper draws on an ongoing World Bank work program on the fiscal dimension of HIV/ AIDS, led by Elizabeth Laura Lule1 (Manager, AFTQK), and conducted by Markus Haacker2 (Con- sultant, based at the London School of Hygiene and Tropical Medicine for the duration of this study). Shantayanan Devarajan (Chief Economist) provided technical guidance, Markus P. Goldstein (World Bank), Antonio C. David (World Bank/IMF), Andy Berg (IMF), and Jacques van der Gaag (Brookings Institution and Amsterdam International Institute for Development) acted as peer reviewers of the under- lying work. Eugenia M. Marinova (AFTHV) coordinated the work program with country governments, agencies, and representatives of other international organizations. Michael A. Obst (Arco Verlag) provided research assistance. Administrative aspects were managed by Sangeeta Raja (AFTHV), and Carolyn Shelton (AFTHV) provided editing support. Yvette M. Atkins (AFTHV) and M. Javed Karimullah (AFTHV) provided administrative support. We are very grateful for the support and inputs from government officials in Botswana, South Africa, and Swaziland, and for the feedback and support from numerous colleagues, listed in the underlying studies. The findings, interpretations, and conclusions expressed in this paper do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organi- zations, or those of the Executive Directors of the World Bank or the governments they represent. 1 Contact details Elizabeth Lule: elule@worldbank.org, phone 1-202-473-3787. 2 Contact details Markus Haacker: contact@devgro.com, http://mh.devgro.org, phone 44-753-5642332. | 1 I. Introduction The impact of and the response to HIV/AIDS [Democratic Republic of Congo] and US$11,400 (human immunodeficiency virus/acquired immu- [Seychelles] in 2010 [IMF 2010]), the state of the nodeficiency syndrome) poses complex policy chal- response to HIV/AIDS, and the extent of external lenges across the Southern African Development support. Indeed, some of the analysis in this paper Community (SADC) region, which includes many builds on these country differences, although many countries facing generalized HIV epidemics as of the fiscal aspects of HIV/AIDS can adequately be well as some of the countries with the highest HIV addressed only in country-specific studies. prevalence rates worldwide. According to UNAIDS Section III provides a cross-section of the re- (2008a), there were 13.6 million people living with sponse to HIV/AIDS across SADC countries from HIV/AIDS (PLHIV) in SADC member countries, a fiscal perspective, focusing on spending and financ- corresponding to over 40 percent of the number of ing data that are available across countries. This yields PLHIV globally, even though the region’s overall some insights regarding the scale of the fiscal burden population accounts for less than 4 percent of the of HIV/AIDS across countries. However, this cross- world population. However, HIV prevalence ranges sectional analysis does not capture crucial aspects of widely across SADC member countries, from 0.1 the fiscal dimension of HIV/AIDS (the role of the percent of the population aged 15–49 (Madagascar) fiscal context, or the persistence of the fiscal costs to 26.1 percent (Swaziland) in 2007. of HIV/AIDS), and therefore encourages the use of The objective of the World Bank’s work program country studies, which adopt a much richer epide- on fiscal dimensions of scaling up the HIV/AIDS miological and fiscal framework. response and this background paper is to highlight Section IV introduces the three country studies and assess fiscal policy challenges arising from the (Botswana, South Africa, and Swaziland) that have response to HIV/AIDS, and to assist countries in se- been conducted so far. Among the countries with lecting HIV/AIDS-related policies and deciding on the highest HIV prevalence globally, these three resource allocations in a sustainable manner. At the countries finance a substantial proportion of their outset, it is important to acknowledge that SADC response to HIV/AIDS from domestic sources. member countries are diverse in terms of the fiscal Beyond these similarities, analysis shows substan- dimension of HIV/AIDS, not only because of dif- tial differences between these countries in terms of ferences in the levels of HIV prevalence, but also in the fiscal dimension of HIV/AIDS. For example, terms of their economic context (with levels of gross although the early and comprehensive response to domestic product [GDP] per capita between US$200 HIV/AIDS in Botswana (section V) translates into 2 | Fiscal Dimension of HIV/AIDS in SADC Member Countries high fiscal costs, the fiscal burden is already declin- ing, as the costs incurred by the reduced number of II. HIV/AIDS Spending and new infections are much lower than current expen- ditures. In South Africa (section VI), the large role Financing across SADC of social grants in the national budget implies that much of the fiscal repercussions of HIV/AIDS oc- Member Countries cur outside the country’s HIV/AIDS program. And The principal sources of data on HIV/AIDS-related in Swaziland (section VII), the increase in the fiscal spending and the financing of the HIV/AIDS pro- costs of HIV/AIDS coincides with a difficult out- gram are the United Nations General Assembly Spe- look for government revenues (shrinking Southern cial Session (UNGASS) reports compiled by national African Customs Union [SACU] revenues). authorities in collaboration with the Joint United Section VIII brings together the different strands Nations Programme on HIV/AIDS (UNAIDS), and of the analysis, addresses some of its limitations, and made available on the UNAIDS Web site.3 Table  1 places the work in the context of the evolving policy summarizes the latest available information from these response(s) to HIV/AIDS. sources (augmented in a few cases where the spend- The work program on the fiscal dimension of ing data in the UNGASS reports were incomplete).4 HIV/AIDS is relevant for at least three types of ac- A word of caution is in order when interpreting these tors involved in the international and national re- data—most data (for nine countries) relate to 2008, sponses to HIV/AIDS: but there are also numerous observations for 2009 ♦ Individuals involved in planning the national (four countries), and two countries for which data are responses to HIV/AIDS, by providing insights on only available for 2007 (Namibia) or 2006 (Swazi- the implications of the epidemic and HIV/AIDS land). As HIV/AIDS-related spending has generally programs, and by highlighting fiscal aspects on a been increasing over the period 2006–9, the differ- range of policy choices. ences in years covered may introduce a distortion. ♦ Individuals (located, for example, in a ministry of With this in mind, this analysis finds that to- finance) in charge of developing and implement- tal HIV/AIDS-related spending has for the years ing the national policy and development agenda, covered accounted for US$3.8 billion (0.8 percent by providing tools to analyze the interactions of GDP) across SADC member countries, corre- between the impact of and the response to HIV/ sponding to US$15 per capita. External financing AIDS on one hand, and the fiscal resource enve- accounted for about half of total spending for the lope on the other hand. years shown. Regarding the cost burden of HIV/ AIDS from a national perspective, the most ap- ♦ Individuals planning or observing the interna- propriate measure is the cost relative to GDP. Data tional response to HIV/AIDS, by providing tools summarized in table 1 indicate how the scale of the to analyze the links between external support and epidemic plays an important role, as well as eco- the fiscal burden of HIV/AIDS. nomic factors. For example, Lesotho has the high- 3 http://www.unaids.org/en/KnowledgeCentre/HIVData/ CountryProgress/ 2010CountryProgressAllCountries.asp. 4 For Swaziland, data are also taken from NERCHA and UNAIDS (2008), and for Zambia, data are drawn