POLICY AND PRACTICE
Shifting the burden: the private sector's response to the AIDS epidemic in Africa
Se défaire de la charge économique, réponse du secteur privé à l'épidémie de SIDA en Afrique
La desviación de la carga como respuesta del sector privado a la epidemia de SIDA en África
Sydney RosenI; Jonathon L. SimonII
I Assistant Professor, Center for International Health and Development, Boston University School of Public Health, 715 Albany St, Boston, MA 02118, USA
II Director, Center for International Health and Development, Boston University School of Public Health, Boston, MA 02118, USA
As the economic burden of human immunodeficiency virus/acquired immunodeficiency syndrome (HIV/AIDS) increases in sub-Saharan Africa, allocation of the burden among levels and sectors of society is changing. The private sector has more scope to avoid the economic burden of AIDS than governments, households, or nongovernmental organizations, and the burden is being systematically shifted away from the private sector. Common practices that transfer the burden to households and government include pre-employment screening, reductions in employee benefits, restructured employment contracts, outsourcing of low skilled jobs, selective retrenchments, and changes in production technologies. Between 1997 and 1999 more than two-thirds of large South African employers reduced the level of health care benefits or increased employee contributions. Most firms also have replaced defined-benefit retirement funds, which expose the firm to large annual costs but provide long-term support for families, with defined-contribution funds, which eliminate risks to the firm but provide little for families of younger workers who die of AIDS. Contracting out previously permanent jobs is also shielding firms from benefit and turnover costs, effectively shifting the responsibility to care for affected workers and their families to households, nongovernmental organizations, and the government. Many of these changes are responses to globalization that would have occurred in the absence of AIDS, but they are devastating for the households of employees with HIV/AIDS. We argue that the shift in the economic burden of AIDS is a predictable response by business to which a deliberate public policy response is needed. Countries should make explicit decisions about each sector's responsibilities if a socially desirable allocation is to be achieved.
Keywords Acquired immunodeficiency syndrome/economics/prevention and control; Cost of illness; Private sector; Workplace; Personnel management; Salaries and fringe benefits; South Africa (source: MeSH, NLM).
A mesure que la charge économique de l'infection par le virus de l'immunodéficience humaine et du syndrome d'immunodéficience acquise (VIH/SIDA) augmente en Afrique subsaharienne, sa répartition au sein de la société se modifie. Le secteur privé a plus de latitude pour éviter le fardeau économique du SIDA que les gouvernements, les ménages et les organisations non gouvernementales, et de fait il se débarrasse systématiquement de cette charge. Parmi les pratiques courantes pour transférer la charge économique vers les ménages et les gouvernements figurent le dépistage à l'embauche, la réduction des indemnités versées aux employés, la refonte des contrats de travail, l'externalisation des tâches peu qualifiées, les coupes sélectives dans les budgets et les changements de technologies de production. Entre 1997 et 1999, plus des deux tiers des grands employeurs d'Afrique du Sud ont réduit leurs versements au titre de l'assurance-maladie ou ont augmenté les cotisations. La plupart des entreprises ont également remplacé les fonds de pension à rente définie, qui exposent l'entreprise à d'importants coûts annuels mais offrent un soutien durable aux familles, par des fonds à cotisation définie, ce qui élimine les risques pour l'entreprise mais profite peu aux familles de jeunes employés qui décèdent du SIDA. L'externalisation de tâches auparavant effectuées par des employés titulaires de postes fixes protège également l'entreprise contre le versement d'indemnités et les coûts de renouvellement du personnel, et permet de transférer aux ménages, aux organisations non gouvernementales et aux pouvoirs publics la responsabilité des soins aux travailleurs et à leur famille. Nombre de ces changements sont en fait des réponses à la mondialisation et auraient eu lieu même en l'absence de SIDA, mais sont catastrophiques pour les ménages d'employés atteints de VIH/ SIDA. Nous estimons que le transfert de la charge économique du SIDA est une réponse prévisible de la part des entreprises, à laquelle les pouvoirs publics doivent réagir avec détermination. Les pays doivent décider clairement quelle est la part de responsabilité de chaque secteur, afin de parvenir à une répartition socialement souhaitable de la charge.
Mots clés SIDA/économie/prévention et contrôle; Coût maladie; Secteur privé; Poste travail; Gestion personnel; Salaire et prime; Afrique du Sud (source: MeSH, INSERM).
