Concerns about the macroeconomic consequences of the human immunodeficiency virus, and the associated acquired immunodeficiency syndrome (HIV/AIDS) have been fueled by several factors. Most obviously, the epidemic has a devastating impact on life expectancy in a number of countries. In the empirical literature on economic growth (not dealing specifically with HIV/AIDS), such a decline is associated with a steep drop in the growth of gross domestic product (GDP). More informally, there are concerns that the epidemic could affect long-term development aspects by destroying human capital and the incentives to invest in education, disrupt the social fabric of a society, and result in an increasing number of disadvantaged young people (mainly orphans).
Second, there have been concerns that the impact of the epidemic is tied up with and exacerbates the challenges of economic development. For example, the 2006 Political Declaration issued by the United Nations (UN) states ‘‘that in many parts of the world, the spread of HIV/AIDS is a cause and consequence of poverty, and that effectively combating HIV/AIDS is essential to the achievement of internationally agreed development goals and objectives.’’
Third, the response to HIV/AIDS in many countries has become a macroeconomic factor in its own right, not only because it partially reverses the adverse direct consequences of the epidemic but also because of the additional demand for (health) services, and the challenges of financing HIV programs.
Against this background, the article focuses on three areas. It sets out with a discussion of the state of the epidemic across countries and its correlation with the state of economic development. This is followed by a discussion of the literature and evidence on the macroeconomic impacts of HIV/AIDS. Finally, the article highlights macroeconomic aspects of the financing of HIV programs, including the role external assistance has played in this.
HIV/AIDS And The State Of Economic Development
The macroeconomic effects of HIV/AIDS depend on the economic context, as well as the state of the epidemic. For example, the impact of HIV/AIDS on affected households depends on available health services and the availability of health and social insurance, companies with high value added per employee have higher stakes in investments to minimize the impact of HIV/AIDS on their staff and operations, and the government’s capabilities in meeting the demand for HIV/ AIDS-related services are constrained by its fiscal resources.
Moreover, the state of the epidemic is partly endogenous, and the quality of the policy response to the epidemic in turn reflects the quality of a country’s institutions and its economic and public policy capacities. From the perspective of global development policy, where HIV/AIDS competes with other causes for external assistance, it is also useful to place the epidemic in an economic context.
HIV/AIDS-related deaths are concentrated in low-income countries, similar to infectious diseases more generally. According to the ‘Causes of Death 2008’ data published by the World Health Organisation, 41% of HIV/AIDS-related deaths and 36% of deaths from infectious diseases occurred in low-income countries in 2008 (which accounted for 12% of the global population). In terms of its association with economic development challenges, HIV/AIDS thus resembles infectious diseases in general, but it is not correlated as closely with basic economic development challenges as malaria deaths are, of which 58% occurred in low-income countries.
However, HIV/AIDS mortality has been declining, reflecting increased access to treatment. According to the data from the 2012 report on the Global AIDS Epidemic by the Joint United Nations Programme on HIV/AIDS (UNAIDS), 542,000 AIDS deaths (32% of global AIDS deaths) occurred in low-income countries in 2011.
The broad distribution of HIV deaths by income group, however, gives a misleading picture of the challenges posed by HIV/AIDS, as HIV/AIDS is distributed across countries very unevenly. Taking, for example, the global distribution of income (Gini coefficient: 0.64) as a References: point, the burden of HIV/AIDS is distributed much more unevenly (Gini coefficient: 0.74). This point is illustrated in Figure 1, which orders the global population by GDP per capita and adds a curve describing HIV prevalence in the respective countries. Indeed, HIV prevalence tends to be higher in countries with lower income. This is evident from the negative coefficient of correlation between HIV prevalence and GDP per capita ( 0.09), substantial HIV prevalence in a number of low-income countries (broadly, those to the left of the 700-million population mark in Figure 1) and an absence of high HIV prevalence among high-income countries (broadly, the rightmost billion in Figure 1). The most striking feature of the distribution of people living with HIV, however, is the high concentration of HIV/AIDS in a few countries with HIV prevalence over 10% of the total population. In this regard, it also differs from malaria, which correlated more strongly with the state of economic development in general (with higher prevalence in low-income countries) but is less concentrated in specific countries.