from NAC Zambia (2010). CONFERENCE EDITION | 3 Table 1 SADC: HIV/AIDS Spending and Financing HIV/AIDS spending Total External financing Country Year (US$ millions) % of GDP Per capita (US$) (% of total) GDP per capita (US$) Angola 2009 33 7 0 05 19 na 3,972 Botswana 2008 348 1 26 194 4 32 1 7,552 Congo, Dem 2008 96 4 08 15 86 0 184 Rep of Lesotho 2008 56 4 36 22 9 53 1 645 Madagascar 2008 12 0 01 06 54 7 468 Malawi 2008 107 4 26 78 97 6 298 Mauritius 2008 na na na na 7,330 Mozambique 2008 146 4 15 71 95 6 478 Namibia 2007 18 5 02 91 49 2 4,341 Seychelles 2009 06 01 68 19 4 8,973 South Africa 2009 2,088 0 07 42 3 27 3 5,824 Swaziland 2006 48 5 18 47 7 61 3 2,698 Tanzania 2008 465 0 23 11 7 98 1 519 Zambia 2008 279 3 26 23 5 97 1 901 Zimbabwe 2009 54 1 12 46 69 8 375 Total (latest years) a 3,745 5 08 14 7 49 9 1,782 Sources: UNGASS Country Reports 2010 for HIV spending, augmented by domestic sources for Swaziland and Zambia, IMF (2010) for GDP. Note: n.a. = not available. a. Excludes Mauritius. est level of spending relative to GDP (3.6 percent, ing reflects that GDP per capita in Malawi (in US$ which is among the countries with the highest level terms) was only 4 percent of the level of GDP per of HIV prevalence along with Botswana and Swa- capita in Botswana. ziland) and a much lower level of GDP per capita. Significant differences in the role of external The next highest spending, at 2.6 percent of GDP, financing are also evident across SADC member occurs in a group of countries including Botswana, countries. While external support accounted for Malawi, and Zambia. While the level of HIV prev- about half of the costs of HIV/AIDS programs in alence in Malawi was only about half of the level the region, this share ranged from about 20 percent for Botswana, and access to HIV/AIDS-related services was much lower,5 the high level of spend- coverage rate of antiretroviral treatment in Malawi was 51 percent (Office of the President and Cabinet, 2010), whereas in Botswana 5 In 2008 (the year the spending data in table 1 relate to), the it was 82 percent (NACA and UNAIDS 2010). 4 | Fiscal Dimension of HIV/AIDS in SADC Member Countries Table 2 SADC: Domestically Financed HIV/AIDS Spending Domestically financed HIV/AIDS Total HIV/AIDS spending Domestically financed HIV/AIDS spending (% of government Country Year (% of GDP) spending (% of GDP) expenditure) Angola 2009 0 05 na na Botswana 2008 26 17 44 Congo, Dem 2008 08 01 05 Rep of Lesotho 2008 36 17 26 Madagascar 2008 01 01 na Malawi 2008 26 01 02 Mauritius 2008 na na na Mozambique 2008 15 01 02 Namibia 2007 02 01 04 Seychelles 2009 01 01 02 South Africa 2009 07 05 16 Swaziland 2006 18 07 21 Tanzania 2008 23 00 02 Zambia 2008 26 00 02 Zimbabwe 2009 12 04 15 Sources: UNGASS Country Reports 2010 for HIV spending, augmented by domestic sources for Swaziland and Zambia, IMF (2010) for GDP. Various IMF country reports for government expenditures. in South Africa to 98 percent in Malawi. The extent three countries facing the highest domestic financ- to which external support has enabled the response ing burden, according to table 2, are the very same to HIV/AIDS in SADC member countries with countries with the highest levels of HIV prevalence relatively low levels of GDP per capita is illustrated (Botswana, Lesotho, and Swaziland). Thus, the inter- in table 2. For example, out of total HIV/AIDS-re- national response to HIV/AIDS has provided partial lated spending of between 1.5 percent of GDP and insurance to countries facing high levels of HIV prev- 2.6 percent of GDP in Malawi, Mozambique, and alence—providing support to HIV/AIDS programs Zambia, high levels of external support (exceeding in countries like Botswana or South Africa that oth- 95 percent of the costs of the HIV/AIDS program) erwise (in light of their relatively high income levels) have reduced the domestic financing needs to around receive limited grant funding. However, this insurance 0.1 percent of GDP or less. is partial because high-prevalence countries also cover Another conclusion from the data presented in substantial program costs from domestic resources. table 2 addresses the link between disease burden As the fiscal costs of HIV/AIDS predominantly and domestically financed HIV/AIDS spending. The arise in a few sectors (notably, health, education, and CONFERENCE EDITION | 5 social spending), it is also instructive to relate the therefore serves as an indicator for the operational level of HIV/AIDS-related spending to such subsec- challenges of scaling up HIV/AIDS-related services. tors. Table 3 compares the level of HIV/AIDS-re- Table 3 offers some illustration on the scale of lated spending with total or public health spending. the policy (and financing) challenges that the HIV/ Because HIV/AIDS-related spending transcends AIDS response poses in a number of SADC mem- the health sector, these figures cannot be interpreted ber countries. Total HIV/AIDS-related spending in terms of HIV/AIDS-related spending absorbing a is equivalent to at least 40 percent of total health certain share of health spending. However, the pro- expenditure in four countries (Botswana, Lesotho, vision of health services is a significant aspect of pub- Tanzania, and Zambia), and exceeds the equivalent lic services—if the level of HIV/AIDS-related ex- of 60 percent of public health expenditure in five penditure is large relative to health expenditure, this countries (the four countries referred to above, plus the Democratic Republic of Congo). From a fiscal perspective, the most immediate fi- Table 3 SADC: HIV/AIDS Spending and nancing challenges arise from domestically Health Spending financed HIV/AIDS-related expenditure; Domestically this exceeded the equivalent of 40 percent Total HIV/AIDS financed HIV/ Total health Public health of public health spending in two countries spending AIDS spending spending (% spending (% (Botswana and Lesotho). Looking further Country (% of GDP) (% of GDP) of GDP) of GDP) ahead, externally financed HIV/AIDS- Angola 0 05 na 2 54 2 04 related spending also poses considerable Botswana 2 57 1 75 5 71 4 26 fiscal challenges, arising from the need to solicit the required external support or re- Congo, Dem Rep of 0 83 0 12 5 75 1 20 sponding to shortfalls in anticipated exter- nal support. Lesotho 3 55 1 67 6 22 3 63 Madagascar 0 13 0 06 4 13 2 74 Malawi 2 63 0 06 9 90 5 90 III. The Global Mauritius na na 4 17 2 04 Mozambique 1 48 0 07 4 93 3 54 Environment and the Namibia 0 21 0 11 7 63 3 21 Financing of HIV/AIDS Programs Seychelles 0 08 0 06 5 13 3 60 South Africa 0 73 0 53 8 62 3 57 Swaziland 1 77 0 68 6 01 3 76 The global financial crisis affects fiscal Tanzania 2 25 0 04 5 32 3 50 space available for the financing of HIV/ Zambia 2 61 0 08 6 16 3 56 AIDS programs through two channels— Zimbabwe 1 23 0 37 8 95 4 14 its impact on SADC member countries, and its impact on main donor countries Sources: UNGASS Country Reports 2010 for HIV spending, augmented by do- mestic sources for Swaziland and Zambia, IMF (2010) for GDP. Data on HIV/AIDS- (and thus, the availability of external sup- related spending relate to years shown in table 1. Data on health spending are port). The role these different channels play from 2008. 6 | Fiscal Dimension of HIV/AIDS in SADC Member Countries Figure1. SADC: GDP Growth, 2005–12 percent of GDP in 2007/8 to 43 percent of GDP in 2012/13 (South Africa 2010). 