Health Expenditure and Finance in South Africa¶
McIntyre, Di; Bloom, Gerald; Doherty, Jane; Brijlal, Prem (1995). Health Expenditure and Finance in South Africa. Health Systems Trust and the World Bank.
| Type | report |
| Scope | Wider apartheid literature |
| Method | policy analysis |
| Period studied | 1982–1995 |
| Regions | South Africa, former homelands (TBVC states and self-governing territories), nine post-1994 provinces |
| Themes | inequality poverty unemployment, macro fiscal finance debt, transition post apartheid, bantustans removals influx, demography urbanisation, methods data sources |
| Citation key | mcintyre1995a |
Argument¶
The Health Expenditure Review maps who finances and who receives health spending in South Africa at the end of apartheid. It argues that the country spends a high share of GDP on health (8.5%), yet resources are distributed grossly unequally between the private and public sectors, between hospital and primary care, and between rich and poor (mostly African) districts. This inheritance of apartheid administration produces excess morbidity and premature death among the poor. Because low growth, debt service and deficit pressure leave little room for more tax-funded spending, the report argues that extending primary care (an extra R1.5-2.5 billion a year) must be paid for mainly by redistribution, hospital efficiency savings, cost recovery and managed transition. It is deliberately descriptive and not policy-prescriptive, and builds its case from the ReHMIS facility survey, a private-sector expenditure review and commissioned technical papers.
Evidence and method¶
ReHMIS (Regional Health Management Information System) facility-level survey of every public health facility (1992/93 expenditure, beds, staff, workload); commissioned technical reports (Valentine and McIntyre on private-sector spending, Deloitte and Touche on capital projects and donors, Price and Broekmann on hospital expenditure trends, Bunting on training, Blecher and McIntyre on research); Registrar of Medical Schemes data; SARB and CSS statistics; 1991 census and magisterial-district income data (van Wyk 1985); the 1993/94 SALDRU living standards survey (preliminary); mortality analysis by Bourne; World Bank comparators; and costing studies of clinics and hospitals.
Place in the debates¶
- Public versus private financing and the role of social health insurance (see Living standards, health and social policy). Neutral and descriptive. It documents the cost spiral in medical schemes and the tax subsidy to them, and treats SHI as an option that needs cost-control safeguards, without recommending it.
- Balance between hospital-based curative care and primary care (resource reallocation versus new money) (see Living standards, health and social policy). Argues the public system is heavily biased to tertiary and academic hospitals and that much of the primary care expansion must be funded by hospital savings. It also accepts that some government and donor 'gap-funding' is needed.
- Fiscal constraint and the social spending trade-off in the transition (macro context) (see Living standards, health and social policy). Takes slow growth, a deficit of 7.8% of GDP and rising debt service as binding limits on health budgets, so health must compete with housing, education and welfare.
Findings¶
-
Health spending was high by international standards, but accessibility and quality were very unequal along racial and income lines. (p. 11)
South Africa spent over 30 billion rands on health services in 1992/93. This amounted to 8.5 percent of its Gross Domestic Product (GDP).
-
Sluggish growth, employment decline and population growth lie behind the fiscal and welfare constraints on health provision. (p. 21)
Since the early 1980s its GDP increased by less than 1.5 percent a year while its population grew by 2.5 percent
-
Apartheid-era racial earnings inequality was very large. (p. 25)
Fallon and da Silva (1994) estimate that, in 1987, Whites earned 9.5 times more per person than Africans.
-
Income inequality is extreme, with Gini estimates between 0.54 and 0.65. (p. 22)
Income-based coefficients for other upper middle income countries range between 0.45 and 0.63, making South Africa one of the most unequal of societies
-
Public health spending was skewed to rich, mainly White districts and away from the poorest, mainly African districts. (p. 37)
Only 26.8 percent of the public sector health budget was spent directly on the poorest 40 percent of magisterial districts, which contain approximately 49 percent of the population
-
Per-person spending on members of private medical schemes was about fifteen times that on residents of the poorest districts. (p. 40)
The ratio of spending on health services for members of medical aid schemes to the average spending by the public health services in the poorest 75 districts was almost 15:1.
-
Public real health expenditure per capita was nearly flat for a decade, so the budget pressure dates from the late-apartheid fiscal squeeze and austerity. (p. 62)
Recurrent expenditure on public health services increased at an annual rate of 18.3 percent a year between 1983/84 and 1992/93, from R2.5 billion to R11.1 billion
-
Government deficit and debt service sharply limit prospects for tax-funded health spending in the transition. (p. 28)
the budget deficit was 27.4 billion rands in 1992/93, or 7.8 percent of GDP
-
Public hospital spending was concentrated in tertiary and academic hospitals at the expense of primary care. (p. 65)
approximately 76 percent of the total was spent on acute care hospitals, 5 percent on chronic hospitals, 11 percent on non-hospital primary health care
-
Medical scheme contributions rose faster than salaries, reflecting a private-sector cost spiral and, via African unionisation, a changed membership base. (p. 52)
Medical scheme contributions were equivalent to 7.1 percent of average formal sector salaries in 1982. Ten years later, in 1992, they amounted to 15.2 percent of average salaries.