Although the correlation between HIV prevalence and GDP per capita is not very strong, the consequences of an HIV infection differ substantially across countries. While mortality among people living with HIV typically was between 1% and 1.5% for high-income countries like France, Spain, or the US, it averaged 4.8% in 34 low-income countries in 2011, and exceeded 8% in Liberia, Nepal, and Somalia (according to estimates from the 2012 UNAIDS Report on the Global AIDS Epidemic). These estimates also illustrate the large impact of increased access to treatment – mortality among people living with HIV in the 34 low-income countries has declined by almost one-half (from 8.6%) since 2005. However, very large differences in the health consequences of an HIV infection across countries with different levels of economic development appear to persist.
Within countries, the correlation between HIV/AIDS and income (or other socioeconomic characteristics) is less straightforward. One of the most important data sources are demographic and health surveys also covering HIV prevalence. Most of these surveys suggest that HIV prevalence tends to be higher for wealthier population groups, but there is no consistent pattern across countries.
In summary, as is the case with infectious diseases more generally, HIV/AIDS deaths occur predominantly in developing countries. However, HIV/AIDS is unusual as it is distributed highly uneven across countries. These observations have implications for the macroeconomic significance of the epidemic. Because the health impact of HIV/AIDS has been so disruptive in specific countries, and because this health shock has emerged as a development threat only over the last 20 odd years, it is plausible that the epidemic has economic consequences (e.g., for GDP growth), which cannot easily be detected for more common and chronic health conditions (e.g., malaria). At the same time, the impact of HIV/AIDS provides a testing ground for theories on health and economic development.
Macroeconomic Impact Of HIV/AIDS
In spite of its devastating impact on health outcomes such as life expectancy in a number of countries, the impact of HIV/ AIDS on economic growth is not obvious. This point is illustrated by Figures 2 and 3, which contrast trends in life expectancy in 10 countries facing the highest HIV prevalence worldwide and the recent growth experience in these countries. (As all of these countries are located in sub-Saharan Africa, the figures also provide averages for the region for comparison.)
According to Figure 2, the impact of HIV/AIDS on life expectancy has been very large, ranging from a loss of 7 years (Uganda) to a loss of 21 years (Zimbabwe) in 2000–05. Moreover, the adverse impact was so strong that life expectancy declined in absolute terms in 8 of the 9 countries, and collapsed to a level last observed in the 1950s or 1960s in Botswana, Lesotho, South Africa, Zambia, and Zimbabwe. In some countries, the negative trend was started to reverse in 2005–10, partly because the HIV epidemic had matured (and the number of AIDS cases was no longer escalating) and partly as a consequence of increased access to treatment.
As evident from Figure 3, the large decline in life expectancy has not resulted in a steep drop in GDP growth. The rate of growth of GDP per capita in 9 countries with high HIV prevalence slowed down, somewhat relative to the rest of sub-Saharan Africa since the mid-1990s. However, the timing of the slowdown precedes or is less persistent than the increase in HIV/AIDS-related mortality. By 2010, the countries with high prevalence can be divided in two groups: (1) Low-income countries like Malawi, Mozambique, Zambia, and Zimbabwe, experiencing large swings in growth rates arguably not caused by HIV/AIDS (this applies especially to the economic crisis in Zimbabwe); (2) South Africa and the enclosed or neighboring middle-income countries Botswana, Lesotho, Namibia, and Swaziland, all experiencing growth rates below the average for sub-Saharan Africa, but which also differ from most countries in sub-Saharan Africa in many regards other than the state of HIV/AIDS.
