Botswana: Selected Issues and Statistical Appendix

This Selected Issues paper and Statistical Appendix provides medium-term estimates of the cost of the government’s current anti-HTV/AIDS policies and programs in Botswana. The paper throws light on the policy challenges that Botswana authorities face in combating the crisis. The findings suggest that the cost of treating HIV/AIDS patients is likely to be high, about 10 percent of GDP, by 2010 (using the baseline estimates). The paper also analyzes Botswana’s approach to medium-term fiscal management.


This Selected Issues paper and Statistical Appendix provides medium-term estimates of the cost of the government’s current anti-HTV/AIDS policies and programs in Botswana. The paper throws light on the policy challenges that Botswana authorities face in combating the crisis. The findings suggest that the cost of treating HIV/AIDS patients is likely to be high, about 10 percent of GDP, by 2010 (using the baseline estimates). The paper also analyzes Botswana’s approach to medium-term fiscal management.

I. Fiscal Implications of HIV/AIDS Treatment in Botswana1

A. Introduction

1. Botswana faces the highest HIV/AIDS incidence rate in the world. According to the July 2002 UNAIDS Report (UNAIDS, 2002), the number of adults (15-49 age group) infected with the virus increased from an estimated 33.2 percent in 1996 to 38.8 percent in 2001. The social and economic consequences include an overburdened health system, a deteriorating public education system, and a threat to the hitherto impressive improvements in the standards of living of Botswana people. The country has slipped down the human development index ranking of 173 nations, falling from a high of 71 in 1996 to 122 in 1999 and 2000.2 Life expectancy in particular has fallen from 67 years in 1996 to 47 years in 1999, and is projected to drop to 29 years in 2010.3

2. Over the last several years, the government has responded with well-targeted interventions designed to combat the spread of the disease and provide needed assistance to victims and their families. Budget allocations to the health sector have seen large increases in the last two years, and is projected to rise by 50 percent in the current fiscal-year, 2002/2003. An aggressive education and public enlightenment campaign has been underway for several years, and an antiretroviral therapy has begun on a pilot basis in the major cities. The government’s effort to draw on support from the domestic private sector, as well as the international donor community, has also yielded positive results. Against this backdrop, the aim of the present study is to provide medium-term estimates of the cost of government’s current anti-HIV/AIDS policies and programs, and thereby throw light on the policy challenges that Botswana authorities face in combating the crisis.

B. Background

3. The extensive and growing literature on the economic and social impact of HIV/AIDS has so far not given detailed attention to the long-term fiscal implications of government efforts to combat the crisis. A staff study on the macroeconomic impact of HIV/AIDS in Botswana in the 2001 Selected Issues and Statistical Appendix (SM/01/72) touched on this issue only briefly. That study assumed that health-related spending due to AIDS would rise by 1.6–2 percentage points of GDP over ten years under moderately-priced treatment options. Far more important was the paper’s finding on the likely impact of HIV/AIDS on economic growth. The paper indicated that as result of the negative impact of HIV/AIDS on growth, the Botswana economy could be 33–40 percent smaller in size in 2010 than it would have been in the absence of HIV/AIDS. A similar study by the Botswana Institute of Policy Analysis (BIDPA 2000), suggested that the impact on the recurrent health budget could be very large. In particular, the study suggested that such a health spending could quadruple (to about 10 percent of GDP) after ten years.

C. Initiatives to Fight the HIV/AIDS Pandemic

4. The government has accorded the pandemic the highest priority, and set an ambitious target for combating the crisis. The stated goal is an “AIDS-free generation of Botswana” by 2016 (NACA 2001). To provide the coordination necessary for the effective mobilization of support from all stakeholders and the international community, a national AIDS coordinating agency (NACA) has been established in the office of the President.

5. International support for the anti-AIDS campaign has taken various forms. A joint government-private sector organization, the Botswana Comprehensive HIV/AIDS Partnership, involving U.S. drug companies, namely, Merck & Co., and Bristol-Myers Squibb Co., as well as the Gates Foundation, together are to contribute a combined sum of US$ 100 million over the next five years. Merck, in addition, has undertaken to donate antiretroviral medicines for the treatment of infected patients. To help build capacity, the U.S. government’s Center for Disease Control and Prevention has sent a number of professionals to assist the government’s effort. Harvard University is assisting the government’s HIV reference laboratory, and a number of universities and hospitals in Europe and the United States are providing other forms of assistance.

6. The domestic private sector is contributing to the prevention, treatment, and family support programs of the government. The most extensive of these private sector initiatives is by the mining company Debswana, which is partly government-owned. Debswana operates two hospitals for its approximately 6,200 workers and pays 90 percent of the cost of antiretroviral treatment for its employees who are not covered by a medical insurance plan, including wife or husband (Economist, 2002). Also, the procurement policy of the company requires that all its suppliers have an HIV/AIDS policy.