A medida que aumenta la carga económica causada por el virus de la inmunodeficiencia humana/síndrome de inmunodeficiencia adquirida (VIH/SIDA) en el África subsahariana, cambia asimismo la distribución de esa carga en las sociedades. El sector privado tiene más margen que los gobiernos, los hogares o las organizaciones no gubernamentales para evitar la carga económica que supone el SIDA, y el resultado es que esa carga se está desviando sistemáticamente fuera del sector privado. Prácticas habituales para transferir la carga a los hogares y los poderes públicos son los análisis biológicos exigidos antes de un contrato, la reducción de las prestaciones a los empleados, la reformulación de los contratos laborales, la contratación externa de los trabajos poco cualificados, los recortes de gastos selectivos y los cambios introducidos en las tecnologías de fabricación. En Sudáfrica, entre 1997 y 1999, más de dos tercios de los grandes empleadores redujeron las prestaciones sanitarias o aumentado las cuotas correspondientes de los empleados. Además, la mayoría de las compañías han reemplazado los fondos de pensiones con prestaciones definidas, que exponen a la empresa a elevados costos anuales pero aseguran apoyo a largo plazo para las familias, por fondos con aportaciones definidas, que eliminan los riesgos para la empresa pero son de escasa ayuda para las familias de los trabajadores más jóvenes que mueren de SIDA. La contratación externa de trabajos antes realizados por personal permanente también protege a las empresas de los costos asociados a las prestaciones y de los costos de rotación, desplazando claramente hacia los hogares, las organizaciones no gubernamentales y el Estado la responsabilidad de la asistencia a los trabajadores afectados y a sus familias. Muchos de estos cambios son respuestas a la globalización que habrían ocurrido también sin el SIDA, pero tienen efectos devastadores para los hogares de los empleados afectados por el virus. En nuestra opinión, la desviación de la carga económica del SIDA es una reacción previsible de las empresas, contra la cual se requiere una respuesta deliberada de política pública. Los países han de tomar decisiones explícitas acerca de las responsabilidades de cada sector, pues sólo así se conseguirá una distribución socialmente aceptable de la carga.
Palabras clave Síndrome de inmunodeficiencia adquirida/economía/prevención y control; Costo de la enfermedad; Sector privado; Lugar de trabajo; Administración de personal; Salarios y beneficios; Sudáfrica (fuente: DeCS, BIREME).
Over the course of 2001 and 2002, a number of prominent multinational corporations announced their renewed commitment to fighting the worldwide epidemic of human immunodeficiency virus/acquired immunodeficiency syndrome (HIV/AIDS). These included some of the largest employers in sub-Saharan Africa, such as Coca-Cola and Anglo American, and some of the most visible businesses globally, such as AOL Time Warner (1, 2). The public pledges of these important and influential companies are a welcome and promising sign, and they could become an important component of the global response to the epidemic. The rush to launch "action plans against AIDS" among a handful of major multinationals, however, has tended to overshadow another important trend that is familiar to many business analysts but is not, so far, a focus of those fighting the HIV/AIDS epidemic: the shifting of the economic burden of AIDS from the private sector to governments, nongovernmental organizations, and households.
The transfer of the AIDS burden from the private sector to others manifests itself in such practices as pre-employment screening to exclude those with HIV from the workforce, smaller employee benefits, restructured employment contracts, outsourcing of low skilled jobs, selective retrenchments, and changes in production technologies that substitute capital for labour. Each of these practices reduces the share of the economic cost of HIV-positive individuals that is borne by private sector employers. Many of the changes would have come about in the absence of AIDS in response to globalization and other changes in the economic or social environment. As a result of the epidemic, however, changes in working conditions that might otherwise have had a mixture of negative and positive consequences for employees' households are becoming unambiguously harmful
The costs to business that arise from HIV/AIDS in the workforce have been documented elsewhere (3). For some firms, they are substantial, increasing labour costs by more than 6%. Wide variation exists among firms in the absolute magnitude of the costs and the relative importance of direct (out of pocket) and indirect (productivity) costs, particularly in the costs associated with retirement, death, and disability benefits (4).