The empirical evidence is also ambiguous. Studies including HIV prevalence or AIDS-related deaths directly in regressions find no or very small impacts of HIV/AIDS on growth. In contrast, studies identifying a large impact of HIV/ AIDS usually build on established findings of the empirical growth literature, notably the positive correlation of growth and life expectancy and then link the variable of interest to HIV/AIDS. In light of the strong impact of HIV/AIDS on life expectancy (or similar variables), this empirical approach returns a large negative impact of HIV/AIDS on growth but rests on two untested hypotheses: (1) The correlation of growth and life expectancy reflects a causal link between health and growth and (2) HIV/AIDS affects economic outcomes in a similar way as changes in the state of population health reflected in changes in life expectancy across countries. Both assumptions are doubtful. Some observers point to common factors like institutions affecting the functioning of health systems, governance, and growth. Also, the health impact of HIV/AIDS has a specific profile that does not simply reverse health gains achieved over the past decades, and it has occurred much more quickly than the gradual improvements in health outcomes achieved over the past decades.
If the links between health and economic outcomes are of a longer term nature, this could mean that the impacts of HIV/ AIDS on economic growth have not fully materialized yet. For example, economic theory suggests that higher mortality risks reduce the returns to education. HIV/AIDS could therefore slowdown the accumulation of human capital and economic growth. As this effect would take several decades to materialize (as cohorts grow from school benches through the working age population), it would barely show up in economic growth data at present, and there would not be a clear contemporary correlation between HIV prevalence and economic growth. Some microempirical evidence points to lower school attendance in areas highly affected by HIV/AIDS, consistent with such a hypothesis about the long-term economic consequences of HIV/AIDS.
Another possible reason why the impacts of HIV/AIDS on growth have been small so far is the fact that economic activity, within countries, is distributed unevenly. It has been observed that HIV is associated with certain economic activities like mining, and that migrant workers also play a large role in disseminating HIV. However, as value added per worker in mining is high, companies can afford to take actions to prevent any disruptions to production from increased mortality or morbidity, at a low cost relative to turnover or value added.
The discourse regarding the macroeconomic effects of HIV/ AIDS has focused on the growth impacts of the epidemic. It is important to take note of the fact that HIV/AIDS also results in a shift in the composition of spending. As governments and households shift expenditures to respond to the epidemic and address its consequences, these funds are no longer available for other purposes, i.e., private or public consumption and investment. Compared with a no-AIDS situation, HIV/AID Srelated spending therefore adds to the economic costs of the epidemic. The discussion in the Section Macroeconomic Aspects of the Response to HIV/AIDS, suggests that public HIV/AIDS spending accounts for several percent of GDP in a number of countries. Private HIV/AIDS spending and shifts in the allocation of time within households add to these economic costs.
The steep declines in life expectancy that can arise because of HIV/AIDS can also be interpreted as an economic cost. Such interpretations of the health impact of HIV/AIDS draw on estimates of the value of statistical life, which typically suggest that a loss in life expectancy of one percent is equivalent to an income loss of 3–4%. A loss in life expectancy of 23% (as in Botswana, 2005–10, compare Figure 2) would then translate into an economic cost exceeding one-half of GDP. Even in countries like the US, with an HIV prevalence of 0.6% and a loss in life expectancy of half an year, the costs of increased mortality, by this count, exceed 2% of GDP.
Small aggregate impacts of HIV/AIDS may mask shifts below the surface of national averages, which are relevant from a welfare perspective. For example, it is plausible that high HIV prevalence increases the risk to material living standards and – for parts of the population – of falling into poverty (even though other households may benefit, taking advantage of employment opportunities vacated by people affected by HIV/AIDS). Also, even though HIV prevalence tends to be somewhat higher among wealthier population groups, differences in access to treatment across population groups, in a country facing an HIV epidemic, can exacerbate inequalities in health prospects. Although demographic and health surveys consistently return higher rates of access to health services for wealthier population groups, little data are available regarding the benefit incidence of HIV/AIDS-related health services and the consequences of increased demand for HIV/AIDS-related health services (and a corresponding scaling-up in the supply of such services) for access to health services more generally.