-
Closing the primary care gap would cost an additional R1.5-2.5 billion a year in recurrent spending. (p. 98)
recurrent spending on non-hospital primary level services will have to increase by at least 1.5 billion rands, and possibly by as much as 2.5 billion rands
-
Growth in African unionisation after legalisation in 1979 raised real wages and formal benefits such as medical cover. (p. 26)
One result of the increasing organisation of the African workforce has been a rise in real wages despite the recession of the 1980s.
Evidence ledger¶
Sources used¶
| Source | Kind | Period | How used |
|---|---|---|---|
| ReHMIS (Regional Health Management Information System) survey database, 1992/93 | author survey or fieldwork | 1992/93 | Principal source for public sector expenditure, facilities, personnel and workload, distributed by province, level of care and income quintile of magisterial district. |
| Valentine and McIntyre (1994), survey of private sector health expenditure (HER Technical Paper No. 9) | author survey or fieldwork | 1992/93 | Establishes private sector finance by source, medical scheme and insurance beneficiaries, hospital data and industry health spending. |
| Registrar of Medical Schemes (1993), Report for year ending 31 December 1993 | commission or official report | 1982-1993 | Time series of medical scheme membership by race, expenditure per beneficiary, contributions and dispensing practitioners. |
| Central Statistical Service, South African Statistics 1993 (CPI and salary data) | official statistical series | 1982-1993 | Provides CPI deflators and average formal sector salaries for real-terms scheme expenditure and contributions. |
| South African Reserve Bank estimates of private consumption expenditure on medical goods and services (1991 and 1994 publications) | national accounts macro | 1983/84-1992/93 | Official routine series shown to understate private health spending by 70 percent relative to the review's own estimate. |
| Chetty (1994), integrated analysis of health facilities (HEU Working Paper No. 13) | secondary literature compilation | 1992-1993 | Supplies hospital bed counts, public/private bed shares and facility needs (additional clinics); used also to validate ReHMIS. |
| Development Bank of Southern Africa (1994), South Africa's nine provinces: A human development profile | secondary literature compilation | 1992-1994 | Source of provincial population, income per capita, GDP shares, SAMDC register numbers and census-adjusted district population. |
| Population Census 1991: Geographic distribution of the Population (CSS 1993c) | population census | 1991 | Urban/rural population distribution and district organisation of population data. |
| Project for Statistics on Living Standards and Development (1994), SALDRU household survey 1993/94 (with Klasen and Doherty 1995 preliminary analysis) | household or labour survey microdata | 1993/94 | Profiles the poor, health-provider use by income quintile, malnutrition, and barriers to care from fees. |
| Bourne (1994), analysis of 1990 mortality data from death registration | official statistical series | 1990 | Potential years of life lost by cause and share of infectious/ill-defined deaths across income quintiles. |
| Department of National Health and Population Development, Health Trends in South Africa 1993 and 1990 nutrition report | official statistical series | 1989-1992 | Infant mortality, maternal mortality, life expectancy and malnutrition indicators by race. |
| World Bank (1993) World Development Report and World Bank (1994) Better Health in Africa | international trade or finance data | 1990-1992 | International benchmarks for health spending, beds, physicians, mortality and basic package costs. |
| Blecher and McIntyre (1994) research expenditure; Bunting (1994) training expenditure (HER Technical Papers 6 and 7) | author survey or fieldwork | 1991/92-1992 | Estimate health research and health personnel training expenditure to complete the national expenditure total. |
| Deloitte and Touche (1994a) capital projects documentation and (1994b) donor funding questionnaire survey (HER Technical Papers 3 and 4) | author survey or fieldwork | 1993/94-1995/96 | Planned public health capital projects and costs, clinic building commitments, and donor health expenditure (R145 million). |
| Price and Broekmann (1994), Trends in provincial hospital expenditure (HER Technical Paper 8) | commission or official report | 1984/85-1990/91 | Time series of real provincial hospital expenditure and unit costs per patient day. |
| McIntyre (1993) Health sector financing and expenditure in South Africa (HEU Working Paper 4) | secondary literature compilation | 1983/84-1992/93 | Public recurrent health expenditure series in nominal and real terms and sources of public finance. |
| Rispel and Behr (1992) health indicators / personnel data | secondary literature compilation | 1989/90 | Distribution of health personnel between public and private sectors. |
| Commission reports: Steinmetz Commission (1993), Melamet Commission (1994), Committee on uniform fee structure (1993) | commission or official report | 1993-1994 | Evidence on pharmaceutical shrinkage, medical scheme insolvency and public hospital fee schedules. |
| Medical Schemes Act (No. 72 of 1967) and Medical Schemes Amendment Act (No. 23 of 1993) | legislation gazette | 1967-1994 | Explains regulatory change affecting scheme structure, risk-rating and cost containment. |
| Department of Finance (1994) Budget Review, 22 March 1994; Fallon and da Silva (1994) World Bank economic data | national accounts macro | 1978-1994 | Government revenue, expenditure, deficit and GDP growth establishing fiscal constraints. |
| Costing studies: Broomberg et al (1992) Diepkloof Clinic; Van den Heever and Price (1994); Zwarenstein et al (1994); Beattie and van den Heever (1995); KaNgwane and Ciskei studies | secondary literature compilation | 1990-1995 | Unit costs and gap estimates used to cost extension of primary care and to model homeland primary care expenditure. |
| Chamber of Mines and other industry-based sources; mine hospital data | chamber of mines or recruiting records | 1992-1993 | Industry health service provision, e.g. 53 mine hospitals with 6,898 beds. |
New data¶
The Health Expenditure Review assembled a national database (ReHMIS) covering every public sector health facility in South Africa and ex-homelands for 1992/93 (beds, staff, workload, expenditure, fees), plus commissioned surveys of private sector expenditure, donor funding, capital projects, research and training, producing a total national health expenditure estimate of R30.15 billion.