Companies that decide to manage these costs have four basic strategy options. These options are not mutually exclusive, and many companies will adopt a mixture of all of them. First, they can invest in HIV prevention programmes designed to reduce the incidence of the disease in their workforces. Second, they can provide treatment, care, and social support to employees who are infected with HIV or who have AIDS, with the objective of keeping these employees in the workforce and delaying or avoiding the costs of AIDS. Third, they can invest in training new workers and broadening the skills of those already employed in order to maintain and replenish their human capital base. Finally, firms can alter their benefits policies, contract structures, and hiring practices to reduce their exposure to AIDS-related costs.
This last strategy, which has been called the "burden shift" (5, 6), is discussed in this paper. It differs fundamentally from the first three strategies, which all aim to reduce the size of the AIDS burden, because it addresses not the absolute magnitude of the burden, but its allocation among various levels and sectors of society. Our interest is in private sector actions that result in employers bearing less of the burden of AIDS, whether those actions are directed at AIDS or are taken for other reasons and affect the AIDS burden coincidentally. We hypothesize that the systematic shifting of this burden is a rational and predictable response by business and an important social and economic phenomenon to which a deliberate public policy response is needed.
The paper describes and analyses burden-shifting practices, both deliberate and inadvertent. Our data are primarily from South Africa Africa's largest economy and the country whose private sector accounts for the largest share of employment (7). After presenting anecdotal and survey evidence and a firm-level analysis of burden-shifting practices, we discuss the implications of the burden shift for businesses, governments, and households.
Evidence for the burden shift
The actions that individual firms can take to reduce or avoid the costs of AIDS among employees fall into three main categories, depending on the companies' intent. First, they can reduce the total number of permanent employees in the workforce, by substituting capital for labour (mechanizing) or increasing the proportion of non-permanent workers at the expense of permanent jobs. Second, they can aim to reduce the number of HIV-positive individuals in the workforce. This type of response takes several forms, including pre-employment screening, altering terms of employment contracts, carrying out selective retrenchments or non-voluntary medical retirements, hiring expatriates to fill senior or highly skilled positions, and in its most extreme form relocating to a country with a lower prevalence of HIV. Third, a company can reduce the cost it bears per HIV-employee, by taking steps such as cutting the levels of retirement, death, health, or sick leave benefits; capping the employer's contribution to benefit premiums; or even relocating to a country that places fewer obligations on employers of HIV-positive workers.
Anecdotal and survey evidence
Anecdotal evidence of burden-shifting practices abounds. In Zimbabwe in 1997, widespread evidence showed illegal pre-employment HIV testing of job applicants by firms and screening of applicants to avoid hiring those with risky lifestyles (8). Recent interviews with several manufacturing firms in Nigeria revealed a number of practices aimed at protecting firms from the burden of AIDS, including covert pre-employment and in-service testing and exclusion of HIV/AIDS-related conditions from medical benefits (9). For example, one firm classified infection with HIV as a "self-inflicted condition" and therefore refused to cover it. A textile firm that routinely tests workers for tuberculosis as part of its occupational safety programme used this opportunity to carry out HIV tests without the workers' knowledge; employees who tested positive for HIV were dismissed but were not told of their infection status. A company in Botswana reduced the number of days of sick leave that employees are allowed to accrue and adopted a policy that requires anyone with a negative sick leave balance to accept medical retirement (10). The in-house health insurance provider of one large South African employer reduced its ceiling for HIV-related claims from R 100 000 per family (US$ 20 408) in 1997 to R 15 000 (US$ 2419) per family in 1999 (10).
More rigorous data that would allow us to quantify the burden shift are harder to find, but a handful of surveys are available. In 1999, Old Mutual a large South African financial services firm asked 15 large, defined-contribution retirement funds if and how they were responding to the rising cost of death and disability insurance. Almost half the funds reported that they are taking steps to limit the company's share of the AIDS burden, such as decreasing death and disability benefits (40%), capping employer contributions (48%), or requiring employees to pay a larger share of premiums for the same benefits (48%) (11).
Shifting employees from defined-benefit to defined-contribution retirement funds has been one of the most common and effective ways for firms to avoid some of the costs of HIV/AIDS (12, 13). Defined-benefit pension funds provide a fixed lifetime annuity to the spouse left behind by an employee who died of AIDS regardless of how many years the employee has worked at the company or of the employee's age at death. Defined-contribution provident funds make a one-off payment of the sum of the employee's contributions and employer contributions up to the last day of employment. The beneficiaries of younger employees with AIDS thus receive only a single payment, which usually is small because of the small number of years worked. A survey of approximately 800 retirement funds in 2000 carried out by Sanlam another South African financial services firm found that 71% of funds were defined-contribution funds compared with just 26% in 1992 (14, 15).