D. Current Programs and Cost Estimates

7. Botswana’s comprehensive HIV/AIDS program aims to achieve by year 2010: (i) a 50 percent reduction of new infections; (ii) a 95 percent rate of access to basic, intermediate, and advanced care; and (iii) a 50 percent reduction of the impact of HIV/AIDS on the population. The programs currently being undertaken can be grouped under two broad areas: prevention, and treatment, care and support (see Box).

Anti-HIV/AIDS Programs

Prevention activities include the following;

  • an Information, Education, and Communication (IEC) program targeted at the general population, youth with a special emphasis on life skills in school curricula, a teacher capacity-building project, radio drama, and workplace programs, including counseling;

  • a condom promotion through social marketing and free distribution through health facilities;

  • a prevention of mother-to-child transmission project (PMTCT):

  • treatment of sexually transmitted diseases;

  • vaccine trials through the Botswana Harvard AIDS Partnership;

  • a total community mobilization (TCM) project that links clients to prevention, treatment, care, and support programs and also provides interpersonal communication for clarification of any myths and misconceptions; and

  • a voluntary counseling and testing program, which also links clients to prevention and other programs.

Treatment, care and support activities comprise the following:

  • prevention of TB;

  • treatment of opportunistic infections;

  • an antiretroviral therapy program;

  • community home-based care for the terminally ill; and

  • orphan care and support.

8. The 2002/03 budget provides P 326.4 million (about 1 percent of GDP) for identified HIV/AIDS spending, including P 28.5 million for antiretroviral treatment (Table I.1). A significant upward revision in the resources allocated to AIDS programs is, therefore, likely over the coming years as the programs become fully operational and the number of patients enrolling for treatment rises.

Table I.1.

Botswana: HIV/AIDS Programs, 1999/2000 - 2002/03

(In thousands of pula)

article image
Source: Botswana authorities.

E. Deriving a Cost Estimate of HIV/AIDS Treatment

Assumptions underlying estimates

9. A key uncertainty in estimating future budget implications of HIV/AIDS is the cost of advanced antiretroviral therapy. Uncertainties arise from the cost of the drugs themselves and from the number of patients who will receive them.

10. To simplify the analysis, we assume that ARV drug costs remain unchanged in current price terms. Thus, the cost to the government would be P600 per patient per month. This estimate includes the provision of some HIV medications by their manufacturers at reduced or no cost. Capital costs are assumed to rise sharply and in rough proportion to the number of new patients entering treatment.

11. Estimates of the number of prospective AIDS patients are based on demographic and epidemiological projections by the International Programs Center of the U.S. Bureau of the Census. These projections of population and death due to AIDS (without antiretroviral therapy) are derived by fitting a demographic and epidemiological model to the available observations on HIV prevalence, mainly from blood tests at antenatal clinics. The number of patients receiving antiretroviral therapy are then projected using AIDS-related mortality rates (in the absence of treatment). More specifically, the number of patients in antiretroviral (ARV) treatment programs is based on the assumption that patients would, on average, reach the threshold “virus load” at which treatment starts, two years before they would die in the absence of treatment, and that antiretroviral therapy would prolong their lives by an average of five years.

12. Two scenarios are considered; (i) The baseline scenario assumes that advanced antiretroviral drugs are gradually introduced, as related infrastructure is put in place; and (ii) a fast-implementation scenario that assumes the government is able to achieve a 95 percent treatment rate of infected patients much sooner, by 2003. Thus, the only difference in assumptions is the coverage rate, with the number of HIV-positive patients falling sick remaining the same in both cases.

The results: gradual expansion—versus fast implementation—of HIV/AIDS programs

13. Under the baseline scenario, the cost of providing advanced treatment would build over time (Table I.2). Costs would increase to 3¾ percent of GDP by 2008, the end of the NDP-9 planning cycle, and exceed 5 percent of GDP by 2010. The number of patients receiving ARV treatment would rise from fewer than 10,000 in 2002 to over 225,000 by 2010 and would therefore prolong the lives of many of those infected. The number of AIDS deaths would rise through 2004, before declining. However, AIDS deaths would rise at the period because those who began treatment in the early phase of the program would begin to die. This pattern follows from the assumption that ARV treatment prolongs life for five years.

Table I.2.

Botswana: Estimated Caste of HIV/AIDS Treatment, 2001-2010

article image

Based on projections by internal programs center at the US bureau of census.

14. Our results are consistent with those of the 2000 BIDPA study, which found that, if Botswana provided generous treatment for HIV/AIDS patients in line with international standards, the cost of this treatment could amount to 5-6 percent of GDP by 2011.