Macroeconomic Aspects Of The Response To HIV/AIDS
The global response to HIV/AIDS has altered the course of the epidemic. The macroeconomic impact of HIV/AIDS therefore partly reflects the consequences of policy interventions, in several dimensions: (1) HIV incidence, (2) the microeconomic consequences, (3) the growth impacts of HIV/AIDS, and (4) the costs of the response to the epidemic.
In many countries, HIV incidence has declined very considerably from its peak. In South Africa, for example, HIV incidence among the population of ages 15–49 years declined from a peak of 2.8% in 1998 to 1.3% in 2011. As a consequence, the health outlook in countries experiencing such declines is improving, and the economic consequences of HIV/AIDS become less forceful.
More immediately, the adverse economic consequences of HIV/AIDS are modified by increased access to treatment. This intuition is supported by empirical analysis on the microeconomic level, illustrating a reversal in worker’s productivity following initiation of treatment. These estimates, however, are available only in settings where labor input and output are directly observable (e.g., tea pluckers) and may not translate one-to-one to other sectors and contexts, such as capitalintensive mining or services, which account for a large share in GDP.
The studies of the macroeconomic effects of HIV/AIDS also provide some pointers regarding the consequences of treatment (and the later studies frequently offer explicit estimates). In addition to mitigating productivity losses, antiretroviral treatment reduces the decline in population growth and reduces the private and public costs of care. Looking ahead, the prospect of access to treatment changes the risks associated with an HIV infection. Along with declining risk of becoming infected, it therefore increases the incentives to invest in education, therefore mitigating one of the most forceful effects through which HIV/AIDS could affect long-term growth.
Macroeconomic studies, which explicitly account for the impact of antiretroviral treatment, illustrate the extent to which increased access to treatment mitigates the economic impacts of HIV/AIDS, frequently suggesting a reversal in the growth impact of approximately one-third to one-half of the unfettered (‘no treatment’) impact of HIV/AIDS. This reversal is less than complete even where the rate of access to treatment is very high because treatment only mitigates and delays the adverse health consequences of HIV/AIDS, and because the costs of treatment crowd out other investments. Some observers argue that access to treatment could be financed from this ‘growth dividend’ (and reduced costs of other HIV/AID Srelated health services). This, however, is not necessarily the case, as the ‘growth dividend’ is not directly available for higher health spending (people surviving longer because of treatment need to eat).
The policy response does not merely reverse the adverse macroeconomic impacts of the epidemic. The costs of the response in many countries have attained a level that is significant from a fiscal perspective, and HIV/AIDS-related external aid may account for a substantial proportion of aid received. Globally, HIV/AIDS accounted for US$ 8.0 billion out of total disbursements of official development assistance of US$ 150 billion in 2011, and out of US$ 19.4 billion in the areas of health and population policies, according to the ‘‘creditor reporting system’’ database maintained by the Organisation for Economic Co-operation and Development.
The high costs of the response to HIV/AIDS in numerous countries are illustrated in Figure 4. The burden of funding the HIV/AIDS program, relative to GDP, is not necessarily the largest in the countries facing the highest HIV prevalence (Botswana, Lesotho, Namibia, South Africa, and Swaziland) but in a number of low-income countries facing HIV prevalence between 3% and 15%. In particular, some least-developed countries face a very large financing challenge, even though HIV prevalence is moderate. This is the case because the unit costs of HIV/AIDS interventions differ across countries much less than the level of GDP per capita.
The spending figures summarized in Figure 4 confirm that HIV/AIDS spending is significant from a fiscal perspective in many countries. In a typical low-income country (the figures are based on the median for this country group), public spending accounts for approximately 25% of GDP, of which 8% (equal to 2% of GDP) go toward health. According to Figure 4, the costs of the national response to HIV/AIDS (whether delivered through the public sector or non-governmental organisations) thus exceed total public health spending in a number of countries. These high levels of spending would be hard to envisage without high levels of external assistance.
