Education + Advocacy = Change


Click a topic below for an index of articles:





Financial or Socio-Economic Issues


Health Insurance



Institutional Issues

International Reports

Legal Concerns

Math Models or Methods to Predict Trends

Medical Issues

Our Sponsors

Occupational Concerns

Board members of HEART

Religion and infectious diseases

State Governments

Stigma or Discrimination Issues


If you would like to submit an article to this website, email us at for a review of this paper

any words all words
Results per page:

“The only thing necessary for these diseases to the triumph is for good people and governments to do nothing.”



A new addition to the HEART is our Forum-check it out

The potential impact of AIDS on democracy in Southern Africa

Robert Mattes, University of Cape Town

Occasional Paper 71, April 2003



Social scientists are only beginning to understand the range of potential impacts the HIV/AIDS pandemic may have on Southern African societies. Belatedly, researchers began compiling evidence about the demographic, economic and social impacts of the disease on infected people, their households and communities, national populations and national economies.2 They have only recently begun to develop propositions about the impacts of HIV/AIDS on the broader processes of governance.3 However, the implications of the pandemic for the survival and consolidation of democratic government, in particular, remain largely unexamined.4 This paper attempts to systematise emerging thinking about the various economic, social and political consequences of HIV/AIDS in the context of political science's best available knowledge about the factors that lead to the consolidation of democracy.

But why a particular focus on the impact of HIV/AIDS on democracy? Would not any disease of this scope—for example malaria, which currently kills more people than AIDS—pose a significant threat to any regime? While this may be true, HIV/AIDS is not just any pandemic. Its primarily sexual mode of transmission allows it to spread quickly and silently throughout a population. The relatively long time span between HIV infection and death due to AIDS complications imposes a virtual death sentence on significant portions of a population. And its sexual mode of transmission brings a range of social stigma that adds to the suffering of its victims.

HIV/AIDS is unique in terms of the shape of the epidemic. Like other infectious disease epidemics, HIV/AIDS follows an 'S' curve. What sets the HIV/AIDS epidemic apart from other infectious disease epidemics is the presence of two 'S' curves: one illustrative of asymptomatic HIV, the other of symptomatic 'full blown' AIDS. The HIV curve precedes the AIDS curve by about five to eight years.5

This long incubation period has helped to make HIV/AIDS more deadly than other life threatening diseases such as cholera or Ebola fever. In the case of the latter diseases, victims progress from infection to visible illness and possible death in a matter of days or weeks. This serves to immobilise sufferers—thereby restricting the spread of the disease—and alerts health authorities who can then act to combat its spread.6 In the case of HIV/AIDS the long period between infection and the appearance of symptoms allows the virus to spread unabated, and facilitates ignorance and denial of the disease.7

But if HIV/AIDS has unique characteristics, is it not true that it poses a common threat to the security and governance of any state?8 Again, the answer is 'no'. Democracy is not just any regime. First of all, researchers should be especially interested in the plight of democracy, as opposed to other types of regimes. Over the past two centuries, democracy has come to enjoy a privileged normative status as the preferred type of political regime because it uniquely recognizes the moral agency and dignity of human beings, and thus their right to determine their individual and collective fates.9 Yet it is precisely this moral agency and human dignity that may be most severely challenged by HIV/AIDS, an argument that will be fleshed out more fully below.

Second, while they may be normatively privileged, democracies are fragile and more difficult to sustain than non-democratic regimes.10 Of the 71 democracies that were established by or after 1950, 33 had died by 1990, lasting an average of just 5.1 years (compared to the 9.4 years of dictatorships that were born after 1950 but died prior to 1990). The 38 democracies surviving after 1990 had an average life span of 13 years (compared to 26.2 years for surviving dictatorships).


In continental Southern Africa, the 1990s saw eight countries achieve a successful transition from an authoritarian regime to a founding democratic election, joining the two existing multi-party systems in Botswana and Zimbabwe. Yet many of these democratisation processes are incomplete and some have undergone reversals. Even where states have implemented regular elections and secured the conditions for political competition, pluralism, and the protection of human rights, democracy remains far from consolidated.

As defined by Larry Diamond, continental Southern Africa contains three emerging 'liberal democracies' (countries that combine genuine political competition with a full range of political freedoms and civil rights).11 Based on Freedom House's (an internationally respected democracy research institute) latest ratings of political and civil rights, South Africa, Botswana, and Namibia are all rated as 'free', thus falling into this category. Yet even these countries run the risk of eventually degenerating to what Diamond calls 'semi democracies' because the existence of single dominant political parties may over time limit effective competition. The region contains four functioning 'electoral democracies': that is they combine genuine political competition with an insufficient protection of rights. Malawi, Tanzania, Mozambique and Lesotho are rated as 'partly free' by Freedom House, thus falling into this category.

The region also contains what Diamond calls 'pseudo democracies' or what Richard Joseph calls 'virtual democracies'. These countries hold elections and allow opposition parties, but competition, pluralism and rights of association, speech and media are actively constrained by the state. Zambia and Zimbabwe are rated as 'partly free' by Freedom House but score sufficiently badly on political rights that they fall into the 'pseudo-democracy' category.12 Freedom House ratings show retrograde trends in political freedom and civil liberties in Malawi and Zimbabwe.13

Thus, even without the presence of HIV/AIDS, the future of democracy in the region is far from certain. None of Southern Africa's democracies can yet be considered to be consolidated: with democratic consolidation meaning little or no probability that a country will abandon regular, free and fair multi-party elections as a way of selecting its rulers.

Social scientists have identified three key factors crucial to sustaining and consolidating democratic rule. The first factor has to do with economics. While wealthier countries are no more likely than poorer ones to have transited from authoritarian to democratic rule, wealthier countries are far more likely to maintain democratic rule. Poor countries can, however, increase the prospects of democratic endurance if their economies grow steadily, and if they reduce inequalities.14

The second factor has to do with political institutions. That is, sustainable democracies require a professional civil service and strong, viable and autonomous courts, legislatures, executives and electoral systems at national and local levels. The 'institutionalisation' of such bodies and processes requires skilled personnel with sufficient resources, developing specialised areas of expertise, and following clear and predictable rules and procedures. As such the rules of democratic governance become routinised and independent of the forces of clientelism, corruption or the whims of the ruling political party.

The third factor has to do with the attitudes of rulers and citizens. Put simply, democracies require democrats. Democracy can only be considered consolidated if it has been legitimated. That is, all relevant elites and the overwhelming majority of citizens must see democracy as ‘the only game in town’.15

Reaching the appropriate thresholds across each of these factors will be difficult for Southern Africa's nascent democracies even under normal circumstances. However, across each factor, the HIV/AIDS epidemic threatens to make consolidation even more difficult, if not completely unobtainable. Of even greater concern, the burdens of the epidemic may so strain political systems as to trigger powerful processes that lead to democratic backsliding and reversal.