Inference¶
Quantification: Descriptive tables.
Descriptive accounting of health expenditure by source, sector, level of care and region, using tabulations from the ReHMIS database sorted by province and by income quintile of magisterial district. Comparison across regions and international benchmarks, a short time series for private and public spending, and scenario-based costing of primary care expansion; no regression or causal analysis.
Key claims¶
| Claim | Basis | Strength |
|---|---|---|
| South Africa spent about R30.15 billion (8.5% of GDP) on health in 1992/93, with about 58-61% flowing through the private sector. (p. 19) | Triangulation of ReHMIS, Valentine and McIntyre private sector survey, Blecher and McIntyre, Bunting, and Deloitte and Touche donor data. | Partial data |
| Public health resources are highly unequally distributed: the poorest districts get far fewer doctors, health inspectors and rands per capita than the richest. (p. xi) | ReHMIS survey tabulated by income quintile of magisterial districts, ranked using van Wyk (1985 income). | Systematic data |
| Public sector spending is hospital-centred: acute hospitals took about 76% of recurrent spending and tertiary/academic hospitals 44%, versus 11% for non-hospital primary care. (p. 47) | ReHMIS expenditure data classified by an algorithm into levels of care; homeland spending allocated by estimation formulas. | Systematic data |
| Private medical scheme costs rose faster than inflation, with contributions rising from 7.1% to 15.2% of average salaries between 1982 and 1992. (p. 34) | Registrar of Medical Schemes reports with CSS CPI and salary data. | Systematic data |
| Real public recurrent health expenditure grew only about 3% a year (0.5% per capita) between 1983/84 and 1992/93, and provincial hospital spending fell in real terms in the 1980s. (p. 44) | McIntyre (1993) series deflated by CPI; Price and Broekmann (1994) for provincial hospitals. | Partial data |
| Members of medical aid schemes use health services costing about 15 times the public spending per person in the poorest districts. (p. 22) | Average medical aid contribution per beneficiary (Valentine and McIntyre) versus ReHMIS public expenditure in poorest districts. | Partial data |
| An additional R1.5-2.5 billion a year is needed for public primary care to meet RDP targets and basic service coverage. (p. 80) | Synthesis of third-party costing studies (van den Heever and Price; Zwarenstein et al; Beattie and van den Heever) plus ReHMIS visit-per-capita shortfalls and unit cost of R20 per visit. | Illustrative or anecdotal |
| Health outcomes are far worse for Africans than Whites (infant mortality roughly seven times higher), though mortality data are unreliable and understate the burden. (p. 16) | DNHPD health statistics, Rispel and Behr, Bourne's mortality database from death registration; author caveats on under-registration. | Partial data |
The author's caveats¶
The authors state that ReHMIS data were collected and computerised without prior validation and were only partially validated; KwaNdebele clinics are missing; homeland expenditure had to be estimated (Transkei, Qwa-Qwa, health-ward allocation formulas). Private sector figures and household out-of-pocket spending are acknowledged underestimates, and capital, training and donor spending are under-reported. Mortality data exclude homelands and about 40 percent of African deaths are unregistered. Magisterial district income rankings rest on 1985 data, and district averages miss poor communities in rich districts. Unit-cost comparisons depend on an assumed 3:1 inpatient/outpatient cost ratio.
Gaps in the evidence¶
The argument lacks utilisation- and need-based data (morbidity surveys, quality of care) to show whether spending differences translate into outcome differences, and has no regional data on medical scheme membership to net private users out of the public denominator. Time series of public spending by district and level of care are missing, so changes in apartheid-era allocation cannot be traced beyond a national total and a 1984-1990 provincial series. Facility-level ReHMIS data and household surveys later became available and could test these claims with proper validation, but ReHMIS itself was not validated at publication.
Replicability: Data described but not public.