8 (Annual growth, average, population-weighted) Given that approximately half of HIV/AIDS- 7 related spending across SADC member countries is 6 5 GDP growth financed through external support, the impact of the 4 global crisis among the main donor countries is as 3 important as the domestic impact. Moreover, exter- 2 nal financing accounts for more than 90 percent of 1 0 HIV/AIDS spending in four SADC member coun- -1 tries (Malawi, Mozambique, Tanzania, and Zambia), Estimates Projections -2 and at least for these countries, the changing global 2005 2006 2007 2008 2009 2010 2011 2012 environment has immediate and significant implica- Source: IMF World Source: IMF (2010). Economic Outlook Database, April 2010 tions for the financing of their HIV/AIDS programs. Figure 2 summarizes estimates and projections for the financing of HIV/AIDS programs depends for key macroeconomic variables (GDP growth on the extent to which a country depends on external and the fiscal balance) for advanced economies support. overall, the United States (a major contributor to For SACU member countries overall, GDP HIV/AIDS programs through the U.S. President’s growth has declined from an average of 7 percent Emergency Plan for AIDS Relief [PEPFAR]), through 2007 to -1 percent in 2009, followed by a the Euro area, and the United Kingdom (major gradual recovery (figure 1). This means that domes- contributors to the Global Fund). As for SADC tic resources expanded much slower than expected member countries, advanced economies have expe- in 2008. Regarding the outlook for the financing of rienced a steep drop in economic growth, although HIV/AIDS programs (and any other public policy the decline is somewhat less pronounced than for objectives), this negative economic impact is com- SADC (about 5 percentage points, rather than 7 pounded by the fiscal repercussions of the crisis. percentage points), and recovery is expected to oc- While data for the budget turnouts for fiscal cur more quickly. At the same time, the fiscal bal- years 2009/10 are not yet available, budget data for ance deteriorates steeply, and remains in deficit for Botswana and South Africa, two countries that fi- the coming years. For the United States, the dete- nance their HIV/AIDS programs predominantly rioration in the fiscal balance is extraordinary, and from domestic sources, can be used as an example. the fiscal deficit estimated for 2009 is the highest In Botswana, government finance depends largely recorded since the end of World War II (Council on revenues from the mining sector, which has been of Economic Advisors 2010). As a result, the In- severely hit by a slump in the demand for diamonds. ternational Monetary Fund (IMF 2010) projects As a consequence, the preliminary budget turnout that the level of public debt in the United States for fiscal year 2009/10 is a deficit of 16 percent of will double relative to GDP between 2007 and GDP, projected to recover only gradually over the 2015 (from 42 percent of GDP to 86 percent of next couple of years. In South Africa, the fiscal bal- GDP). In the United Kingdom, the economy con- ance has deteriorated from a surplus of 1.7 percent of tracted by 5 percent in 2009, and the deterioration GDP in 2007/8 to a deficit of 7.3 percent of GDP in in the fiscal balance is of a similar magnitude as for 2009/10, and public debt is projected to rise from 28 the United States: net debt is projected to increase CONFERENCE EDITION | 7 Figure 2. Macroeconomic Trends, Advanced Economies, 2005–12 Figure 2A. GDP Growth, Advanced Economies, 2005-12 Figure 2B. Fiscal Balance, Advanced Economies, 2005-12 (Percent) (Percent) 4 0 3 Advanced -2 economies 2 -4 1 Euro Area 0 -6 Advanced economies -1 United States -8 Euro Area -2 United -10 United States United -3 Kingdom Kingdom -12 -4 -5 -14 2005 2006 2007 2008 2009 2010 2011 2012 2005 2006 2007 2008 2009 2010 2011 2012 Source: IMF World Economic Outlook Database, April 2010. Source: IMF World Economic Outlook Database, April 2010. Source: IMF (2010). from 38 percent of GDP in 2007 to 82 percent of grams cannot adequately be addressed through GDP in 2012. While fiscal deterioration has been recourse of the fiscal and macroeconomic summary less severe in the Euro area, the fiscal position was indicators available across countries discussed above. weaker in some countries at the outset, resulting in One important aspect of the HIV/AIDS response is the “Greek crisis� and putting the treasury bonds of the persistence of the fiscal commitments—current a number of countries under pressure. treatment and impact mitigation programs require Thus, the fiscal environment in advanced econo- long-term expenditures over the next decades. To mies over the coming years will be much tighter understand the extent to which the response to HIV/ than expected previously, and the fiscal repercus- AIDS absorbs the available fiscal space, it is neces- sions of the global economic crisis will extend well sary to rely on projections, based on country-specific into the economic recovery. Consequently, increasing data, of the state of the epidemic and parameters demands for funding of HIV/AIDS programs face describing the response to HIV/AIDS. tighter competition. It is against this background Botswana, South Africa, and Swaziland face a that UNAIDS (2010) has shifted the emphasis from high disease burden and corresponding high costs for estimating international funding gaps to develop- their responses to HIV/AIDS (overall, and in terms ing “smarter, faster, lower cost and more effective of domestically financed spending). Summarized be- solutions.� This particularly applies to a number of low are the findings from the studies conducted in SADC member countries, as the combination of these countries. high costs of HIV/AIDS programs and high levels of external financing makes them particularly vulner- able to shortfalls in external support. V. Fiscal Dimension of HIV/ IV. Introduction to Country AIDS in Botswana Studies Botswana is among the countries with the highest level of HIV prevalence in the world; according to Many of the questions relevant to managing the UNAIDS (2008a), prevalence among the population fiscal dimensions and planning of HIV/AIDS pro- of ages 15–49 was 23.9 percent, and 300,000 people 8 | Fiscal Dimension of HIV/AIDS in SADC Member Countries were living with HIV. As a consequence of HIV/ Figure 3. Botswana: Health Expenditure by AIDS, key health indicators have deteriorated cata- of 8.2. Botswana: Health Expenditure by otherwise) SourceFigureFinancing axis), unless statedSource of Financing (US$ millions (left strophically—life expectancy at birth has declined Percent of Total Health Expenditures 700 40 Externally financed from 64 years in 1990 to 51 years in 2000 (World 600 (right scale) 35 30 Bank 2010), and the probability of reaching age 50 500 US$ millions 25 400 Private has dropped to 55 percent (compared to 88 percent 300 Private Households - Households - 20 Prepaid without AIDS) for the 2005–10 period (United 200 Out-of-pocket Public Health Expenditure 15 10 Nations Population Division 2009). 100 Non-Profit Institutions 5 0 0 Numerous studies have addressed the macroeco- 1995 1997 1999 2001 2003 2005 2007 nomic consequences of HIV/AIDS. A recent com- Source: WHO 2010. Source: WHO (2010). prehensive study by Jefferis, Siphambe, and King- Between 2001 and 2005, public health expendi- horn (2006) estimates that annual GDP growth is tures accelerated rapidly relative to GDP, to about reduced by about 1.2 percentage points, implying a 5 percent, before falling back to a level of less than moderate increase in the growth rate of GDP per 4 percent. In absolute terms, expenditures increased capita of about 0.4 percentage points. However, the from a level of about US$125 million in 2002 epidemic has resulted in a disconnect between eco- (US$110 per capita) to US$473 million in 2005 nomic development and human development. Life (US$388 per capita), and remained at about this expectancy is about 20 years lower than in coun- level. External financing played a subordinate role tries with a similar level of GDP per capita (such as (less than 1 percent of total health spending), at least Turkey, Romania, Malaysia, and Argentina), and at until 2002, but increased to around 5 percent of total about the same level as in Kenya (GDP per capita health spending by 2008. one-ninth of Botswana’s) and Ethiopia (GDP per Estimates of HIV/AIDS-related spending are capita one-sixteenth of Botswana’s). Consequently, available from two National AIDS Spending As- Botswana’s ranking according to