Health is an area in which external assistance is playing a large role across developing countries in general. Owing to the uneven distribution of HIV/AIDS across countries, and the high costs of HIV/AIDS in a number of countries, the role of external assistance is even more pronounced in the area of HIV/AIDS spending, as illustrated by Figure 5. For low-income countries (broadly, those with GDP per capita of less than US$ 1000 in Figure 5), external financing usually accounts for more than 80% of the total costs of the HIV/AIDS program and in some cases close to 100%. In contrast, external assistance for public health spending rarely exceeds two-thirds of total spending. The differences in external funding between HIV/AIDS and health are even more pronounced for middle-income countries including countries which are not facing very high HIV prevalence rates.
Looking ahead, two aspects of the fiscal dimension of HIV/ AIDS are worth noting. First, the costs of HIV/AIDS programs are going to remain high for a long time, even where HIV incidence is declining. This is the case because the number of people receiving treatment are still rising, and an increasing number of people who have contracted HIV in the past will require treatment. Second, there is a perception (and some early evidence) that external funding for HIV/AIDS is stagnating or even declining. This will place the funding of HIV/AIDS programs under pressure, especially in low-income countries where HIV/AIDS spending is high relative to the government’s fiscal resources.
The impact of HIV/AIDS on economic growth has been small so far. This finding raises some questions regarding the empirical literature on health and growth (which would predict a large impact), but it could also be the case that the link from increased mortality to growth occurs so slowly and has not fully materialized yet. In many countries, HIV/AIDS programs have attained a scale that is significant from a fiscal perspective. The response to HIV/AIDS has been enabled by high rates of external assistance in the past, but the availability of funding is perceived to decline. Under these circumstances, sustaining the funding of HIV/AIDS programs will present a challenge especially for a number of low-income countries.
- Bachmann, M. O. and Booysen, F. L. R. (2006). Economic causes and effects of AIDS in South African households. AIDS 20(14), 1861–1867.
- Botswana Institute for Development Policy Analysis (2000). Macroeconomic impacts of the HIV/AIDS epidemic in Botswana. Gaborone, Botswana: BIDPA.
- Case, A. and Ardington, C. (2006). The impact of parental death on school outcomes: Longitudinal evidence from South Africa. Demography 43(3), 401–420.
- Case, A. and Paxson, C. (2011). The impact of the AIDS pandemic on health services in Africa: Evidence from demographic and health surveys. Demography 48(2), 675–697.
- Deaton, A. (2006). Global patterns of income and health: Facts, interpretations, and policies. NBER Working Paper No. 12269. Cambridge, MA: NBER.
- Ellis, L., Laubscher, P. and Smit, B. (2006). The macroeconomic impact of HIV/ AIDS under alternative intervention scenarios (with specific References: to ART) on the South African economy. Stellenbosch, South Africa: Bureau for Economic Research, University of Stellenbosch.
- Haacker, M. (ed.) (2004). The macroeconomics of HIV/AIDS. Washington, DC: International Monetary Fund.
- Nattrass, N. (2003). The moral economy of AIDS in South Africa. Cambridge: Cambridge University Press.
- Parkhurst, J. O. (2010). Understanding the correlations between wealth, poverty and human immunodeficiency virus infection in African countries. Bulletin of the World Health Organization 88(7), 519–526.
- Sahn, D. E. (2010). The socioeconomic dimensions of HIV/AIDS in Africa. Ithaca, NY and London: Cornell University Press.
- UNAIDS (2012). Report on the global AIDS epidemic 2012. Geneva: UNAIDS.
- Whiteside, A. (2008). HIV/AIDS: A very short introduction. Oxford and New York: Oxford University Press.