Ever since the pioneering work of Seymour Martin Lipset, social scientists have been aware of the strong correlation between national wealth and democracy.16 The most recent work by Adam Przeworski and his colleagues has given clear specification of the linkages between economic development and democratic endurance. Wealthy democracies do not die. The wealthiest democracy to revert to authoritarian rule was Argentina, in 1974, with a Gross National Product (GNP) per capita of US$6,055. Under that threshold, the death rate of democratic systems increases as national wealth declines. While those with a GNP per capita between US$6,001 and US$7,000 have a probability of reversion in any single year of 0.008 (or an average life expectancy of 125 years), those with a GNP per capita of US$1,000 or less have a probability of reversion of 0.122 (or an average life expectancy of approximately eight years).17

Given these findings, in theory none of Southern Africa's multiparty systems has a very long 'life expectancy'. Botswana and South Africa are the region's wealthiest countries, yet all other things being equal, their GNP per capita suggests a democratic life expectancy of just 36 years. And democracies with Namibia's level of wealth have lasted, on average, just 18 years. Given their levels of national wealth, the rest of the region's multiparty systems have an average calculated life expectancy of just over eight years (Table 1).18

Table 1: National wealth and life expectancies of continental Southern African democracies


GNP per capita US$ (1999)

GNP per capita: global rank (1999)

Probability of democratic breakdown

Average 'democracy life expectancy' in years

South Africa















Source: World Development Report, 2000/2001; and Przeworksi et al.






But Przeworski et al also demonstrate that poor countries can sustain democratic rule provided their economies grow and income inequalities are reduced (India being the most prominent example). Where a democracy's economy has contracted, its probability of extinction is 0.05 (around 1 in 20), but when incomes are growing, this decreases to 0.015 (or 1 in 66). If growth is 5% or higher, the probability of democratic reversal declines even further to 0.013 (or 1 in 77). The odds of a democratic reversal fall to just 1 in 135 if its economy grows for three or more consecutive years.19

However, the opposite is also true. In the words of Diamond and Linz, "economic crisis represents one of the most common threats to democratic stability".20 Where incomes contract in one year, a democracy is three times as likely to break down in the following year compared to a democracy that experienced growth (1 in 20 versus 1 in 66, respectively). When incomes decline for two or more consecutive years, the chances of democratic breakdown rise to 1 in 13. Of the 39 'democratic deaths' observed between 1950 and 1990, 28 were accompanied by a fall in per capita income in either one or two years preceding the end of democratic rule.21

Democracies are more sensitive to economic stagnation and crisis than authoritarian regimes, and poorer democracies are more sensitive than richer ones. Poorer democracies, those with per capita GNP under US$2,000, fail in approximately one out of every ten occasions after incomes decline in the previous year. Middle income democracies, those with a per capita GNP between US$2,000 and US$5,000, fail approximately one out of every 16 times after incomes decline. As noted earlier, no democracy with a per capita GNP of above US$6,055 (Argentina, 1974) has ever failed.22

None of this bodes well for Southern Africa's low to lower-middle income democracies. From 1998/99 to 2000/2001, regional growth has generally been low with only one country (Mozambique, 6.6%) above the 5% threshold. Positive growth rates were also recorded in Malawi (4.4%), Tanzania (3.1%), Botswana (3%), Namibia (0.6%) and Zambia (0.4%). However per capita income declined in Lesotho (-3%), Zimbabwe (-1.8%) and South Africa (-0.9%).23

Finally, Przeworski et al have found that inequality reduction is an important factor in poorer democracies. While the available data is limited, they found that out of 358 country-years where a democracy had a GINI Index below the international mean (0.35), there was only one reversal.24 However, of the 379 country-years where a democracy had a GINI Index above 0.35, five democracies broke down. Democracies are particularly unstable when the income share of the bottom 40% of the population declines.25

While data is incomplete, inequality (as represented by the GINI Index) is extremely high in at least three countries in the region: South Africa (0.59), Zimbabwe (0.57) and Lesotho (0.56). But it is still above the international mean in three other countries: Zambia (0.50), Mozambique (0.40) and Tanzania (0.38).26 Moreover, indications are that income inequality is increasing in South Africa, and that the proportion of the national income going to the bottom two-fifths of the population has declined since 1994.27

This is not encouraging news, especially as the HIV/AIDS pandemic is widely expected to limit growth, increase inequality and reduce national wealth across the region. Because HIV infection is spread predominantly through sexual activity, AIDS related illness and death occurs disproportionately amongst younger, economically active people. This is expected to reduce household earnings and personal savings, as well as human capital and the size and skills of future work forces. Firms are expected to face higher wage bills because of increased employer contributions to pension, life and medical benefits, as well as higher training and replacement costs. Productivity is expected to decrease as a result of a decreased skills base, lower worker morale, increased absenteeism and the necessity of constantly replacing skilled workers.

Based on these assumptions, economists project an average reduction in per capita income growth of between 1% and 1.5% per annum for the entire sub-Saharan region.28 For South Africa, studies forecast declines between 0.2% to 0.3% until 2005 and between 0.3% and 0.4% thereafter.29 In cumulative terms, growth will be 0.5% lower in year two of the projection and 2.6% lower by year eleven.30 The impact on overall per capita income, however, is less clear. While AIDS is expected to slow income growth, it will also slow population growth. If populations fall faster than national incomes, GDP per capita income may actually increase.31

Economists expect the pandemic to increase inequality, at least in South Africa. Skilled workers may benefit from greater demand and higher wages for their labour, a larger supply of goods produced for their niche market, and a relatively longer life span because of access to anti-retroviral drugs provided by their firms. The relatively unskilled and unemployed will face declining income, fewer and more expensive goods produced for them, and greater morbidity and mortality. While total GDP shrinks, the skilled may enjoy an increasing share.32

It is important to exercise caution with regard to these projections. Final projections may vary in important ways depending upon the assumptions underlying the projective model. Take, for example, assumptions about how the costs of HIV infections affect firms' decisions on investment, pricing and hiring. The cost of anti-retrovirals has plummeted so much that firms may soon be able to share costs with workers and provide access to medication, at least to their highly skilled personnel. If this happens, existing models will have over-estimated the effect the HIV pandemic will have on the workforce. At an even more fundamental level, economists have assumed that firms have the necessary basic information to calculate the labour costs of AIDS and its impact on their profits, and that they do in fact perform such calculations.

Recent research in South Africa suggests that firms do not collect or use existing information and thus have only the vaguest sense of indirect costs. Moreover, direct costs may comprise less than 5% of the total wage bill because most firms provide so few pension and medical benefits to unskilled workers in the first place, but also because many employees choose to leave their job once they discover they are HIV-positive.33 In fact, a Department for International Development (DFID) sponsored survey of South African firms found that most firms expected HIV/AIDS to have either a moderate (53%), or little or no (38%) impact on their business.34

Existing projections also fail to 'trace through' the 'second order' impacts that occur after people, firms and governments respond to the 'first order' impact of AIDS on things like consumption, investment, savings and the size of the labour force. Will firms respond to higher medical costs by replacing workers with machines, raising costs, reducing profits or sharing costs with workers? Will governments increase health spending, and if so will this be at the expense of other expenditure items? Will governments finance rising expenditure through increased taxes or increased borrowing? Nicoli Natrass, an economist with the Centre for Social Science Research at the University of Cape Town, has shown how differing answers to these questions can result in widely varying projections.35

Existing models are linear, in that they model the particular economic impacts of a range of factors and then sum them. However, Alex de Waal, an adviser on HIV/AIDS and governance to the Economic Commission for Africa, points out that development should be understood as a process, and thus model the interactions among the different economic impacts, as well as second order reactions to first order effects. Dynamics may accelerate if they reach critical thresholds or co-occur in specific combinations.