the United Nations sessments, the first one (NACA and UNAIDS Development Programme (UNDP) Human Devel- 2007) covering 2003–5 and the second one cover- opment Index slipped from 71 in 1996, to 125 as of ing 2006–8 (NACA and UNAIDS 2010). HIV/ 2007; and whereas Botswana ranks 60th in terms of AIDS-related spending increased from 1.9 percent GDP per capita, it ranks 159th (among 181 coun- of GDP in 2003 to 2.6 percent of GDP in 2008, tries covered) in terms of life expectancy. corresponding to a nominal increase from US$150 Because of high mineral revenues (typically close million to US$348 million. Much of the increase to 20 percent of GDP through 2007), government in HIV/AIDS-related spending was enabled by ex- spending is relatively high in Botswana. Botswana ternal support, increasing from 0.1 percent of GDP was hard hit by the global financial crisis, and fol- to 0.8 percent of GDP (and from US$12 million lowed an expansionary fiscal policy in response. The to US$112 million in absolute terms). Bilateral fi- coming budget years will see an adjustment from nancing (largely from the United States) accounted the very high fiscal deficits, exceeding 10 percent of for about two-thirds of external support in 2006–8. GDP. As the role of the mineral sector and the cor- Notably, about 30 percent of external support came responding fiscal revenues are expected to decline from other international sources, reflecting high over the coming years, longer-term fiscal projections levels of support from private international sources, need to consider this shrinking resource envelope. largely through the African Comprehensive HIV/ CONFERENCE EDITION | 9 AIDS Partnership (ACHAP) funded by Melinda Figure 4. Botswana: HIV Incidence and HIV/ and Bill Gates Foundation and Merck, and the AIDS-Related Mortality, 1980–2030 (% of Clinton Foundation. Meanwhile, spending from do- population, age 15+) mestic sources (dominated by public spending) has 4.0 remained flat relative to GDP. 3.5 The analysis of the fiscal dimension of HIV/ 3.0 AIDS combines three elements: i) estimates and 2.5 Incidence projections of the state of the epidemic; ii) estimates 2.0 HIV/AIDS-related mortality and projections of the fiscal costs of HIV/AIDS; and 1.5 iii) a model and assumptions describing the macro- 1.0 economic and fiscal context. Assumptions regarding 0.5 the state and course of the epidemic were taken from 0.0 1980 1985 1990 1995 2000 2005 2010 2015 2020 2025 2030 NACA (2008) and Stover (2008), and updated in a number of places (for example, to incorporate the Source: Author's estimates and projections. Source: Author’s estimates and projections. latest estimates of access to treatment). Underlying estimates of the size and the structure of the popu- Figure 5. Botswana: PLHIV, 1980–2030 (% of lation were taken from the United Nations Popula- population, age 15+) tion Division (2009). Looking forward, certain as- sumptions (HIV incidence and coverage of a number 30.0 R ec eiving 2 nd Lin e of interventions directly affecting the course of the T reatm ent 25.0 R ec eiving 1s t Lin e epidemic, such as treatment access and prevention 20.0 T reatm ent U nm et D em and , 1s t of mother-to-child transmission) were calibrated 15.0 Lin e T reatm ent N ot R eq uirin g in line with targets in the draft National Strategic T reatm ent 10.0 Framework for HIV and AIDS 2010–16. 5.0 The projections envisage further gradual declines 0.0 in HIV incidence rates (and—after a small rebound 1980 1985 1990 1995 2000 2005 2010 2015 2020 2025 2030 as the immediate effects of scaling up treatment wear Source: Author’s estimates and projections. off—declines in HIV/AIDS-related mortality, figure 4). HIV prevalence declines over the projection period, Pula (P)  3 billion to Pula 5.5 billion, but (at least even though the absolute number of PLHIV continues from 2015) decline slowly in relation to GDP. The to increase slowly over the projection period. While biggest component of the fiscal costs of HIV/AIDS, HIV prevalence (ages 15+, somewhat lower than and the factor that dominates the increase in costs, prevalence for ages 15–49) decreases steadily from a is care and treatment, rising from P 1.3 billion to peak of 24 percent in 2002 and 21 percent in 2010 P 2.5 billion, reflecting the increasing number of to 18 percent in 2030, the number of people receiving people receiving treatment (rising from 119,000 in treatment increases sharply, rising from close to zero 2010 to 168,000 in 2030), as well as the increasing in 2000 to 8.6 percent of the adult population (and role of second-line treatment (about 30 percent of 40 percent of PLHIV) and 9.6 percent of the adult people receiving treatment in 2030). Another impor- population (52 percent of PLHIV) by 2030 (figure 5). tant factor is the increase in the costs of mitigation, The fiscal costs of HIV/AIDS increase steadily which reflects the increase in the number of orphans over the projection horizon, almost doubling from through much of the projection period. 10 | Fiscal Dimension of HIV/AIDS in SADC Member Countries In terms of their long duration, the fiscal costs also yield insights regarding the size of the fiscal of HIV/AIDS resemble a liability that needs to be burden of HIV/AIDS. At a discount rate of 3 per- repaid over a long period of time. This means that cent, the present discounted value of the fiscal costs instruments used to assess the sustainability of debt of HIV/AIDS amounts to 192 percent of GDP, of which costs equivalent to a present discounted value Figure 6. Botswana: Fiscal Costs of HIV/AIDS, of 73 percent of GDP are incurred by 2030. 2010–30 (pula billions, 2009 prices) While the fiscal costs of HIV/AIDS are ultimate- 6.0 ly driven by infections, the costs at any point of time Impact on public 5.0 servants Overhead reflect infections that occurred many years before. 4.0 Conversely, a change in HIV incidence affects the fis- 3.0 Mitigation cal costs only very slowly. To clarify the link between 2.0 HIV incidence and the fiscal costs of HIV/AIDS, first Care and treatment 1.0 Prevention the costs incurred by one new infection are estimated. 0.0 For example, for an infection occurring in 2010, over- 2010 2015 2020 2025 2030 all costs incurred are estimated to be equivalent to P Source: Author’s estimates and projections. 92,000, corresponding to about two times GDP per capita. On the macroeconomic level, the estimated fis- Figure 7. Botswana: Fiscal Costs of HIV/ AIDS, 2010–30 (% of GDP) Figure 9. Botswana: Costs of One Additional 4.0 Impact on public HIV infestion (pula, 2009 prices) 3.5 servants 7,000 3.0 Overhead Total 6,000 Present discounted 2.5 value: Pula 92,000 Mitigation 5,000 2.0 4,000 1.5 1.0 Care and treatment 3,000 Treatment 0.5 2,000 Prevention 0.0 1,000 2010 2015 2020 2025 2030 0 Source: Author’s estimates and projections. 2010 2015 2020 2025 2030 2035 2040 2045 Source: Author’s calculations. Figure 8. Botswana: Present Discounted Value of the Fiscal Costs of HIV/AIDS, as of 2010 (% Figure 10. Botswana: Fiscal Costs of HIV/ of GDP) AIDS, “Commitment Basis,� 2010–30 4.0 400 Actual expenditures 350 3.0 300 All time 250 2.0 Total costs, 200 C osts through 2030 "commitment basis" 150 C osts incurred 1.0 by new infections 100 Population-based spending 50 Discount rate (percent) 0.0 0 0 2 4 6 8 10 2010 2015 2020 2025 2030 Source: Author’s estimates and projections. Source: Author’s estimates and projections. CONFERENCE EDITION | 11 Figure 11. External Financing of HIV/AIDS Pro- exceeded 1 percent of the population aged 15–49 gams across Countries (% of HIV/AIDS-related only from 1993. From that level, it escalated rapidly, spending) reaching 10 percent only five years later in 1998, and 100 increased further to just under 19 percent by 2006. 