Our health in brief - part four: How will the health system cope over the next four years?
This is the fourth of a four brief series on health expenditure in the public and private sectors. The first brief deals with commonly ignored components of health expenditure: provision in workplaces, and medical expenditure on health financed from the Compensation Fund and the Road Accident Fund. The second considers the impact of local government on health. The third brief sets out information on the pattern of health expenditure from 1 April 2019 to the 31 March 2020. This forms the baseline for examining the implications of the 2021 Budget for health expenditure over the medium term, the topic of this brief.
Introduction
Given the fiscally constrained medium term projections in the 2021 Budget, this brief explores the financing of health care expenditure and its implication for service delivery up until the 2023/24 financial year. The assumptions and projections in the Budget are used throughout the table below. Once the baseline projection has been presented, a brief discussion follows of what might happen if the assumptions are violated.
The analysis
The analysis proceeds by means of a row by row discussion of the table. All values are in current prices, unless otherwise stated.
Panel A
Rows 1 to 5 present successive estimates and projections from the 2019 to the 2021 Budgets, including the 2019 and 2020 Medium Term Budget Policy statements. Row 6 extracts current estimates and projections. Rows 7 and 8 add in medical expenditure financed by the Compensation Fund and Row 9 presents aggregate expenditure in the public sector. Row 10 converts Row 9 into an index.
Panel B
Panel B presents an economic and demographic framework for what follows. Row 11 presents the GDP deflator, which permits conversion of GDP in current prices to GDP in constant prices. Row 12 presents GDP itself in billions of rand and Row 13 presents compensation of government employees in billions of rand and Row 14 converts Row 13 into an index. Row 15 presents United Nations estimates of the South African population in thousands and Row 16 converts Row 15 into an index.
Panel C
Panel C calculates per capita expenditure on health in the public sector. Row 17 calculates average aggregate health expenditure per capita in current prices. Row 18 presents the consumer price index and Row 19 calculates aggregate health expenditure in constant 2020/21 prices. Row 20 converts Row 19 into an index.
Panel D
Row 20 presents too pessimistic an indicator of the quantity of health services provided in the public sector, because the intention of the 2021 Budget is to hold employee compensation costs to below inflation. So a unit cost index is computed in Row 21, assuming employee costs will evolve according to the compensation index in Row 14 and other costs according to the GDP deflator in Row 11. Row 22 rebases Row 21 from 2018/19=100 to 2020/21=100. Row 23 then calculates a quantity index, based on per capita expenditure in current prices from Row 17 and the unit cost index in Row 22. This indicates that the quantity of health services per capita will drop by 5.9% between 2020/21 and 2023/24.
Panel E
Panel E takes the analysis forward by considering the gap between the financing needed to keep the quantity index in Row 23 constant and the projected financing. Row 24 calculates the gap. Row 25 calculates the impact on the gap associated with spreading resources more thinly by not allowing expenditure to rise with population. This is bound to happen on projection assumptions, as the system battles to continue to afford the resources it currently has. This by itself does not close the gap, as Row 26 indicates. The only other source of savings would be to squeeze some of the rent out of the costs of procurement by the public health system, no easy task as spending on COVID-19 attests. A saving of 4% in non-employee costs (see Rows 27 and 28) would close the gap in 2021/22 and 2022/23, but would still leave a gap in 2023/24, as Rows 24, 29 and 30 show.
Panel F
Panel F presents estimates and projections of household consumption expenditure. The index in Row 31 is taken from the 2021 Budget projection and Row 32 calculates household consumption expenditure itself.
Panel G
Panel G presents estimates and projections of the components of private health expenditure (Rows 33 to 37), all projected to grow at the same rate as household consumption expenditure. Row 38 presents total private health expenditure and Row 39 displays the percentage of total health expenditure provided by the private sector. By 2022/23, the percentage recovers to its 2019/20 level of 51.3% and in 2023/24 it rises further to 52.8%.