A new addition to the HEART is our Forum-check it out

No models, for example, attempt to forecast the way in which sharply decreased adult longevity may reshape economic rationality and thus the decisions of firms, household and individuals. Most economic theory, rooted firmly in the logic of developed economies, is based on the principle that individuals rationally consider future expectations, which in many ways is based on their anticipated lifespan. De Waal points out that over the past century, most people in developed societies—once they have reached their teens—could expect to live into their 70s. But when adult lifespan becomes highly uncertain, individuals' rationale to invest time and resources in their education or training, or to save and invest is less rational. Rather people's emphasis turns to spending and consuming, to liquidating assets to pay for health care and to enjoy their short lifespan.36 As de Waal puts it:

Why save for a future that does not exist? ... Just as a doubling of longevity would entail major structural transformations of developed economies, a halving of adult lifespan in much of sub-Saharan entails a structural change in the region's economies that make it impossible for it to follow existing models for economic development.37

Budgets and taxes: increased demands and decreased supply

Existing levels of HIV infection mean that Southern African societies will experience drastically increased levels of severe illness. Resultant increasing demands for access to government clinics, medical stocks, and available hospital beds threaten to push the region's limited public health systems to, and past, their limits. As governments increase health spending to cope with the increasing demand for medicine and hospital beds, they will be faced with a growing wage bill to replace dying nurses and doctors from an increasingly scarce labour pool and growing expenditure to assist families caring for orphans. For most countries in Southern Africa, public spending on HIV/AIDS related matters threatens to consume their entire health budget, or increase the overall national budget.

Evidence, albeit often anecdotal and episodic, is beginning to accumulate that suggests these impacts are already occurring. In South Africa, government reports have admitted that substantial increases in AIDS patients are steadily displacing other patients.38 In South Africa's industrial heartland, Gauteng province, Soweto's Chris Hani Baragwanath Hospital, the country's largest, has seen a 500% increase in HIV patients since 1996. One half of all beds in the province's public hospitals are occupied by AIDS patients.39 Yet the country's health system is unable to meet these increased demands. A 2002 survey of AIDS affected households in four South African provinces found that 40% to 60% of people living with AIDS had never been admitted to a hospital.40 Just 16% of eligible households were able to obtain a state grant.41 South African government estimates predict the number of orphans rising from the present 150,000 to two million by 2010, far outstripping the capacity of the existing welfare system.42

What potential impacts will these rising demands on national budgets have on the region's multi-party systems? Economic policy makers will come under increasing pressure to devote increasing shares of the national budget to public health and welfare expenditure for AIDS orphans and the families or institutions who care for them. Significant shifts have already occurred. In 1997 public health spending on AIDS exceeded 2.5% of GDP in seven of 16 sampled African countries, an extremely high proportion since total health spending accounted for only 3% to 5% of GDP in these countries.43 As of 1998, 12% of South Africa's national budget went to health, and 21% of provincial budgets have been allocated specifically to HIV/AIDS related expenditure.44 The total proportion of the South African government's budget devoted to AIDS will increase by 22% (R1.4 billion to R1.8 billion) between 2003 and 2004.45 Yet according to South African government reports, even with present levels of real growth in future health budgets, the health system will not be able to keep pace with increased demand for services.46

In Zambia, government AIDS spending rose from US$1.7 million in 1990 to US$12.9 by 1995, and is expected to rise to US$21 million by 2005.47 Zimbabwe spends almost half of its health budget on treating AIDS patients.48 This is expected to rise to almost two-thirds by 2005.49 In Malawi, senior government health officials have publicly expressed their concern that half of their health budget will soon have to be devoted to treating AIDS patients.50

Regardless of whether or not they decide to provide drugs that reduce mother-to-child transmission, or anti-retroviral therapies, countries with any form of public health system will ultimately have to confront these costs. Nattrass and Jolene Skordis conclude from a costing of mother-to-child transmission drugs in South Africa,

unless the government is planning to deny hospital care to children with HIV/AIDS (which would be unconstitutional in South Africa), it costs the government more to let the children contract HIV from their mothers, get sick and die, than it does to save them.51

While relatively wealthy countries like South Africa or Botswana may be able to redistribute expenditures between budget sectors, or embark upon limited, disciplined deficit spending, the poorest countries in the region will be forced to increase their dependence on foreign assistance. International donors already account for approximately two-thirds of HIV/AIDS budgets in low and middle income countries across the world, largely accounted by overseas development assistance.52 As of 1996/97, sub-Saharan African governments are contributing an average of just 9% of all HIV/AIDS spending in their countries.53 Even in South Africa, which is significantly better able to support its own efforts, a 2000 Futures Group International study found that it still received around half of its total HIV/AIDS funding from external sources.54

HIV/AIDS may consequently increase the reliance of Southern African states on international donor funding. Zimbabwe's foreign assistance needs have reportedly already increased by 27% because of AIDS.55 Such increased demand for health related donor assistance may crowd out other development assistance funds that countries badly need.56 And if widely known, such dependence on donor funding may reduce the recipient government's popular legitimacy because of the conditionality with which donor funds usually come.

Because HIV/AIDS has a disproportionate impact on the most productive part of the labour force, it will disproportionately reduce the numbers of those most able to contribute to the national treasury through payroll taxes. Thus, at the same time that the HIV/AIDS pandemic raises public demands for increases in public health expenditures, increased levels of death and illness will simultaneously reduce the tax base from which governments finance their budgets.

How will governments pay for increased budgetary demands in the face of a decreasing tax base? If they choose deficit financing, they will crowd out the private investment that could otherwise increase employment and growth. If they choose expenditure switching, they will crowd out government investment in infrastructure or development programmes (e.g. housing, education or land redistribution) that might otherwise help reduce inequality and build public confidence in democratic government.


Democracy is a system of rules and procedures by which free and equal people elect representatives to make decisions for them, and a system of rules and procedures by which those representatives make decisions. A consolidated democracy is one in which these rules are widely known, and predictable—where the processes of democratic governance become a matter of habit.

These rules are given effect through political institutions—such as legislatures, executives, courts and regulatory and security agencies—which both embody and enforce these rules. To work effectively, these institutions require people with sufficient skills, expertise and resources to develop sufficient political autonomy and power to fulfil their functions, whether to make laws, oversee the executive, prosecute criminals, or deliver public services impartially. In this sense, the development of a strong and effective state is a necessary, though certainly not sufficient, condition for the consolidation of democracy.

There is little reason to believe that state employees across Southern Africa are more immune to HIV infection than the rest of the population. While the greatest number of infections and deaths are projected to occur in the unskilled and semi-skilled part of the workforce, skilled and highly skilled sectors will nevertheless be hit heavily. In South Africa, HIV infection rates have been projected to peak at 23% of skilled and 13% of highly skilled workers by 2005. By 2015 this will result in a skilled workforce that is 18% smaller, and a highly skilled force that is 11% smaller.57 Thus, the pandemic is likely to devastate large portions of policy-makers, national legislators, local councillors, election officials, soldiers and civil servants—including doctors, nurses, teachers, ambulance drivers, fire-fighters and police. South African government reports conclude that AIDS will have become the leading cause of death among public servants by 2002, resulting in an estimated 250,000 deaths in the public service by 2012, or 23% of the present workforce of 1.1 million employees.58

According to media reports, HIV/AIDS has resulted in the death of a senior presidential advisor, a sitting cabinet minister, and an influential legislative back-bencher in South Africa's ruling party. In Zimbabwe at least three cabinet ministers have succumbed to AIDS.59 At the local council level, a 2002 analysis of Durban's (eThekwini) metropolitan council records over a 21-month period shows sharp increases in the extent of councillor absenteeism because of illness (from less than one in the first half of 2001, to over four in mid 2002), as well as in the proportion of total absenteeism due to illness (from less than 5% in the first half of 2001 to 37% in late 2002).60 The Parks, Recreation and Culture Department reported a 32% turnover in personnel in the previous six months. The electricity department estimated that they experience four to five employee deaths and two medical boardings per month over the past two years double their previous rates.61

Besides killing increasing numbers of public servants and elected officials, the pandemic could severely damage the process of political institutionalisation in several ways. A shrinking proportion of civil servants, policy-makers and legislators will be at their jobs long enough to develop the specialised skills, expertise and professionalism needed to do their jobs.