90 80 2005 UNAIDS (2008) estimates that 5.7 million people were living with HIV/AIDS in South Africa at 2006 External Financing 70 2007 60 50 end-2007. Data from antenatal clinics suggest that 40 30 Botswana HIV prevalence has stabilized in recent years (South 20 Africa 2009).7 During 2005–10, mortality increased 10 0 to a level last observed in the early 1960s. Life expec- 0 2,000 4,000 6,000 GDP per capita (U.S. dollars) 8,000 10,000 tancy (52 years) decreased to the level observed in Source: Author’s calculations, based on UNAIDS 2008 and IMF the mid-1960s, and currently is 20 years lower than (2008). in Brazil, although the level of GDP per capita is about the same. cal costs incurred by new infections account for less Unlike the impact of HIV/AIDS on key health than half of projected spending. This reflects that re- outcomes, the most comprehensive studies available duced and declining HIV incidence already result in find that the macroeconomic impact of HIV/AIDS lower new spending commitments, whereas the high has been moderate so far. Ellis, Laubscher, and Smit current expenditures primarily address the large num- (2006), adapting the macroeconomic model main- ber of infections that have occurred in the past. tained by the Bureau for Economic Research at the Botswana receives relatively high levels of inter- University of Stellenbosch, estimate that HIV/AIDS national support for its HIV/AIDS program (for is reducing GDP growth by 0.4 percent annually, and example, US$232 million, equivalent to US$130 per will continue to do so through 2020. In addition to capita, from official sources in 2008). The high levels aggregate impacts, HIV/AIDS also has distribution- of external support to Botswana reflect the extraordi- al implications. For example, surveys typically show nary burden of disease (and fiscal burden) the coun- that HIV prevalence is higher and HIV awareness try is facing. External support normally accounts for lower in population groups with lower education less than 20 percent (or 30 percent if external support levels. Moreover, access to health insurance is very from private sources is included) of the total costs limited outside the top three income deciles. of the HIV/AIDS program, broadly in line with in- In 2006/7 and 2007/8, government revenues ac- ternational practice considering Botswana’s level of counted for about 30 percent of GDP, and the bud- economic development. get returned a surplus (1.2 percent and 1.7 percent of GDP, respectively). However, the impact of the glob- al crisis has resulted in a deterioration of the fiscal VI. Fiscal Dimension of HIV/ situation and outlook. GDP growth declined from AIDS in South Africa over 5 percent in 2006 and 2007 to -2 percent in Dorrington, Bradshaw, Johnson, and Budlender (2006). Johnson and Dorrington (2006) provide a discussion of some of the under- According to the Actuarial Society of South Africa lying assumptions and methods. (ASSA 2006),6 HIV prevalence in South Africa 7 HIV prevalence among pregnant women has remained broadly 6 The data included in ASSA (2006) are further discussed by constant, at 29.1 percent in 2006 and 29.4 percent in 2007. 12 | Fiscal Dimension of HIV/AIDS in SADC Member Countries Figure 12. South Africa: Health Expenditure by Health, the Department of Education, and the De- Source, 1995–2008 (% of GDP) partment of Social Development. The largest item is 10.0 0.10 the “HIV/AIDS and STDs� in the Department of 9.0 8.0 0.08 Health, which, by 2008, was the largest item under 7.0 Private Insurance Externally financed the Department’s Strategic Health Programmes. Ex- 6.0 0.06 5.0 penditures increased from R181 million in 2000/01 4.0 Private - out of pocket 0.04 to R4.8 billion in 2009/10 (or from US$25 million 3.0 2.0 NGOs 0.02 to US$473 million) and are expected to rise to R9.3 1.0 Public billion by 2012/2013. The structure of expenditure 0.0 0.00 1995 1997 1999 2001 2003 2005 2007 changed over this period. Health services are ad- Source: WHO (2010) for health spending, IMF (2010) for GDP. ministered through the provincial budgets. As HIV/ Note: External financing refers to right scale, other categories to left AIDS-related health services expanded, and increas- scale. ing share of HIV/AIDS-related allocations under the Department of Health are accounted for by spe- 2009, and is expected to recover only slowly. Mean- cific allocations to provinces (conditional grants). while, government revenues have dropped by about 3 The national response to HIV/AIDS is guided percent of GDP, and expenditures increased by over by the HIV & AIDS and STI Strategic Plan for 5 percent of GDP, so that the fiscal balance dete- South Africa, 2007–11 (SANAC 2007), organized riorated to -7 percent of GDP by 2009/10. For the around the aims of reducing the rate of new HIV financing of the national HIV/AIDS program—as infections by 50 percent by 2011 and reducing the for other categories of public spending—this means impact of HIV and AIDS on individuals, families, that the available fiscal resources are tighter than communities and society by expanding access to what might have been expected two years ago. The appropriate treatment, care, and support to 80 per- level of GDP is expected to recover only slowly, and cent of all HIV-positive people and their families by may permanently remain lower than expected earlier, 2011. and the government expects that by 2012/13 it will In addition to the costs of the national response to have accumulated additional public debt (compared HIV/AIDS, an important aspect of the fiscal dimen- to 2008/9) equivalent to 15 percent of GDP (South sion of HIV/AIDS is the impact on social expendi- Africa 2010). tures. South Africa has established a fairly extensive Public health expenditure has remained fairly public social security system, accounting for 12 per- constant relative to GDP since 1995 (at about 3 cent of total government expenditures (3.5 percent of percent of GDP), even though the impact of HIV/ GDP) in 2009/10 (South Africa 2010). HIV/AIDS AIDS has escalated. At first glance, this would sug- affects the incidence of the conditions targeted by so- gest that the expansion of HIV/AIDS services has cial grants, such as orphanhood or disability, and ar- resulted in a reallocation from other types of health guably has contributed to the increase in the number services. However, as GDP expanded rapidly over of recipients of foster care grants (from 276,000 in this period, it turns out that the costs of HIV/AIDS 2000/01 to 569,000 in 2009/10) and disability grants accounted for only about a quarter of the increase in (from 613,000 in 2000/01 to 1,423,000 in 2006/7). health spending since 2000. On the other hand, certain categories of social grants In the national budget, HIV/AIDS-related line (for example, old-age pensions or child support grants) items occur in the budgets of the Department of are likely to decline as a consequence of HIV/AIDS. CONFERENCE EDITION | 13 Figure 13. South Africa: Adults Receiving ART, Figure 15. South Africa: Fiscal Costs of HIV/ 2000–2031 (millions) AIDS, “Narrow NSP,� 2007–31 3.5 1.4 1.2 3.0 Bas eline 1.0 2.5 Narrow NSP Prevention 0.8 Ex panded NSP 2.0 0.6 Hard Choic es Treatment (public sector) 1.5 0.4 0.2 1.0 0.0 2007 2012 2017 2022 2027 0.5 -0.2 Payroll -0.4 Other Mitigation 0.0 Social grants -0.6 Source: Guthrie and others (2010). Source: Guthrie and others (2010), and author’s calculations. Figure 14. South Africa: HIV Prevalence, 1990– Figure 16. South Africa: Fiscal Costs of HIV/ 2031 (% of population 15+) AIDS, “Expanded NSP,� 2007–31 (% of GDP) 25 1.4 1.2 20 1.0 Prevention 0.8 15 0.6 Bas eline 0.4 Treatment (public sector) 10 Narrow NSP 0.2 Ex panded NSP 0.0 5 2007 2012 2017 2022 2027 Hard Choic es -0.2 Payroll Other Mitigation 0 -0.4 Social grants -0.6 Source: Guthrie and others (2010). Source: Guthrie and others (2010), and author’s calculations. This analysis has been conducted in tandem with 35.5 in 2016, but subsequently fall to Rbn 20.8 by the ongoing “2031� analysis of the long-run costs and the end of the projection period. Relative to GDP, financing of HIV/AIDS in South Africa (Guthrie the costs peak at over 1 percent of GDP in 2012–19, and others 2010), builds on the costing