For medical benefits, two similar data sets are available. Old Mutual surveyed a random sample of 56 large South African employers, stratified by size and location, in 1999. Of the firms surveyed, 44 (78%) reported having restructured their health care benefits in the previous two years mainly by shifting more of the cost onto the employees, capping company contributions, reducing benefit levels, or a combination of these. An average of 36% of employees with access to company-sponsored medical-aid schemes had opted out entirely primarily because of cost (16). The Johannesburg Chamber of Commerce and Industry surveyed 1500 of its members in 2000 (17): some 600 (40%) had changed to lower premium medical-aid schemes that provided fewer benefits, while the proportion of staff participating in medical-aid schemes had declined in 53% of the surveyed companies.
Finally, quite a few of the burden-shifting practices related to contract conditions were documented by a World Bank survey of 325 large manufacturing firms in the greater Johannesburg metropolitan area in 1999. When asked about steps taken in response to labour legislation enacted between 1995 and 1998, about 40% of firms reported that they had chosen to hire fewer workers, use more temporary workers, use more machinery, or a combination of the three. About one- third also reported using more subcontractors in response to the new laws (18).
Although none of the surveys summarized above was of a nationally representative sample, all attest to a pervasive decrease in the level of retirement, death, disability, and medical benefits being provided to employees. Beyond this, South African firms increasingly are outsourcing their non-core service jobs, and even some core production jobs, to independent companies whose function it is to provide workers. The mining and agribusiness sectors among the country's most important exporters are very large contractors of independent labourers, who provide the services of full-time employees but receive few of the benefits.
To put some hard numbers on these trends, we analysed the benefits policies and employment contract structures of two large companies in KwaZulu Natal the province of South Africa that has been hardest hit by the AIDS epidemic. Table 1 gives a snapshot of conditions for four classes of less skilled workers: permanent employees, fixed-term contractors, casual (day) labourers, and employees of an outsourcing firm that provides cleaning services to other companies. Both firms report that they either have in the recent past or intend in the near future to reduce the number of permanent unskilled employees in favour of more casual or contract workers.
In general, the retirement, death, and disability benefits provided to permanent employees exceed by several times those offered to non-permanent workers. Salaries are also considerably higher for permanent employees. For workers with AIDS, this translates into permanent employees having much larger, although usually still inadequate, financial resources for their own care and their families' future welfare than non-permanent workers. If a less skilled permanent employee of Company 2 (see Table 1), for example, dies in service, his beneficiaries receive death benefits of about R 120 000 plus the amount accrued in the employee's provident fund; the beneficiaries of a casual worker at the same company receive nothing.
None of the types of employees included in Table 1 has ready access to medical care beyond that provided by on-site, adult, first-level care clinics and the public medical system. The non-permanent workers are offered no medical-aid cover at all. Permanent employees can opt for medical aid, and both firms contribute a substantial share of the premiums. The employee copayment remains high enough, however, to preclude most low-paid staff from joining.
Where will the burden go?
When an employer-subsidized health insurance plan caps benefits for patients infected with HIV at far less than the costs of the treatment needed, employees with HIV must pay for their own treatment; rely on services provided by the government, religious organizations, or other nongovernmental organizations; or forgo treatment. When an employer reduces its death or retirement benefits or hires non-permanent workers who are not eligible for such benefits, the families left behind by those who die of AIDS must find other sources of support, such as social insurance for the few who have access to it, or fall back on their own resources. (To some extent, all households even those without an AIDS death will bear a greater burden, because AIDS-related claims will drain the resources from retirement funds and reduce the benefits available to everyone (13).)
Governments and nongovernmental organizations will meet some of the demand for services created by the epidemic, but in the end, households and extended families will bear the brunt of the costs. Health care facilities funded by government and nongovernmental organizations have already been overwhelmed by HIV/AIDS patients, who accounted for 54% of adult hospital inpatients and 62.5% of child hospital inpatients at a major hospital in Durban, South Africa, in 2001, for example (19, 20). A small but growing evidence base shows the impact on households of the loss of an adult member to AIDS (21, 22). In rural Kenya, for example, the loss of a male household head was associated with a 68% decline in the net value of agricultural output, implying a substantial blow to the family's economic welfare (23).