Panel H
Panel H presents estimates and projections of health expenditures across beneficiaries: all (Row 40) for expenditures benefiting the entire population, medical aid beneficiaries (Row 41), the medically insured (Row 42), and those who are neither medical aid beneficiaries nor medically insured (Row 43). Row 44 presents aggregate health expenditure for the system as a whole. Row 45 presents aggregate health expenditure as a percentage of GDP. It rises from 9.0% in 2019/20 to 9.8% in 2020/21 (because of the contraction in the denominator) and then declines again to 9.0% in 2023/24.
General interpretation
Given the projection assumptions, the health system will hold together, but it will be under increased strain. In the public sector, it will not be possible to increase the size of the provision to compensate for a rising population, and if value for money in procurement in not improved, services will have to be cut. In the private sector, continuing rises in medical aid premiums above the inflation rate will press against modest rises in household consumption expenditure, prompting some to trade down in options provided by medical aid, or even to cancel it altogether.
What happens if the projection assumptions are violated?
The principal risk is that the unions manage to punch a hole in the government’s wage restraint policy. If this happens, the unit cost index will rise. It may be that the government will accommodate the change in employee compensation costs fully in future medium term expenditure projections, increasing health expenditure projections accordingly. With such accommodation, there would be no change in the quantity index. With less than full accommodation, the strain on public health provision would increase.
The second risk is that economic growth comes in below or above the projection in the 2021 Budget. We must hope that the latter is the case. If the former happens, the health sector is in for a very difficult time.
|
Panel |
|
Row |
2018/19 |
2019/20 |
2020/21 |
2021/22 |
2022/23 |
2023/24 |
|
|
|
|
R billion |
|||||
|
|
|
|
|
|
|
|
|
|
|
A |
Government health expenditure |
|
|
|
|
|
|
|
|
|
2019 Budget |
1 |
208.8 |
222.6 |
238.8 |
255.5 |
|
|
|
|
2019 MTBPS |
2 |
209.7 |
222.7 |
238.5 |
257.2 |
272.9 |
|
|
|
2020 Main Budget |
3 |
|
222.0 |
229.7 |
244.0 |
257.6 |
|
|
|
2020 MTBPS |
4 |
|
219.1 |
226.2 |
235.3 |
242.0 |
246.3 |
|
|
2021 Budget |
5 |
|
|
247.0 |
248.8 |
245.9 |
245.0 |
|
|
Best projection |
6 |
209.7 |
219.1 |
247.0 |
248.8 |
245.9 |
245.0 |
|
|
Compensation Fund |
7 |
2.5 |
2.7 |
3.0 |
3.5 |
3.9 |
4.3 |
|
|
Road Accident Fund |
8 |
3.6 |
3.4 |
3.8 |
4.4 |
4.9 |
5.5 |
|
|
Best projection with CF and RAF |
9 |
215.8 |
225.2 |
253.9 |
256.6 |
254.7 |
254.8 |
|
|
Best projection Index (2018/19=100) |
10 |
100.0 |
104.4 |
117.6 |
118.9 |
118.0 |
118.1 |
|
B |
Prices and population |
|
|
|
|
|
|
|
|
|
GDP deflator |
11 |
100.0 |
104.0 |
108.5 |
112.7 |
117.4 |
122.5 |
|
|
GDP current prices |
12 |
4924.0 |
5152.3 |
5007.2 |
5350.2 |