developed and decline to about 0.4 percent of GDP by 2031. in this context, and is organized along three scenar- Under the expanded NSP scenario, the build-up in ios: a “narrow NSP� scenario, which is based on the costs is faster, but they decline faster over the last National Strategic Plan 2007–11, and applies 2011 decade of the projection period (as a result of the coverage rates for projections; “hard choices,� which more aggressive prevention measures early on), and builds on the narrow NSP scenario, but reallocated by 2031—at Rbn 14.6—are considerably lower than funds to prevention measures; and an “expanded in the narrow NSP scenario. NSP� scenario, envisaging enhanced eligibility for In some regards, these projections resemble es- treatment and higher treatment coverage rates (at- timates available for other countries—the costs of tained by 2015), and a scaling up of certain interven- treatment are the most important driver of the fiscal tions through 2021. costs over the next decade, and the enhanced pre- Under the narrow NSP, the fiscal costs of HIV/ vention efforts in the “expanded NSP� scenario lower AIDS almost double from Rbn 19.0 in 2009 to Rbn the fiscal costs later on. A unique feature of the fiscal 14 | Fiscal Dimension of HIV/AIDS in SADC Member Countries costs of HIV/AIDS in South Africa is the impact by one additional infection. These costs are estimated of HIV/AIDS on social grants. Notably, HIV/AIDS to dominate because of the increased need for treat- reduces the number of people reaching age 60 who ment over the 20 years following an infection, and could qualify for old-age grants (an income thresh- reduced costs of social (such as old-age) grants later old also applies). For this reason, the impact of HIV/ on. Using a discount rate of 3 percent, the costs of an AIDS on the costs of social grants, which is positive additional infection come out at R8,800 (about one- initially (as increased costs of disability grants and third of GDP per capita) for the narrow NSP (of foster care grants dominate), declines from about which R33,300 can be attributed to treatment and 2017 as cohorts highly affected by HIV/AIDS reach R30,900 to the decline in the costs of social grants), age 60. Thus, the thrust of the fiscal impact of HIV/ and R7,800 for the expanded NSP (of which treat- AIDS in South Africa occurs through two channels: ment accounts for R37,500). (i) an increase in the demand for public health ser- vices, and (ii) a slowdown of the rise in the costs of Among the factors complicating an assessment old-age grants (which would increase steeply other- of the fiscal costs of HIV/AIDS are the long time wise, reflecting demographic and health factors). These two main factors are also apparent in the Figure 19. South Africa: Fiscal Costs of HIV/ microeconomic analysis of the expected cost incurred AIDS, “Commitment Basis,� “Narrow NSP,� 2010–31 (% of GDP) Figure 17. South Africa: Costs of Additional 0.7 Infection, “Narrow NSP� Scenario (ZAR, 2009 0.6 Total costs (excluding social grants) prices) 0.5 0.4 4,000 Total costs 3,000 Total (including social grants) 0.3 2,000 Treatment 0.2 1,000 Prevention and population-based spending 0 0.1 2010 -1,000 2015 2020 2025 2030 2035 2040 2045 2050 0.0 -2,000 2010 2015 2020 2025 2030 -3,000 Social grants Source: Author’s calculations. -4,000 Source: Author’s calculations. Figure 20. South Africa: Fiscal Costs of HIV/ AIDS, “Commitment Basis,� “Expanded NSP,� Figure 18. South Africa: Costs of Additional 2010–31 (% of GDP) Infection, “Expanded NSP� Scenario (ZAR, 2009 prices) 0.9 0.8 5,000 Total 0.7 4,000 Total costs (excluding social grants) 0.6 3,000 Treatment 0.5 2,000 0.4 Total costs 1,000 (including social grants) 0 0.3 2010 -1,000 2015 2020 2025 2030 2035 2040 2045 2050 0.2 Prevention and population-based spending -2,000 -3,000 Social grants 0.1 -4,000 0.0 -5,000 2010 2015 2020 2025 2030 Source: Author’s calculations. Source: Author’s calculations. CONFERENCE EDITION | 15 Figure 21. Human Development Index, Select- ed Countries, 1980–2007 VII. Fiscal Dimension of HIV/ 0.9 AIDS in Swaziland 0.8 China Egypt 0.7 Swaziland Swaziland is considered the country with the highest 0.6 HIV prevalence in the world; an estimated 26 per- Guatemala 0.5 cent of the population is HIV positive (UNAIDS 0.4 2008a,2008b). As a consequence of HIV/AIDS, 0.3 Nepal Pakistan mortality in Swaziland has risen from 0.9 percent in 0.2 1990–95 to 1.6 percent in 2005–10 (United Nations 1980 1985 1990 1995 2000 2005 Population Division 2009), and the probability of a Source: UNDP (2009). newborn reaching age 50 has dropped from around 80 percent to just over 40 percent. CSO and Macro International (2008) report that 20 percent of young lags involved—between infection and treatment Swazis of ages 10–14 had lost at least one parent, and need and the long duration during which treatment 7.5 percent had lost both parents. is required. For this reason, prevention measures may Economic growth in Swaziland has been disap- affect the fiscal costs of HIV/AIDS only very slow- pointing in recent years, with average GDP growth ly, and enhanced prevention efforts may increase the of just 2.2 percent from 2000–10, and GDP growth costs in the short run. To get a clearer measure of per capita of 1.9 percent over the same period. How- the extent to which HIV/AIDS affects fiscal space ever, the extent to which the impact of HIV/AIDS in the longer run, the estimation starts with the fact has affected growth is difficult to establish. The slow- that most of the costs are ultimately incurred as a down in economic growth from its high levels in the consequence of new infections, and current expen- 1980s (averaging 7 percent) occurred in the early ditures predominantly serve needs caused by past 1990s, preceding the period in which HIV/AIDS infections. To measure the evolving fiscal burden could plausibly affect growth. Surprisingly few stud- of HIV/AIDS, it is necessary to estimate the costs ies address the impact of HIV/AIDS on economic caused by new infections (based on estimates of the growth in Swaziland; the bulk of evidence points costs caused by one additional infection). From this to a moderate impact of HIV/AIDS on economic perspective, the costs of HIV/AIDS on a “commit- growth so far. ment basis� are estimated at just under 0.4 percent However, as a result of the steep deterioration of GDP initially, and decline over the projection pe- in health outcomes, Swaziland has lost considerable riod. Thus, while expenditures continue to increase ground in terms of more comprehensive develop- over the coming years (serving infections that have ment indicators such as the UNDP’s Human De- occurred in the past), the underlying fiscal burden velopment Index, where between 1990 and 2005, (in terms of the amount needed to be put aside now Swaziland dropped from a cluster including coun- to cover the future costs of HIV/AIDS) is already tries such as China, the Arab Republic of Egypt, and declining. Guatemala, to the level of countries such as Nepal or Pakistan. 