Although in some cases government facilities are accepting the burden of AIDS, in others the government itself is shifting the burden onto households or back onto the private sector. In Nigeria, the National Health Insurance Scheme specifically excludes HIV/AIDS on the basis that the cost of AIDS treatment would bankrupt the scheme (24). The recently enacted Medical Schemes Act is forcing South African firms to pay for medical care for a larger proportion of workers than they have before (25). Facing many of the same financial constraints as businesses, governments are likely to pursue some of the same strategies used by businesses, including outright avoidance of the AIDS burden.
Is AIDS the reason for the observed trends?
Many African companies, particularly those in South Africa, would have undertaken parts of the cost avoidance strategy even in the absence of HIV/AIDS. The second half of the 1990s brought to South Africa a difficult combination of rising labour costs because of new labour legislation, affirmative action goals that led to high rates of employee turnover, high inflation in health care costs, and exposure, for the first time, to competitive global markets. All of these factors are encouraging companies to restructure their workforces, reduce production costs, limit employee benefits, and shift to more capital-intensive production technologies the same strategy being used to protect against the costs of the HIV/AIDS crisis.
The coincidence of the epidemic and changes in the social and economic environment makes it difficult to ascertain the true cause of many business decisions. One very large South African company, for example, dissolved its shipping department and established its truck drivers as independent "ownerdrivers," on the stated premise of supporting the formation of a black entrepreneurial class (26). Creating independent businesses owned by black Africans is indeed a priority of the South African government and is regarded as a "social investment." On the other hand, the company no longer has a responsibility to provide any benefits to drivers, although its business will suffer if its distribution network is disrupted by high morbidity and mortality among drivers. The shift from defined-benefit to defined-contribution pension funds also took place largely in response to social forces other than HIV/AIDS in this case, pressure from labour unions in the years before the AIDS epidemic, which were based on unions' perceptions, at the time, that defined-contribution funds were in their members' interest (6).
In these cases and others, the changes that caused the burden of AIDS to shift elsewhere would probably have been undertaken even in the absence of AIDS. When the types of burden-shifting practices described earlier are categorized by cause (Table 2), the synergies between AIDS and globalization readily are apparent
It is clear from Table 2 that AIDS and globalization mutually reinforce many burden-shifting practices. The AIDS epidemic accelerates some changes that would have taken place anyway, such as mechanization, while globalization facilitates some practices that begin as direct responses to AIDS, such as capping of benefits contributions. Often it is impossible to determine which of the two causes came first. The pressures of a global economy are forcing businesses to become ever leaner and, in many cases, ever meaner. In a world without AIDS, the changes to employment conditions would, like other aspects of globalization, have had both positive and negative consequences for businesses and workers. When the AIDS epidemic is added to the mix, the unintended impact on workers and their families may be ruinous.
Should we be surprised?
The costs that businesses are avoiding through the types of practices described above are substantial. The Metropolitan AIDS Research Unit, which represents one of the largest insurance companies in South Africa, has warned repeatedly, for example, that without active intervention, AIDS will double the average cost of employee benefits by 2005 and will treble them by 2010 adding 15% to the average wage bill (27). Irrespective of whether their own costs have started to rise, businesses have reason to be worried.
The capping of costs borne by companies, and thereby the transfer of those costs to government, households, and, to a lesser extent, other companies is a rational response by profit-maximizing businesses, and it should be expected. Of all those who are affected by the HIV/AIDS epidemic, private firms have the most scope to contain and avoid the costs. Companies, and to some extent governments, will avoid costs because they can; households will bear those costs because, in most cases, they cannot avoid them (5). (Note that this applies only to some of the costs of HIV/AIDS. It is not the case for the market impacts of the epidemic, such as increasing wage rates and falling demand for companies' products.)