15. The fast-implementation scenario assumes that advanced ARV drugs are provided to 95 percent of those infected and needing treatment by 2003. The number of patients receiving treatment rises sharply from under 10,000 in 2002 to over 48,600 in 2003 and to 300,000 by the year 2010. AIDS deaths would fall sharply in the near term as treatment extended the life expectancy of those with high viral loads from two to seven years. Deaths, however, would rise sharply in 2010 as those who began treatment in 2003 died. Costs of the treatment are estimated to reach 6 percent of GDP by 2008 and 6½ percent by 2010, resulting in cumulative total costs of P 12 billion over the 2003-10 period.

16. The two scenarios provide an estimate of the range of costs of providing ARV therapy. The baseline scenario assumes a measured pace of implementation not because this is a desired policy, but as an illustration of a possible outcome if time is required to expand the current pilot program. It would seem reasonable to assume that actual costs will depend on the availability of hospitals, laboratories, and skilled health professionals. These resources will take time to develop.

17. On the other hand, it would be desirable to provide ARV therapy to all those in need as quickly as possible. Putting aside the question of feasibility, the fast-implementation scenario indicates that costs would be higher than in the slower implementation baseline case, but they could be financed utilizing existing resources, including continued support from the donor community.

18. The provision of ARV therapy is only one part of the government’s action plan. When other health-related expenses are included, HIV/AIDS spending is estimated to expand from 8 percent of total health expenditure to just under 70 percent in 2010 (Table I.3). Its share of total central government spending would rise from under 1 percent to 14 percent.

Table I.3.

Botswana: Central Government AIDS Expenditure, 1999 - 2010 1/

(Baseline Scenario)

article image
Source: Botswana authorities and staff estimates and projections.

Fiscal year, March-April. AIDS expenditure includes provision of ARV, and all other AIDS programs.

Constant 2001 prices.

Sensitivity analysis

19. The cost estimates in both scenarios are sensitive to a few assumptions in addition to those described above. For example, both scenarios assume that all eligible patients will seek out ARV therapy. This may not be the case if individuals are reluctant to be tested for the HIV virus or identify themselves to family and co-workers as HIV-positive by undertaking the treatment regime. Another variable is private sector participation. Both scenarios assume that the full burden of treatment is borne by the public sector. Costs to the government would be reduced to the extent that private firms take over treatment for their own employees. The diamond mining company, Debswana, has already begun providing HIV/AIDS treatment to its employees and their spouses. The fiscal impact of this action, however, is limited because Debswana is partly government owned.

F. Conclusions

20. Estimates of the impact of HIV/AIDS on Botswana’s government finances are based on many assumptions and simplifications. Nevertheless, the cost of treating HIV/AIDS patients is likely to be very high, around 10 percent of GDP, by 2010 (using the baseline estimates). And, estimates presented above do not include indirect costs such as increased sick leave. In view of the potential magnitude involved, it would seem prudent to study further the budget implications, perhaps in the context of NDP-9.


  • Bank of Botswana, 2001, Annual Report (Gaborone, Botswana: Bank of Botswana).

  • BIDPA 2000, Macroeconomic Impact of the HIV/AIDS Pandemic in Botswana, Final Report (Gaborone, Botswana: Botswana Institute for Development Policy Analysis).

    • Search Google Scholar
    • Export Citation
  • Economist, 2002, “Fighting Back”, May 11th–17th, page 25.

  • Haacker, Markus, 2001, “Providing Health Care to HIV Patients in Southern Africa”, IMF Working Paper 01/03 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Haacker, Markus, “The Economic Consequences of HIV/AIDS in Southern Africa”, IMF Working Paper 02/38 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • NACA, 2001, Botswana 2001 HIV Sero–Prevalence Sentinel Survey Amongst Pregnant Women and Men with Sexually Transmitted Diseases (Gaborone, Botswana: National AIDS Coordinating Agency).

    • Search Google Scholar
    • Export Citation
  • Rosen, S., and J.L., Simon, 2002, “Shifting the Burden of HIV/AIDS” (unpublished; Boston, Massachusetts: Center for International Health, Boston University School of Public Health).

    • Search Google Scholar
    • Export Citation
  • UNAIDS 2002, Report on the Global HIV/AIDS Epidemic (Geneva, Switzerland: Joint UN Program on HIV/AIDS).

  • UNDP, 2000, Botswana Human Development Report, 2000 (New York: United Nations Development Program).

  • USAID,2001, “The HIV/AIDS Crisis: How Are Finance and Planning Ministries Responding?”, paper presented by the U.S. Agency for International Development at the plenary session on HIV/AIDS, African Growth and Opportunity Act (AGOA) Forum, Washington, October 31.

    • Search Google Scholar
    • Export Citation
  • World Bank, 2001, “Botswana: Selected Development Impact of HIV/AIDS” World Bank Macroeconomic Technical Group, Africa Region, Report No. 22043-BT (Washington: World Bank).

    • Search Google Scholar
    • Export Citation

Prepared by Mzwanele Mfunwa.


Same source as above.