It will be increasingly difficult for legislatures, ministries and government agencies to pass on the skills that they do have. There will be fewer experienced officials available to train younger personnel in key formal skills (such as programme design, budgeting, cost/benefit analysis, monitoring and evaluation, and personnel management), or pass on more informal standard operating procedures or norms such as ministerial accountability, bureaucratic neutrality and official ethics. Identifying training needs and grooming replacements is likely to be made more difficult by the stigma of AIDS which means that civil servants may leave work and die with little warning.

In South Africa, government reviews have concluded that the country is not training enough nurses and teachers to cope with current demand, and pointed to similar problems looming in the police and justice system.62 In Zambia and Malawi, deaths among doctors and nurses have exceeded the rate at which replacements can be trained.63 In Durban, fire department managers have noted that while it takes three months to train a fire-fighter, it takes years to create one with enough skills to pass knowledge onto younger members through informal training.64 When asked about their ability to co-ordinate and plan the recruitment and planning or replacements, several city managers related stories of processing applications for medical boarding on a Friday only to find out the following Monday that the person had died over the weekend.65

Where civil servants endeavour to deliver public services according to rational principles of need or merit, the rapidly changing demographic impacts of the HIV/AIDS pandemic will make it difficult to anticipate demand accurately and plan the types, amounts and locations of services to be supplied. National and local governments may invest in services that end up under-utilised because of an unanticipated fall in demand, or they may face unanticipated demands because their ability to supply has fallen faster than the decline in the overall population.66

For example, the managers of Durban's municipal subsidised housing programme admit that they know little about how HIV/AIDS will affect patterns of household formation. South Africa awards subsidies to households, not individuals. Thus, even if it were possible to make relatively accurate projections of decreases in eligible individuals, it would be difficult to predict changes to the number of potentially subsidised households. The pandemic may reduce demand for housing if splintered families combine into new households; but it may also increase demand if families splinter and scatter into new, smaller households, or if it spurs increased urban migration. The increase in orphans also places new challenges on housing planners who must find new ways to increase the capacity of communities to absorb these children.67

All this is particularly worrying since the region's nascent democracies lack strong political institutions. The dominant characteristic of political institutions across the region, and indeed most of sub-Saharan Africa, is one of neo-patrimonialism whereby 'strong man' political leaders use patronage to gain and maintain political loyalty. Patronage relationships shape the behaviour of legislators and civil servants as much as any legal-rational principles of a bureaucratic state.68 A civil service characterised by a high degree of staff turnover and growing proportions of inexperienced personnel will be even less likely to develop and enforce institutional boundaries and autonomy. Conversely, such a civil service is more likely to succumb to patronage or corruption payoffs by a neo-patrimonial executive branch, party officials or business people.

HIV/AIDS threatens at least two political institutions which are intrinsically important for the development of young democracies. The first is the set of institutions responsible for organising and conducting regular free and fair elections, the irreducible minimum of democratic government. The loss of non-partisan supervisory officials, combined with the complicated voter registration procedures of Southern Africa's multi-party systems, will increase opportunities for voter fraud. Increased deaths may necessitate more regular vetting of voters' rolls. Alternatively, increasing death rates will increase opportunities for governments to utilise 'ghost voters' to inflate vote totals. Increasing proportions of ill voters may also necessitate more, or more strategically located, polling places and greater use of absentee ballots to enable the ill to vote.69

Thus, better electoral administration skills may become more necessary at the same time that electoral commissions begin to lose skilled personnel. Moreover, the funds necessary to register voters, maintain voters' rolls and hold free and fair elections may be crowded out by increasing shares of national budgets going to health care or anti-HIV/AIDS programmes.70 These prospects may threaten to damage popular perceptions of the impartiality of elections.

Countries with specific types of electoral systems may come under additional pressures. Increased deaths among Members of Parliament (MPs) and local councillors in constituency based systems will increase the number of by-elections. A steady flow of by-elections may increase government sensitivity to shifts in public opinion, but may also be financially unsustainable. Countries may be forced to abandon constituency representation in favour of party list proportional representation systems. While the list system provides for the swift and cheap replacement of sick or dead legislators it also removes a fundamental linkage between governors and the governed.

The second set of key democratic institutions theatened by HIV/AIDS includes national, regional and local legislative bodies, the sine qua non of representative democracy. Legislatures best represent constituent views when they develop institutional autonomy vis-à-vis the executive. This is usually achieved through the development of a seniority system that encourages the accumulation of skills in the use of legislative procedures and rules, as well as substantive specialisation and expertise in specific policy areas to enable informed oversight of executive policy. Such skills help create stronger and more autonomous parliamentary portfolio committees. This applies to both elected members as well as parliamentary researchers, administrative assistants and clerks. Rapid membership turnover due to AIDS illness and death threatens these processes.


Democracies require democrats. They require citizens who believe democracy is preferable to all alternatives, and who give life to the democratic processes by obeying the law, participating in democratic life, refraining from supporting elites who could endanger or end democratic processes, and who are willing to stand up and defend democracy if it is under threat.71

As noted earlier, democracy as a political regime uniquely recognises and is designed to maximise human agency. But the unique combination of the characteristics of HIV/AIDS attacks this sense of human agency. First, because it disproportionately affects large numbers of younger age cohorts, it results in a drastically reduced adult lifespan. Second, the combination of its scope and its incubation period means that at any given moment, a large proportion of society will be living under a death sentence. Third, the length of the incubation period means that many people will live under this 'death sentence' for a prolonged period.

Fourth, at least in Southern Africa, HIV infection and the onset of AIDS illness imposes significant economic burdens on individuals and their households in the form of increased medical costs, the prospect of losing a wage earner, as well as significant burial costs.72

Finally, given that few people can be confident about the HIV status of current or prospective sexual partners, the uninfected are also likely to experience a sense of helplessness and lack of control over their future. For example, where national prevalence is 15%, and this rate applies throughout a person's lifetime, more than half of today's 15-year-olds will die from AIDS.73 Given the lack of any clear, positive message of hope to uninfected people in those age groups about how they can confidently avoid infection over an extended period of time, a large number of teenagers and young adults may conclude that they have little hope of avoiding this death sentence themselves.

Akin to de Waal's arguments about the need to revise traditional assumptions about homo economicus, we also need to consider what such a sharp reduction in adult life expectancy means for our assumptions about the behaviour of homo politicus. It is likely to recalibrate the context of citizens' rational decision making, in particular reducing the incentives for co-operative behaviour and increasing incentives for opportunistic behaviour.74 Thus, the pandemic not only damages the human body, but may also 'damage' the 'body politic'.75 This is likely to have several important political consequences.