5671.8 |
6021.8 |
|
|
Budget: compensation of employees |
13 |
584.4 |
623.8 |
637.0 |
650.4 |
656.0 |
659.3 |
|
|
Compensation of employees index |
14 |
100.0 |
106.7 |
109.0 |
111.3 |
112.3 |
112.8 |
|
|
Population ('000s) |
15 |
57793 |
58558 |
59309 |
60042 |
60756 |
61453 |
|
|
Population index (2018/19=100) |
16 |
100.0 |
101.3 |
102.6 |
103.9 |
105.1 |
106.3 |
|
C |
Per capita expenditure - public sector |
|
|
|
|
|
|
|
|
|
Current prices (Rand) |
17 |
3734 |
3846 |
4281 |
4274 |
4192 |
4146 |
|
|
Consumer price index (2018/19=100) |
18 |
100.0 |
103.0 |
114.6 |
114.5 |
112.3 |
111.0 |
|
|
Constant 2020/21 prices (Rand) |
19 |
4051 |
4012 |
4281 |
4115 |
3874 |
3672 |
|
|
Constant price index (2020/21=100) |
20 |
|
|
100.0 |
96.1 |
90.5 |
85.8 |
|
D |
Unit costs - public sector |
|
|
|
|
|
|
|
|
|
Unit cost index (2018/19=100) |
21 |
100.0 |
103.4 |
112.2 |
113.8 |
114.3 |
115.5 |
|
|
Unit cost Index (2020/21=100) |
22 |
89.1 |
92.1 |
100.0 |
101.4 |
101.8 |
102.9 |
|
|
Revised quantity index (2020/21=100) |
23 |
|
|
100.0 |
98.5 |
96.2 |
94.1 |
|
Panel |
|
Row |
2018/19 |
2019/20 |
2020/21 |
2021/22 |
2022/23 |
2023/24 |
|
|
|
|
R billion |
|||||
|
|
|
|
|
|
|
|
|
|
|
E |
Gap analysis(current prices) |
|
|
|
|
|
|
|
|
|
Gap (R billion) |
24 |
|
|
|
3.9 |
10.1 |
15.9 |
|
|
Spread more thinly |
25 |
|
|
|
3.2 |
6.1 |
8.9 |
|
|
Remaining gap |
26 |
|
|
|
0.7 |
4.0 |
7.1 |
|
|
Limit on non-compensation costs cut |
27 |
|
|
|
4.0% |
4.0% |
4.0% |
|
|
Per cent cut in non-compensation costs |
28 |
|
|
|
0.7% |
4.0% |
4.0% |
|
|
Cut in non-compensation costs |
29 |
|
|
|
0.7 |
4.0 |
4.0 |
|
|
Remaining gap |
30 |
|
|
|
0.0 |
0.1 |
3.1 |
|
F |
Household consumption expenditure |
|
|
|
|
|
|
|
|
|
HCE index (2020=100) |
31 |
|
|
100.0 |
102.9 |
105.4 |
107.5 |
|
|
HCE R billion |
32 |
3958.3 |
4129.9 |
4100.8 |
4383.1 |
4675.4 |
4976.1 |
|
G |
Private health expenditure |
|
|
|
|
|
|
|
|
|
Medical aid expenditure |
33 |
172.2 |
185.9 |
184.6 |
197.3 |
210.5 |
224.0 |
|
|
Medical aid out of pocket |
34 |
16.3 |
17.0 |
16.9 |
18.0 |
19.2 |
20.5 |
|
|
Medical insurance |
35 |
9.4 |
9.4 |
9.3 |
10.0 |
10.6 |
11.3 |
|
|
Other private expenditure |
36 |
23.8 |
23.8 |
23.6 |
25.3 |
26.9 |
28.7 |
|
|
Workplace expenditure |
37 |
0.9 |
0.9 |
0.9 |
1.0 |
1.0 |
1.1 |
|
|
Total |
38 |
222.6 |
237.0 |
235.3 |
251.5 |
268.3 |
285.6 |
|
|
Share of privately provided |
39 |
50.8% |
51.3% |
48.1% |
49.5% |
51.3% |
52.8% |
|
H |
Health expenditure by category |
|
|
|
|
|
|
|
|
|
All |
40 |
8.0 |
8.3 |
9.4 |
9.5 |
9.4 |
9.4 |
|
|
Medical aid beneficiaries |
41 |
188.5 |
202.9 |
201.5 |
215.3 |
229.7 |
244.5 |
|
|
Medically insured |
42 |
9.4 |
9.4 |
9.3 |
10.0 |
10.6 |
11.3 |
|
|
Others |
43 |
232.5 |
241.6 |
269.1 |
273.4 |
273.2 |
275.2 |
|
|
Total |
44 |
438.4 |
462.2 |
489.2 |
508.1 |
523.0 |
540.4 |
|
|
Medical aid, insurance (per cent of total) |
45 |
46.0% |
46.8% |
43.9% |
45.2% |
46.8% |
48.2% |
|
I |
Health expenditure as per cent of GDP |
46 |
8.9% |
9.0% |
9.8% |
9.5% |
9.2% |
9.0% |