Governments around the world can and do constrain the actions of private companies through regulations. If governments demand too much of the private sector, however, companies might fail, relocate to lower-cost countries, or hasten the transition to capital-intensive technologies that require fewer unskilled employees. Private sector bankruptcies, relocations, and retrenchments are an undesirable outcome for everyone: governments lose tax revenue, employees lose jobs, and communities lose investment and commercial activity. To a lesser extent, policies that force medical-aid schemes and retirement funds into deficit will diminish the welfare of vast numbers of employees and the families who rely on these benefits. A recent public opinion survey in 12 sub-Saharan countries found that job creation, poverty alleviation, and economic development were people's highest priorities for government action in almost every country, ranking far ahead of AIDS and serving as a reminder of the importance of private sector growth (28).
The private sector clearly has an important role to play in preventing HIV infection among employees and financing care for those who are infected, but it appears inevitable that primary responsibility for prevention of HIV and care of infected people will continue to fall on governments, nongovernmental organizations, and households. The potential contribution of the private sector should not be neglected, but it should not be overestimated either.
What should be done?
The HIV/AIDS epidemic has confronted decision-makers in sub-Saharan Africa with a public policy optimization problem. They must do everything in their power to foster economic growth and retain and create jobs that their populations urgently need and want. At the same time, they must induce the private sector to do as much as it can to fight the AIDS epidemic and care for those affected. If governments push too hard on the latter, they risk losing ground on the former, as businesses respond to the higher cost of labour by substituting technology for labour, outsourcing jobs, or relocating. The burden shift thus is another instance of the fundamental tension between private profit and the public good a tension played out in domains of public policy that range from environmental protection to regulation of financial markets.
Given the importance of developing realistic national strategies to manage the epidemic and the discrepancy between public pledges of action against AIDS and private measures that shift the burden of AIDS, we see a need for action at three levels. First, each country must decide how it wants the burden of HIV/AIDS to be allocated. The burden is huge, and in the end the largest share will almost inevitably fall on individuals and households. The private sector has a clear incentive, and some ability, to shift the burden unless governments take action to prevent it. Deliberate decisions on social policy must be made, and enforced, if the ultimate allocation of the burden is to be socially desirable. These decisions are likely to differ widely among countries, based on differences in their economies, forms of government, regulatory capacities, and burden of AIDS. The decisions ultimately will reflect each country's own solution to the optimization problem defined above: balancing economic growth and employment with business investment in the fight against AIDS.
Second, researchers and international organizations should begin to develop a set of strategies and tools that help countries achieve the balance they desire. This effort can draw on extensive experience in other fields to regulate business practices and balance private sector interests with the public good. Such a balance is difficult to achieve but certainly not impossible. Successful examples include environmental regulations that phased out leaded gasoline in a way that protected children's health, while minimizing the cost to business; and antitrust legislation that ensures competitive pricing for consumers, while guaranteeing a level playing field for businesses.
Finally, the trend we have described in this paper should be monitored and analysed. We have presented a hypothesis, with some preliminary supporting evidence drawn largely from a single country in the region. Before policy-makers can develop response strategies, they need a better understanding of baseline conditions. Systematic data collection and ongoing monitoring of levels of benefits, hiring practices, and employment structures are needed to understand the nature and magnitude of the trend, determine where and for what types of industries or employers it is most important, and evaluate the impacts of policy changes. Using experience from other fields and other countries, as well as information generated by monitoring and analysis, we can encourage both governments and businesses to recognize and bear their fair share of the burden and to do their best to support the households who will bear the rest.
Funding for the research presented in this paper was provided by the South Africa Mission of the US Agency for International Development (USAID) through the Child Health Research Project, G/PHN/HN/CS, Global Bureau, USAID, under the terms of Cooperative Agreement No. HRN-A-00-96-90010-00, the Applied Research on Child Health (ARCH) Project. The opinions expressed are those of the authors and do not necessarily reflect the views of USAID. We thank Alan Whiteside and the AIDS Economics Team of the Center for International Health and Development for their participation in discussions about the ideas developed in this paper.
Conflicts of interest: none declared.
1. Top businesses pledge to act on HIV/AIDS. New York: Global Business Council on HIV/AIDS; 26 June 2001.
2. GBC president, Richard Holbrooke, welcomes mines' leadership on AIDS, calls on other multinational corporations to act. New York: Global Business Council on HIV/AIDS; 30 August 2002.
3. Rosen SB, MacLeod WB, Vincent JR, Fox MP, Thea DM, Simon JL. Investing in the epidemic: the impact of AIDS on businesses in southern Africa. XIV International AIDS Conference, Barcelona, 2002, July 7-12: Abstract ThOrG1505.