16 | Fiscal Dimension of HIV/AIDS in SADC Member Countries Even more so than in other countries, the response to 2.3 percent by 2014. The provision of antiretro- to HIV/AIDS in Swaziland takes place against the viral therapy (ART) through the public sector was backdrop of tightening fiscal resources. Following launched in December 2003. The number people a boom in revenues from SACU—increasing from obtaining ART rose from about 6,000 people at about 12 to 13 percent of GDP before 2003/4 to 28 end-2004 (out of estimated 45,000 in need of treat- percent of GDP in 2006/7, largely on account of an ment) to 47,000 at end-2009 (WHO, UNAIDS, economic boom in South Africa—SACU revenues and UNICEF 2008; NERCHA and UNAIDS have dropped to 7 percent of GDP in the 2010/11 2010). GoS (2009) envisages an increase in the cov- budget, which envisages a fiscal deficit of 13.4 per- erage of ART to 85 percent of adults and 90 percent cent of GDP. Therefore projections assume that of children by 2014. public expenditures will decline to just under 30 per- The latest complete data on HIV/AIDS-related cent of GDP (compared to a peak of 40 percent in spending and financing of the HIV/AIDS program 2008/9), even though public debt is projected to in- are included in the National AIDS Spending Assess- crease from 15 percent to about 50 percent of GDP, ment (NERCHA and UNAIDS 2008). Over the two and stabilize at this level. years covered, spending increased from US$40 mil- From 1995 to 2002, total health expenditures lion in 2005/6 to US$49 million in 2006/7. The most hovered at or just below 4 percent of GDP, and public important program components over these two years health expenditures accounted for about half. Since were prevention (about 20 percent of total), treatment 2002, public health spending accelerated from 2 per- and care (about 25 percent of total), and orphans and cent of GDP in 2002 to 3.8 percent of GDP in 2008. vulnerable children (absorbing about 30 percent of the Relative to government spending, the share of public total). External financing accounted for about US$30 health spending increased from 6.5 percent in 2002 million in each year, about 60 percent of total spend- to 8.5 percent in 2008. According to the latest bud- ing, largely through a grant from the Global Fund. get figures for 2010/11 (Sithole 2010), the increase This analysis of the fiscal dimension of HIV/ in public allocations for health continued through AIDS combines three elements: i) estimates and 2010/11, reaching 4.7 percent of GDP and 12.5 per- projections of the state of the epidemic; ii) estimates cent of government expenditures. External support and projections of the fiscal costs of HIV/AIDS; and increased from 2 percent of total health expenditures iii) a simple model and assumptions describing the in 2002 to 10 percent in 2008 (and the equivalent of 18 percent of public health expenditures). Figure 22. Swaziland: PLHIV, 1980–2030 (% of The national response is currently guided by the population, age 15+) National Multisectoral Strategic Framework for 30.0 HIV and AIDS 2009–14 (GoS 2009). Most HIV/ R eceiv ing 2nd Line T reatm ent R eceiv ing 1s t Line T reatm ent AIDS-related services are delivered through the 25.0 U nm et D em and, 1s t Line T reatm ent N o t R equiring T reatm ent public sector. The objectives of the strategic frame- 20.0 work can be grouped in three areas: prevention; 15.0 treatment, care, and support; and impact mitigation 10.0 (and response management, concerning the overall 5.0 efficacy of the program). In the area of prevention, 0.0 the government of Swaziland (GoS 2009) envisages 19 8 0 19 8 5 19 9 0 19 9 5 2000 2005 2 0 10 2 0 15 2020 2025 2030 a decline in adult HIV incidence from 2.9 percent Soure: Author’s estimates and projections. CONFERENCE EDITION | 17 Figure 23. Swaziland: Fiscal Costs of HIV/ Figure 26. Swaziland: Costs of Additional AIDS, 2010–30 (% of GDP) Infection, 2010 (emalangeni, 2009 prices) 7.0 7,000 Impact on public servants 6.0 Overhead 6,000 5.0 5,000 Mitigation 4.0 4,000 3.0 3,000 Present discounted value: 2.0 Care and treatment 2,000 E103,000 1.0 1,000 Prevention 0.0 0 2010 2015 2020 2025 2030 2010 2015 2020 2025 2030 2035 2040 2045 2050 Source: Author’s estimates and projections. Source: Author’s estimates and projections. Figure 24. Fiscal Costs of HIV/AIDS Figure 27. Swaziland: Fiscal Costs of HIV/ 35 AIDS, “Commitment Basis,� 2010–31 (% of 30 % of current expenditures GDP) 25 8.0 % of government revenues 7.0 Actual expenditures 20 6.0 15 5.0 10 4.0 Total costs, Costs incurred "commitment basis" 5 % of GDP 3.0 by new infections 0 2.0 2010 2015 2020 2025 2030 Population-based spending 1.0 Source: Author’s estimates and projections. 0.0 2010 2015 2020 2025 2030 Source: Author’s estimates and projections. Figure 25. Swaziland: Present Discounted Value of the Fiscal Costs of HIV/AIDS, as of 2010 (2009). Key parameters, such as treatment coverage rates, were set in line with the National Strategic Framework 2009–14. The estimates and projections of the fiscal costs of HIV/AIDS are based on the tar- gets specified in the National Strategic Framework 2009–14 (GoS 2009). Until about 2000, HIV prevalence increased sharply, driven by very high HIV incidence rates. Starting in 2003, increased access to treatment has become a critical factor underlying trends in HIV Source: Author’s estimates and projections. prevalence, while HIV incidence has been declining since about 1995. By 2010, estimates suggest that 7 macroeconomic and fiscal context. Underlying esti- percent of the population aged 15+ will be receiving mates of the size and the structure of the population treatment; this rate is projected to increase to 13 per- were taken from United Nations Population Division cent by 2020 and remain at about that level until 2030. 18 | Fiscal Dimension of HIV/AIDS in SADC Member Countries Based on the epidemiological projections, and E103,000 (about four times GDP per capita). As of the targets of the National Strategic Framework, the 2010, the costs of new infections are higher than actu- costs of HIV/AIDS and the HIV/AIDS program al spending, but decline steadily in line with targeted are estimated to be 4.4 percent of GDP in 2010, and reductions in HIV incidence. A 10 percent reduction that they will rise to 6.8 percent of GDP by 2020, in the number of new infections (relative to projected slowly declining to 6.5 percent of GDP by 2030. numbers) would result in savings in the costs of new The most important component of costs are those infections equivalent to 0.6 percent of GDP. of care and treatment, doubling from 1.4 percent of External assistance has played a key role in financ- GDP in 2010 to 2.8 percent of GDP in 2020. The ing the response to HIV/AIDS in Swaziland, ac- second largest item, mitigation, stands at 1.6 percent counting for about 60 percent of the costs (broadly in of GDP initially, rising to 2.3 percent of GDP by line with international practice). However, the fiscal 2020. These estimated costs are very large from a fis- burden of HIV/AIDS remains severe even after an cal or macroeconomic perspective, especially as fiscal allowance for anticipated external financing is made, resources are shrinking. As a consequence, the fiscal and is highly sensitive to a slowdown in the rate of costs of HIV/AIDS are projected to rise to over 30 external financing. However, in light of the steep in- percent of current expenditures. crease in the fiscal costs of HIV/AIDS projected over In terms of their long duration, the fiscal costs the coming years (and a tighter global economic en- of HIV/AIDS resemble a liability that needs to be vironment), the level of external support enjoyed by repaid over a long period of time. This means that Swaziland so far may not be sufficient in the future. instruments used to assess the sustainability of debt also yield insights regarding the size of the fiscal bur- den of HIV/AIDS. At a discount rate of 3 percent (about the level of the real interest rate that applies to VIII. Concluding Remarks Swaziland’s public debt), the present discounted val- In many SADC member countries, the impact of and ue of the fiscal costs of HIV/AIDS amounts to about the response to HIV/AIDS have attained a level that three times the level of annual GDP (290 percent), of is significant from a macroeconomic and fiscal per- which costs equivalent to a present discounted value spective. This paper analyzes the cost of HIV/AIDS of 127 percent of GDP are incurred by 2030. This from a fiscal angle, interpreting the response to HIV/ means that the fiscal burden of HIV/AIDS is far AIDS as a long-term fiscal commitment, and broad- above the levels that would be considered unsustain- ening the scope of the analysis to identify fiscal costs able if the costs represented debt service. of HIV/AIDS (such as certain social grants) that are One of the reasons for the persistence of the fiscal not normally included in HIV/AIDS costing stud- costs of HIV/AIDS is the long period between infec- ies, but nevertheless contribute to the fiscal costs of tion and treatment need, and the long period of time HIV/AIDS. The analysis in this report should cata- in which patients receive treatment. To get a clearer lyze policy dialogue in the three countries in terms of understanding of the links between HIV incidence long-term financial sustainability of national HIV/ and the costs of HIV/AIDS, the costs of one addi- AIDS programs; exploration of innovative financ- tional infection over time are estimated, and these es- ing; program efficiency and effectiveness, including timates are used to attribute the overall fiscal costs of allocative efficiency between program components HIV/AIDS to the years in which the infections occur. of prevention, treatment and care and support; and One infection is estimated to result in fiscal costs of strengthening of private public partnerships. CONFERENCE EDITION | 19 References ASSA (Actuarial Society of South Africa). 