Decreased citizen support for democratic government

At its most extreme, HIV/AIDS may turn citizens into authoritarians because mounting death and sickness drives them, in desperation, to try any set of political entrepreneurs who promise to offer a solution, whether they use democratic means or not. These could, for example, be political or religious movements who seek to place the blame for the pandemic on personal immorality, religious transgressions, minority groups, or external forces.76 Such a scenario could become more likely if the region's democratic governments are widely perceived to be incapable of preventing the pandemic from growing, caring for its victims or preventing a sharp deterioration in quality of life.77

More likely, the pandemic may reduce the importance which people attach to democracy because of more urgent priorities such as simple survival. The degree to which democratic versus authoritarian government matters to someone infected with a fatal disease, or whose life is burdened with caring for such people, or who believes they have little prospect of avoiding infection, is an open question.

Decreased citizen participation

HIV/AIDS may also reduce overall levels of public participation in democratic politics. Mounting AIDS deaths and illness will reduce the absolute number of citizens able to vote or participate in public life. But the question also arises whether the 'death sentence' imposed on the infected, or the threat of infection facing the uninfected will reshape the usual incentives to becoming involved in public affairs. Moreover, the burden of caring for ill family members or friends is likely to reduce those people's time and resources available to participate in public affairs. For example, many countries in the region require people to obtain multiple forms of identification to register to vote, which places both the ill and caregivers at a disadvantage.78

However, it is also possible that some counter-trends will occur. As medical research brings down the costs of anti-AIDS medication, or comes closer to developing an AIDS vaccine, the infected and their loved ones might be galvanised into a strong and active constituency to demand that their governments make these drugs widely available. Though it is composed of a relatively small number of activists, South Africa's Treatment Action Campaign (TAC) may be one example of a broad social movement.

A damaged civil society

The level of public participation in a democracy is not determined solely by the choices of ordinary citizens. Mass participation in the political system is also facilitated by civil society organisations and interest groups who mobilise, channel and structure public participation between elections. Civil society across Southern Africa is weak. However, it could be further weakened because the types of people who form the backbone of most civil society organisations may be especially susceptible to HIV infection: while they tend to be better educated, they also tend to be younger, more mobile, and often spend time away from their home office during the course of their work. It is estimated that a medium to large civil society organisation with 30 staff and 80 volunteers, based in South Africa's KwaZulu-Natal province, will lose one to two employees and one to two volunteers annually by 2009 to AIDS, totalling 19 staff and 22 volunteers by 2021. By 2009, this would mean the loss of 158 days of staff time and 212 days of volunteer time annually due to AIDS related illness.79

Many organisations may be particularly vulnerable since lost staff members take with them unique skills that took many years to develop. The loss of long-term staff members or volunteers is likely to be particularly devastating. As the director of a KwaZulu-Natal democracy promotion organisation pointed out, such staff possess experience vital to understanding the culture, traditions and dynamics of the communities where they work, and have build up mutual trust and understanding with communities, churches and traditional leaders; relationships that are the very strength of these organisation.80

The pandemic may thus reduce overall levels of public participation in democratic processes by inhibiting the capacity of those civil society organisations who organise and channel political activity.81 Some evidence of this dynamic may be found at the societal level in an Afrobarometer survey of seven Southern African countries. It found that countries with the highest measured levels of severe illness (due to any cause) also have the lowest overall levels of attendance in local community meetings and participation in local service and welfare groups. This relationship persists even after controlling for levels of poverty. This may reflect the fact that AIDS has already resulted in the death of critical proportions of those who organise and drive community meeting or local welfare groups.82 And at the organisational level, a survey of 59 KwaZulu-Natal civil society organisations found that three-quarters reported some form of AIDS-related impact on their organisation. One-third of the Durban-based organisations noted absenteeism or loss of staff members to HIV/AIDS.83

The 'uncivil' society

Finally, the pandemic may have important effects on the 'civility' of society, decreasing popular compliance with the law, and increasing violent protest, social intolerance and criminal activity.

Individuals or households who suffer an AIDS illness, or have lost a wage-earner to severe illness or death, will be less able to pay local taxes or rates, or for public services such as electricity or user fees for schools or health clinics. Moreover, HIV infection may reduce or totally remove any incentive to do so, especially if payment is required after services have been provided rather than before. In contrast, the incentive would appear to dictate getting whatever you can for as little money as possible. De Waal notes: "Those who feel they have nothing to lose cannot be deterred by a judicial system that imposes custodial sentences, or even the death sentence."84

Another impact on political culture may come from the tendency of people infected with HIV or ill with AIDS to be lonely and depressed. Such conditions often lead to hopelessness and apathy, and even frustration and aggression, which could result in the non-compliance with laws or even political violence.85 Paradoxically, falling prices of some anti-retroviral drugs may aggravate and inflame these simmering frustrations because firms, and eventually governments will be in the position of deciding who does and does not receive life prolonging drugs. The crucial period will occur after it becomes affordable to provide anti-retroviral treatment to some employees and citizens, but before it becomes possible to provide them to all. During this period firms will have an incentive to provide limited drug therapy for their most highly skilled workers. If done widely enough, and for a prolonged period, this could reinforce class divisions and inequalities between different socio-economic classes in society.86 Even if firms are able to provide drug therapy for all their employees, in the context of high unemployment levels, this could still fuel class-based conflict between those with a decent job and everyone else who either works in the informal sector or is unemployed.

The wide-scale private sector provision of anti-retroviral treatment for employees may increase the pressure on governments across the region to also provide drug therapy to public sector employees and citizens not covered by their employers.87 If resources are limited, governments will be confronted with decisions about where, and for whom, to prioritise access to drug therapy. As with firms, governments may be tempted to use skills and education as criteria of eligibility for treatment, to protect its investments in training and education.88 Alternatively, governments may be tempted to prioritise the allocation of life-extending drugs to its supporters. While such possibilities might seem outlandish, the government of Zimbabwe, under the presidency of Robert Mugabe, has shown itself quite capable of manipulating the distribution of food aid to favour its political supporters. In Zambia, applicants for management positions in the mining industry are screened for HIV, and in Botswana prestigious study-abroad scholarships are restricted to HIV-negative applicants.89

Regardless of whether or not governments actually do this, the perceptions of politically expedient preferences in the provision of anti-retroviral drugs may be bolstered by the fact that South Africa's highest HIV rates are located in KwaZulu-Natal, political heartland of the opposition Inkatha Freedom Party, or that Zimbabwe's non-Shona population suffer higher HIV infection rates and have less adequate heath care than the largely pro-government Shona population. Or, that Namibia's Herero and Damara populations have less access to government resources and health care.90 Even with the best of intentions, the uneven distribution of existing health care systems may advantage certain groups and areas. For example, while Botswana has an extensive primary health care system by Southern African standards, only one out of every 12 pregnant Batswana women receive mother-to-child-transmission treatment, most of these live in urban areas.91

Thus, political decisions around AIDS treatment hold the potential to aggravate tensions and political conflict between workers and unemployed; urban and rural populations; or between different ethnic groups.

Another dimension of 'uncivility' may arise not amongst the infected, but in the form of public intolerance targeted at stereotyped 'high risk' groups such as homosexuals, truck drivers, orphans, sex workers, miners or released prisoners. South Africans, for example, already tend to be intolerant of disliked social groups, political opponents and foreigners.92 Given the fear created by the HIV/AIDS epidemic, governments may succumb under popular pressure to deny political and economic rights to perceived 'high risk' groups. Alternatively, local groups may take it upon themselves, using terror and intimidation, to force HIV-positive people out of their schools and communities.