4. AIDSCAP Project. African Workplace Profiles. Washington (DC): Family Health International; 1995.
5. Simon JL, Rosen S, Whiteside A, Vincent JR, Thea DM. The response of African businesses to HIV/AIDS. In: Commonwealth Secretariat, editors. HIV/AIDS in the Commonwealth 2000/01. London: Kensington Publications; 2000.
6. Van den Heever A. HIV/AIDS and employee benefits. AIDS Analysis Africa 1998;9:3-4.
7. Blunch N-H, Canagarajah S, Raju D. The informal sector revisited: policy challenges for poverty reduction. Washington (DC): World Bank; 2001.
8. Collins J. Leveraging private sector support for HIV/AIDS prevention: opportunities and obstacles. A report on Zimbabwe and Brazil. Washington (DC): Family Health International; 1997.
9. Center for International Health. The implications of HIV/AIDS for Nigerian manufacturing firms. Results of the HIV/AIDS component of the RPED Nigeria survey, March-April 2001, World Bank Africa Region. Boston (MA): Boston University School of Public Health; 2001.
10. Center for International Health and Development. The cost of HIV/AIDS to businesses in Africa (forthcoming).
11. Will your retirement fund survive AIDS? Johannesburg: Old Mutual Trustee Times; February 2000.
12. Sidley P. Middle-aged workforce to carry rising AIDS costs. Johannesburg: Business Day; 14 April 1999.
13. Death and disability: funds face critical issues. Johannesburg: Old Mutual Trustee Times; February 2002.
14. Sanlam's new system targets retirement funds. Johannesburg: Business Day; 22 March 2001.
15. Trends in the retirement industry. Johannesburg: Business Times; 18 October 1998.
16. The 1999 Old Mutual Healthcare Survey. Johannesburg: Old Mutual; 1999.
17. Shevel A. Profits suffering from medical aid costs. Johannesburg: Business Report; June 15 2000.
18. Chandra V, Moorty L, Rajaratnam B, Schaefer K. Constraints to growth and employment in South Africa. Report no. 1: statistics from the Large Manufacturing Firm Survey. Informal Discussion Papers on Aspects of the Economy of South Africa No. 14. Washington (DC): The World Bank Southern Africa Department; 2001.
19. Colvin M, Dawood S, Kleinschmidt I, Mullick S, Lallo U. Prevalence of HIV and HIV-related diseases in the adult medical wards of a tertiary hospital in Durban, South Africa. International Journal of STD and AIDS 2001;12:386-9.
20. Williams BG, Gouws E, Colvin M, Sitas F, Ramjee G, Abdool Karim SS, et al. Patterns of infections: using age prevalence data to understand the epidemic of HIV in South Africa. South African Journal of Science 2000;96:305-12.
21. Barnett T, Whiteside A. AIDS in the twenty-first century: disease and globalization. New York: Palgrave MacMillan; 2002.
22. The impact of AIDS on households. In: Ainsworth M, Fransen L, Over M, editors. Confronting AIDS: evidence from the developing world. Selected background papers for the World Bank Policy Research Report Confronting AIDS: public priorities in a global epidemic. Brussels: European Commission; 1997.
23. Yamano T, Jayne TS. Measuring the impacts of prime-age adult death on rural households in Kenya. Staff Paper Series 2002-26, Department of Agricultural Economics. East Lansing (MI) Michigan State University; 2002.
24. Okonkwo A. Nigeria set to launch health insurance scheme. Lancet 2001;358:131.
25. Gill S. Medical aid revolution. Johannesburg: Daily Mail and Guardian; 1 February 2000.
26. Tawney C, Levitsky J. Small enterprise development as a strategy for reducing the social cost of restructuring and the privatization process: public and private initiatives. Interdepartmental Action Programme on Privatization, Restructuring and Economic Democracy Working Paper IPPRED-6. Geneva: International Labour Office; 1997.
27. Van Niekerk M. AIDS could add 15% to salary bill. Johannesburg: Business Day; 21 September 2000.
28. Afrobarometer. Key findings about public opinion in Africa. Afrobarometer Briefing Paper No. 1. Cape Town: Afrobarometer, 2002. Available from: URL: http://www.afrobarometer.org
Address to correspondence