2006. “ASSA Africa.� Centre for Economic Governance and AIDS Select Model.� Accessed May 2009, http://www.ac- in Africa, Cape Town. tuarialsociety.org.za. IMF (International Monetary Fund). 2010. World Eco- Botswana. 2010. UNGASS Country Progress Report, Re- nomic Outlook Database, April 2010 Edition. Wash- porting Period: January 2008–December 2009. www. ington, DC. unaids.org/en/KnowledgeCentre/HIVData/Coun- INLS (Instituto Nacional de Luta Contra a Sida, Angola), tryProgress/2010_UNGASS_Reporting.asp and UNAIDS ( Joint United Nations Programme on CNCS (National AIDS Control Council), and UN- HIV/AIDS). 2010. Relatório sobre o Progresso do País AIDS. 2010. UNGASS Progress Report, 2008–9, para dar Seguimento aos Compromissos da Sessão Especial Mozambique. CNCS, Maputo. sobre VIH e SIDA da Assembleia Geral das Nações Unidas, CNLS (Conseil National de Lutte contre le SIDA). 2008–09. Luanda. 2010. Rapport de situation national a l´intention de l´UNGASS, Madagascar. Période considérée: Janvier Jefferis, Keith, Happy Siphambe, and Anthony Kinghorn. 2008–Décembre 2009. Antananarivo. 2006. The Economic Impact of HIV/AIDS in Botswana. Gaborone: Econsult. Council of Economic Advisors. 2010. “Economic Re- port of the President: 2010 Report Spreadsheet Johnson, Leigh F., and Rob E. Dorrington. 2006. “Model- Tables.� Washington, DC. ling the Demographic Impact of HIV/AIDS in South CSO (Central Statistical Office), and Macro Interna- Africa and the Likely Impact of Interventions.� Demo- tional, Inc. 2008. Swaziland Demographic and Health graphic Research 14 (art. 22) 541–74. Survey 2006–7. Mbabane: Central Statistics Office. Malawi, Office of the President and Cabinet. 2010. Ma- Dorrington, Rob E., Debbie Bradshaw, Leigh Johnson, lawi HIV and AIDS Monitoring and Evaluation Report: and Debbie Budlender. 2006. “The Demographic 2008–9, UNGASS Country Progress Report, Reporting Impact of HIV/AIDS in South Africa: National Period: January 2008–December 2009. Lilongwe. Indicators for 2006.� Centre for Actuarial Research NAC (National AIDS Commission). 2010. Lesotho, UN- and South African Medical Research Council. GASS Country Report, Status of the National Re- Ellis, Linette, Pieter Laubscher, and Ben Smit. 2006. sponse to the 2001 Declaration of Commitment on HIV The Macroeconomic Impact of HIV/AIDS under Al- and AIDS, January 2008–December 2009. Maseru. ternative Intervention Scenarios (with Specific Refer- NACA (National AIDS Coordination Agency). 2008. HIV/ ence to ART) on the South African Economy. Bureau AIDS in Botswana—Estimated Trends and Implications for Economic Research, University of Stellenbosch, Based on Surveillance and Modeling. Gaborone: NACA. Stellenbosch, South Africa. NACA, and UNAIDS ( Joint United Nations Programme GoS (Government of the Kingdom of Swaziland). on HIV/AIDS). 2007. Botswana National AIDS 2009. “The National Multisectoral Strategic Frame- work for HIV and AIDS 2009–14.� Mbabane. Spending Assessment 2003/04 – 2005/06. Gaborone. Guthrie, Teresa, Nhlanhla Ndlovu, Farzana Muhib, ———. 2010. Progress Report of the National Response to Robert Hecht, and Kelsey Case. 2010. “The Long- the 2001 Declaration of Commitment on HIV and AIDS, Run Costs and Financing of HIV/AIDS in South 2008–9. Gaborone. 20 | Fiscal Dimension of HIV/AIDS in SADC Member Countries Namibia, Ministry of Health and Social Services. 2010. South Africa National Department of Health, and United Nations General Assembly Special Session (UN- SANAC (South African National AIDS Council). GASS), Country Report, Reporting Period 2008–9. 2010. Country Progress Report on the Declaration of Windhoek. Commitment on HIV/AIDS, Reporting Period: January 2008–December 2009. Pretoria. National AIDS Council, Seychelles. 2010. UNGASS Country Progress Report, Seychelles, Reporting for the Pe- South Africa National Treasury. 2010. Budget Review. riod: January 2008–December 2009. Victoria. Pretoria. National AIDS Council, Zimbabwe. 2010. United Na- Stover, John. 2008. “Estimated HIV Trends and Program tions General Assembly Special Session Report on HIV and Effects in Botswana.� PLoS ONE 3 (11). AIDS, Follow-up to the Declaration of Commitment on Tanzania, United Republic of. 2010. UNGASS Reporting HIV and AIDS, Zimbabwe Country Report, Reporting for 2010, Tanzania Mainland and Zanzibar. Dodoma. Period: January 2008–December 2009. Harare. UNAIDS ( Joint United Nations Programme on HIV/ National AIDS Secretariat, Mauritius, Prime Minister’s AIDS). 2008a. 2008 Report on the Global AIDS Epi- Office, and UNAIDS ( Joint United Nations Pro- demic. Geneva. gramme on HIV/AIDS). 2010. Republic of Mauritius, Country Progress Report 2010, United Nations General ———. 2008b. “Annual Estimates of HIV Prevalence, Assembly Special Session on HIV/AIDS Declaration of People Living with HIV/AIDS, and HIV/AIDS- Commitment, Reporting Period: January 2008–December Related Deaths, 1990–2007.� Released in conjunction 2009. Port Louis. with UNAIDS (2008a), accessed at UNAIDS Web NAC (Zambia National HIV/AIDS/STI/TB Council). site in October 2008. 2010. National AIDS Strategic Framework 2011–15, ———. 2010. UNAIDS Outlook Report 2010. Geneva. Draft 1. Lusaka. UNDP (United Nations Development Programme). 2009. NERCHA (National Emergency Response Council on Human Development Report 2009. New York: UNDP. HIV and AIDS), and UNAIDS ( Joint United Na- tions Programme on HIV/AIDS). 2008. “The King- United Nations Population Division. 2009. World Popula- dom of Swaziland National AIDS Spending Assess- tion Prospects: The 2008 Revision—Population Database. ment 2005/06 and 2006/07.� Mbabane. New York: United Nations. ———. 2010. Monitoring the Declaration of the Commit- WHO (World Health Organization). 2010. National ment on HIV and AIDS (UNGASS), Swaziland Country Health Accounts. Accessed April 25, http://www.who. Report, March 2010. Mbabane. int/nha/en/. PNMLS (Coordination Nationale du Programme Na- WHO, UNAIDS ( Joint United Nations Programme on tional Multisectoriel de Lutte Contre le VIH/SIDA). HIV/AIDS), and UNICEF (United Nations Fund 2010. Rapport National UNGASS, République Démocra- for Children). 2008. “Epidemiological Fact Sheet on tique du Congo, 2008–9. Kinshasa. HIV and AIDS—Core Data on Epidemiology and SANAC (South African National AIDS Council). 2007. Response: Swaziland (2008 Update).� Geneva: WHO. HIV & AIDS and STI Strategic Plan for South Africa World Bank. 2010. World Development Indicators 2010. 2007–11. Pretoria: SANAC. Washington, DC: World Bank. Sithole, Majozi V. 2010. “Budget Speech 2010.� Ministry Zambia Ministry of Health, and National AIDS Council. of Finance, Mbabane. 2010. Zambia Country Report, Monitoring the Declara- South Africa National Department of Health. 2009. 2008 tion of Commitment on HIV and AIDS and the Universal National Antenatal Sentinel HIV and Syphilis Prevalence Access, Biennial Report, Reporting Period: January 2008– Survey: South Africa Report. Pretoria. December 2009. Lusaka.