A 2002 survey in South Africa revealed that a tenth of AIDS-affected households had faced hostility or rejection from their community. Many reported being chased away, refused help, or confronted with uncaring attitudes by health care workers.93 A three-country study of 41 schools in Botswana, Malawi and Uganda found AIDS orphans were subject to insensitive treatment by teachers and administrators at schools. However instances of deliberate discrimination were quite rare.94

Finally, some analysts are concerned that HIV/AIDS may contribute to lawlessness by orphaning large numbers of children across the region over the next two decades.95

Without AIDS, orphans in developing countries around the world comprise an average of 2% of all children aged 15 and under.96 According to the United Nations Children's Fund (UNICEF), however, orphans exceed 20% of the under 15 population in Congo-Kinshasa, Malawi, Rwanda, Uganda, Zambia and Zimbabwe. A large part of sub-Saharan Africa's unusually high proportion of orphans has usually been attributed to political conflict. As of 1999, however, UNICEF estimated that just under one-third of all orphans in the sub-continent were AIDS orphans.97 Within Southern Africa, USAID estimated the total number of maternal or double orphans, as of 2000, at 2.9 million, or 8% of all children 15 and under, of which 65% were AIDS orphans.98

UNICEF projects the total number of maternal and double AIDS orphans to increase to 14 million across sub-Saharan Africa by 2010. Particularly sharp increases in the orphan population are expected in Botswana, Central African Republic, Lesotho, Mozambique, Namibia, South Africa, Swaziland and Zimbabwe, amounting to 30% to 40% of all children.99 In Southern Africa, USAID expects the total number of orphans to increase to 5.5 million by 2010, by then accounting for 16% of all children 15 and under, of which 87% will be AIDS orphans.100 In South Africa, a Medical Research Council report estimates that a third of children born in 2002 will be orphaned by 2015, with a total of around 1.9 million orphans under 15 and three million under 18.101

All orphans suffer the trauma of losing parents, and of higher levels of impoverishment, exclusion and abuse. The education of orphaned children is often one of the first casualties if extended families suffer losses of income.102 Demographic and Health Surveys from 15 sub-Saharan African countries in the 1990s found that 'out of school' rates of orphans aged 10 to 14 were on average 19% higher than non-orphans.103 A Botswana, Malawi and Uganda study of schools in high prevalence districts found that orphans, especially double orphans (i.e. those who lost both their parents), were considerably more likely to interrupt their schooling than other children.104

But AIDS orphans are likely to suffer even higher levels of trauma as a result of watching parents endure prolonged, often painful periods of illness and depression prior to death. Social stigmatisation may produce even higher levels of discrimination and maltreatment.105

This may lead to higher levels of crime and other anti-social behaviour over the next two decades as this growing, and disproportionately large, share of children move through the 15 to 24 age range; the years of greatest propensity to commit crime.106 British and South African studies of children or young adults who had committed violent crimes, found that above average proportions of them had been orphaned, abandoned or rejected by parents or guardians.107


Democracy is the preferred political system in the world today. It is the only political regime that is based on, and designed to maximise, human equality and freedom. Southern Africa was not immune from the 'Third Wave' of democratisation that swept the world in the 1980s and 1990s. Beginning in 1990, multi-party systems with regular elections emerged in Namibia, Zambia, Malawi, South Africa, Lesotho, Mozambique and Tanzania. Recent political changes suggest that Angola may not be far behind.

However, in none of these countries can democracy be characterised as consolidated. 'Strong man' presidents and firmly entrenched ruling parties threaten political pluralism in the face of weak legislatures, weak party systems, and weak civil societies. Yet with the important exception of Zimbabwe, civil and political rights, and reasonably free and fair elections persist in most of these newly democratised states.

HIV/AIDS, however, threatens to block and even reverse democratic development across the region. Lost incomes, increasing health costs, shrinking tax bases, increased labour costs and decreasing productivity all conspire to threaten the economic growth necessary to sustain democratic practice in poor countries. Increasing death and illness in cabinets, legislatures and government ministries threatens the institutionalisation that young democracies need, to create the strong and effective states that give effect to the rules of democracy. Sharply decreasing adult life expectancy, and increasing proportions of people living with effective death sentences, removes incentives for large sections of the populace to participate in democratic politics or comply with the rules of the democratic state. Stigma, discrimination and conflict over scarce resources threaten to increase political conflict and criminal behaviour.

However, almost all of what has been outlined here is based on the assumption that in Southern Africa, HIV-infection is akin to a death sentence. Two developments may force a revision of these assumptions and all the projections based thereon. First, sharply decreased costs of anti-retroviral drugs. Or, increased political will by national leaders across the region to provide such drugs to their citizens may significantly mitigate the economic, social and political consequences of HIV infection. Second, bold and imaginative prevention campaigns may convey to the uninfected how to avoid infection, thus reversing the cycle of gloom and despair.

Finally, much of what is presented above is based on logic and conjecture, rather than fact. Little is known about why or how children, citizens, elites and institutions infected, affected or threatened by HIV/AIDS change their social and political behaviour. Clearly, much more research is needed. The spread and extent of the pandemic across the southern African region and within individual countries should provide sufficient evidence to support comparative cross-sectional and longitudinal research by inquisitive and dedicated researchers. This paper has been written in an attempt to lay out a research matrix into which prospective research can be slotted, and hopefully entice researchers to translate conjecture into testable hypotheses.


An earlier version of this paper was presented to a University of Natal Health Economics and AIDS Research Unit (HEARD), University of Cape Town Democracy in Africa Research Unit, and Institute for Democracy Governance and AIDS Programme Workshop on 'Democracy and AIDS in Southern Africa: Setting the Research Agenda', Cape Town, 22-23 April 2002. For proceedings, see <>.

See for example, A Hamoudi and J Sachs, The Economics of AIDS in Africa, Centre for International Development, Cambridge, MA, undated; S Freudenthal, A Review of Social Science Research on HIV/AIDS, Unpublished paper, Sida/SAREC, November 2001.

For an examination of the political implications of economic impacts, see W Parker, U Kistner, S Gelb, K Kelly and M O'Donovan, The Economic Impact of HIV/AIDS on South Africa and its Implications for Governance: A Literature Review, USAID, Washington DC, 2000, p 1. For projections on the impact of HIV/AIDS on security forces and regime stability, see L Heinecken, Living in Terror: The Looming Security Threat to Southern Africa, African Security Review 10(4), 2001, pp 7-18; R B Cheek, Playing God With HIV: Rationing HIV Treatment in Southern Africa, African Security Review 10(4), 2001, pp 19-28; P Fourie and M Schönteich, Africa's New Security Threat: HIV/AIDS and Human Security in Southern Africa, African Security Review 10(4), 2001, pp 29-44. For initial evidence on impacts in the education sector, see P Bennell, K Hyde and N Swainson, The Impact of the HIV/AIS Epidemic on the Education Sector in Sub-Saharan Africa, Centre for International Education, University of Sussex Institute for Education, Sussex, February 2002.



A recent literature review concludes that while there is a growing, though unorganised, body of theoretically informed speculation about the impact of HIV/AIDS, there is virtually no body of substantive evidence. See R Manning, AIDS and Democracy: What Do We Know?, Paper presented to a University of Natal Health Economics and AIDS Research Unit (HEARD), University of Cape Town Democracy in Africa Research Unit, and Institute for Democracy Governance and AIDS Programme Workshop on 'Democracy and AIDS in Southern Africa: Setting the Research Agenda', Cape Town, 22-23 April 2002.

T Barnett and A Whiteside, AIDS in the Twenty First Century: Disease and Globalisation, Hampshire and New York, Palgrave Macmillan, 2002, p 48.

R Pharaoh and M Schönteich, AIDS, Security and Governance in Southern Africa: Exploring the Impact, ISS Paper 65, January 2003, p 1.

Barnett and Whiteside, op cit, p 48.

No author, AIDS and Violent Conflict In Africa, Peace Watch 8(4), United States Institute of Peace, Washington DC, June 2001.

A Sen, Democracy as a Universal Value, Journal of Democracy 10(3), 1999, pp 3-17.

A Przeworski, M Alvarez, J Antonio Cheibub and F Limongi, Democracy and Development: Political Institutions and Well-Being in the World, 1950-1990, Cambridge University Press, Cambridge, 2000, p 47.

L Diamond, Introduction, in: L Diamond & M Plattner (eds) Democratization In Africa, Johns Hopkins University Press, Baltimore, 1999.

See L Diamond, Is the Third Wave Over?, Journal of Democracy 7(3), July 1996, pp. 20-27;and R Joseph, Africa, 1990-1997: From Abertura to Closure, Journal of Democracy 9(2), April 1998, pp 3-17.

A Karatnycky, The Freedom House Survey, Journal of Democracy, 13(1), January 2002, pp 99-112; R Mattes and A Leysens, Southern Africa, in: E Kolodziej (ed) A Force Profound? The Politics, Power, and Promise of Human Rights, University of Pennsylvania Press, Philadelphia, (forthcoming), 2003.

Przeworski et al, op cit, pp 117-122.

J Linz and A Stepan, Towards Consolidated Democracies, Journal of Democracy 7(2), April 1996, pp 14-33.

S M Lipset, Some Social Requisites of Democracy: Economic Development and Political Legitimacy, American Political Science Review 53(1), 1959, pp 69-105.

Przeworski et al, op cit, chapter 2.

In using Przeworki et al's results to assess life expectancy of southern Africa's democracies, an important caveat is in order. No country in continental southern Africa would qualify as a democracy under Przeworski et al's criteria, which include at least one turnover in government. However, I am proceeding on the basis that had they coded Botswana as a democracy in their data set of countries from 1950 to 1990, the results would have been essentially the same, and thus can be used to assess the life expectancy of regular elections and multi-party competition.

Przeworski et al, op cit, chapter 2.

L Diamond and J Linz, Introduction: Politics, Society and Democracy in Latin America, in: L Diamond, J Linz, and S L Lipset (eds.), Democracy in Developing Countries: Latin America, Lynne Reiner, Boulder, 1989, pp 1-58.

Przeworski et al, op cit, chapter 2.

Ibid, pp 109-111.

World Development Report, 2000/2001, Oxford University Press, Washington D.C., 2001, pp 274-275.

The Gini index measures inequality over the entire distribution of income or consumption. A value of 0 represents perfect equality, and a value of 1 perfect inequality.

Przeworski et al, op cit, pp 120-121.

World Development Report, 2000/2001, op cit, pp 282-283.

A Whiteford and D E Van Seventer, Winners and Losers: South Africa's Changing Income Distribution in the 1990s, WEFA, Johannesburg, 1999, pp 3 and 11-19.

M Over, The Macroeconomic Impacts of AIDS in Sub-Saharan Africa, Technical Paper 3, World Bank, Washington DC, 1992, cited in: Parker et al, op cit, p19.

K Quattek and T Fourie, The Economic Impacts of AIDS in South Africa: A Dark Cloud On the Horizon, ING Barings, Johannesburg, April 2000, p 21.

Arndt and Lewis, in: Parker et al, op cit, p 17. Parker et al note a recent trend toward cautious optimism, however, citing cross national studies that find no evidence of slower growth in economies where HIV/AIDS is advanced (Bloom & Mahal, 1995). Indeed, modelling the impacts of HIV/AIDS is so difficult because the disease is different from other epidemics. Parker et al, op cit, pp 10-12.

N Nattrass, AIDS, Growth and Distribution in South Africa, CSSR Working Paper 7, Centre for Social Science Research, University of Cape Town, Cape Town, March 2002, p 5.

Ibid, p 1.

Ibid, pp 12-21; C Kennedy, From the Coalface: A Study of the Response of a South African Colliery to the Threat of AIDS, CSSR Working Paper 5, Centre for Social Science Research, University of Cape Town, Cape Town, April 2002.

B Beresford, The Cost of Doing Nothing, Mail & Guardian, 17 to 23 May 2002.

Nattrass, op cit, pp 6-12.

A de Waal, Modelling the Governance Implications of the HIV/AIDS Pandemic in Africa, AIDS and Governance Discussion Paper 2, Unpublished draft, March 2002.

Ibid, p 9.

L Ensor, AIDS To Outstrip SA's Future Health Budgets, Business Day, 10 October 2001.

Cheek, op cit, p 22.

B Beresford, Failing to Deliver, Mail & Guardian, 11 to 17 October 2002.

L Clarke, Study Highlights Ravages of HIV/AIDS, Sunday Independent, 3 November 2002.

No author, New Priorities Needed, Business Day, 10 October 2001.

Heinecken, op cit, p 10; Cheek, op cit, p 20.

No author, Comparing Thailand, Brazil, South Africa, Botswana and Uganda, Budget Watch, Idasa, Cape Town, November 2002, p 8; No author, Focus on Brazil, Thailand, Uganda, Botswana?, Budget Watch, Idasa, Cape Town, November 2002), p 2.

Proceedings, Democracy in AIDS in Southern Africa: Setting the Research Agenda, op cit.

Ensor, op cit.

Cheek, op cit, p 22.

Heinecken, op cit, p 10.

Cheek, op cit, p 21.

G Hiwa, Malawi Ministry of Health, cited in: Cheek, op cit, p 22.

N Nattrass and J Skordis, Paying to Waste Lives: The Affordability of Reducing Mother to Child Transmission of HIV in South Africa, CSSR Working Paper 4, Centre for Social Science Research, University of Cape Town, Cape Town, December 2001, p 2. Also see Nathan Geffen, Nicoli Nattrass & Chris Raubenheimer, The Cost of HIV Prevention and Treatment Interventions in South Africa, CSSR Working Paper 28, Centre for Social Science Research, University of Cape town, Cape Town, January 2003.

Focus on Brazil, Thailand, Uganda, Botswana?, op cit, p 2.

Ibid, p 4.

Ibid, p 4.

L Bollinger, J Stover, R Kerkhoven, G Mutangadura and D Mukurazita, The Economic Impact of AIDS in Zimbabwe, POLICY Project, Washington DC, 1999, <>, cited in: J Youde, All the Voters will be Dead: HIV/AIDS and Democratic Legitimacy and Stability in Africa, International Foundation for Electoral Assistance, Stockholm, 2001, p 26.

Youde, op cit, p 27.

Quattek and Fourie, pp 1, 7 and 10.

No author, AIDS Could Cripple Public Service, Business Day, 29 June 2001; Youde, op cit, p 19.

Youde, op cit, p 8.

R Manning, The Impact of HIV/AIDS on Local Democracy: A Case Study of the eThekwini Municipality, KwaZulu-Natal, South Africa, CSSR Working Paper, Centre for Social Science Research, University of Cape Town, Cape Town, forthcoming, 2003, pp 25-26.

Ibid, pp 11 and 22.

L Ensor, Treasury Sounds the Alarm Over AIDS, Business Day 10 October 2001.

Cheek, op cit, p 22.

Manning, op cit, p 20.

Ibid, p 22.

Ibid, pp 12-13 and 18-19.


M Bratton and N van de Walle, Democratic Experiments In Africa, Cambridge University Press, Cambridge, 1997.

See E Costarelli, HIV/AIDS, Democracy and Citizenship, Electoral Institute of Southern Africa, Johannesburg, pp 8-10.

Youde, op cit, pp 10-20.

R Mattes and H Thiel, Consolidation and Public Opinion In South Africa, Journal of Democracy 9(1), January 1998, pp 95-110.

In South Africa average household expenditure on health is approximately 4% of total income. However, a survey of 771 AIDS affected households in four provinces found that they were spending on average a third of their income on health care, rising to over half of all income in rural households. Half of the families had paid for a funeral in the previous twelve months, usually consuming more than three months income. The vast majority of households were poverty stricken with an average household income of under R1,000 per month. Two-thirds reported that they had lost some income due to AIDS. See B Beresford, Families Tipped Into Destitution, Mail & Guardian, 27 September to 3 October 2002; and B Beresford, Failing to Deliver, Mail & Guardian, 11 to 17 October 2002; and L Clarke, Study Highlights Ravages of HIV/AIDS, op cit.

Fourie and Schönteich, op cit, p 31.

De Waal, op cit, p. 9.

A Whiteside and D FitzSimons, The AIDS Epidemic: Economic, Political and Security Implication, Conflict Studies 251, Research Institute for the Study of Conflict and Terrorism, May 1992, p 26.

De Waal, op cit, p 8.

J Brower and Chalk, The Security Implications of Infectious Disease and Its Impact on National and International Institutions and Policies, in America's Real Achilles Heal: The Threat of Infections Disease to National and International Security, RAND, Santa Monica, 2002, p 6, cited in: Manning, op cit, p 14; Youde, op cit, p 5.

Youde, op cit, pp 10-20. For a discussion of ways to empower HIV-affected people to vote, see E Costarelli, op cit, pp. 8-10.

R Manning, The Impact of HIV/AIDS on Civil Society: Assessing and Mitigating Impacts: Tools & Models for NGOs and CBOs, Health Economics and HIV/AIDS Research Division, University of Natal, Durban, 2002, pp 14-26.

Manning, The Impact of HIV/AIDS on Civil Society, op cit, pp 16 and 29.

D Gordon, Plague Upon Plague: AIDS and Violent Conflict in Africa, cited in: Manning, AIDS and Democracy: What Do We Know?, op cit.

A Whiteside, R Mattes, S Willan and R Manning, Examining HIV/AIDS In Southern Africa Through the Eyes of Ordinary Southern Africans, CSSR Working Paper 11, Centre for Social Science Research, University of Cape Town, Cape Town, 2002, pp 36-37.

Manning, The Impact of HIV/AIDS on Civil Society, op cit, p 1, pp 10-11.

De Waal, op cit, p 8.

R Shell, Halfway to the Holocaust: The Economic, Demographic and Social Implications of the AIDS Pandemic to the Year 2010 in the Southern African Region, HIV/AIDS: A Threat to the African Renaissance? Konrad Adenauer Stiftung Occasional Papers, Kondrad Adenauer Stiftung, Parktown, June 2000, pp 19-20. On the relationship between frustration and violence, see T R Gurr, Why Men Rebel, Princeton University Press, Princeton, 1970.

Nattrass, op cit, p 1.

De Waal, op cit, p 10.

Cheek, op cit, pp 24-25.


Ibid, pp 25-27.

Ibid, p 24.

J Gibson and A Gouws, Overcoming Intolerance in South Africa: Experiments in Persuasion, Cambridge University Press, Cambridge, 2002; R Mattes, C Africa and H Taylor, Intolerance and Intimidation in South Africa, Opinion '99 Series 2(9), Idasa, Cape Town, 1999, <>. For South Africans' attitudes toward ordinary and skilled immigrants, see R Mattes, D Taylor, D McDonald, A Poore and W Richmond, South African Attitudes To Immigrants and Immigration, in: D McDonald (ed.), On Borders: Perspectives On International Migration In Southern Africa, St. Martin's Press, New York, 2000), pp196-218; and R Mattes, J Crush and W Richmond, The Brain Gain and Legal Immigration to Post-Apartheid South Africa, Africa Insight 30(2), October 2000, pp 21-30.

Beresford, Families Tipped Into Destitution, op cit; Beresford, Failing to Deliver, op cit.

Bennell, Hyde and Swainson, op cit, p xi.

Fourie and Schönteich, op cit, p 38.


S Hunter and J Williamson, Children on the Brink: Strategies to Support A Generation Isolated by HIV/AIDS, UNICEF/USAID, New York, June 2000, cited in: Bennell, Hyde and Swainson, op cit, p 48.

Cited in Fourie and Schönteich, op cit, p 38.

Bennell, Hyde and Swainson, op cit, pp 66 and 68.

Cited in Fourie and Schönteich, op cit, p 38.

J A Smetherham, 33% will be AIDS Orphans by 2015, says MRC Report, Cape Times, 18 June 2002.

Cited in Fourie and Schönteich, op cit, p 38.

The Progress of Nations, UNICEF, New York, 2000, cited in: Bennell, Hyde and Swainson, op cit, p 58.

Bennell, Hyde and Swainson, op cit, p 59.

Fourie and Schönteich, op cit, p 39.

M Schönteich, Age and AIDS: South Africa' Crime Time Bomb?, African Society Review 18(4), 1999; Fourie and Schönteich, op cit, p 40.

Fourie and Schönteich, op cit, p 40. However, for a recent comprehensive review of the evidence supporting current thinking about the social and political impact of growing numbers of AIDS orphans, see Rachel Bray, Predicting the Social Consequences of Orphanhood in South Africa, CSSR Working Paper, Centre for Social Science Research, University of Cape Town, forthcoming, 2003.


Across Southern Africa HIV/AIDS threatens to block and even reverse democratic consolidation. Lost incomes, higher healt and labour costs, shrinking tax bases and decreasing productivity undermine the economic growth necessary to sustain democratic government in the countries of Southern Africa.

Sustainable democracies require a professional civil service and strong, viable and autonomous courts, legislatures, executives and electoral systems at national and local levels. AIDS related death and illness among cabinets ministers, legislators and public officials threaten the political institutions that are necessary to give effect to the rules of democracy. Sharply reduced adult life expectancy, and increasing proportions of people living with AIDS, remove incentives for large sections of the populace to participate in democratic politics or comply with the rules of the democratic state.


ROBERT MATTES is associate professor of Political Studies, and director of the Democracy in Africa Research Unit in the Centre for Social Science Research, University of Cape Town. He is also cofounder and co-director of the Afrobarometer, a regular cross-national survey of citizens' attitudes to democracy, markets and civil society in selected African countries..