Trends in Public Spending on Health

47 downloads 0 Views 91KB Size Report
Appendix 1 Utilisation of Government Hospital Services .... Reddy K.N. and V.Selvaraju (1994), Health Care Expenditure by Government in India, Seven Hills.
TRENDS IN PUBLIC SPENDING ON HEALTH IN INDIA

V.Selvaraju Vinod B.Annigeri* National Institute of Public Finance and Policy 18/2, Satsang Vihar Marg, Special Institutional Area New Delhi – 110 067

Background Paper for

Commission on Macroeconomics and Health (India Study) Indian Council for Research on International Economic Relations India Habitat Centre, Lodi Road, New Delhi - 110003 May 2001 * Vinod B.Annigeri is with the Centre for Multi-Disciplinary Development Research, D.B.Rodda Road, Jubilee Circle, Dharwad – 580 001 The authors are grateful to Shri.R.L.Misra and Mrs.Rachel Chatterjee for their constructive comments and suggestions on the earlier draft of this paper.

CONTENTS 1. Introduction .................................................................................................................. 1 2. Methodology and Data ................................................................................................. 2 3. Analysis of Public Spending on Health ........................................................................ 5 4. Role of Central Government in Public Spending on Health ....................................... 8 5. Analysis of Sectoral Spending on Health ................................................................... 10 6. Nexus between Health Expenditures and Health Outcomes ..................................... 12 7. Summary and Policy Implications ............................................................................. 14

2

LIST OF TABLES

Table 1 Share of Health in Revenue Budget of Major States (in %)............................ 5 Table 2 Real Per-Capita Public Spending on Health (in Rs.) ...................................... 6 Table 3 Average Household Expenditure per Hospitalised Case in Government Hospitals (in Rs.)..................................................................... 6 Table 4 Percentage Share of Salaries and Wages in Total Public Spending on Health .......................................................................................... 7 Table 5 Utilisation of Hospital Services for In-Patient Treatments (in %) ................. 8 Table 6 Health Sector Infrastructure in Selected States ............................................... 9 Table 7 Percentage of Central Spending in Total Public Spending on Health............. 9 Table 8 Real Per-Capita Public Spending on Health by Central Government (in Rs.) ......................................................................... 10 Table 9 Trends in Sectoral Spending on Health (in %) ............................................ 11 Table 10 Trends in Per-Capita Real Public Spending on Health by Selected Major States (in Rs.) ..................................................... 12 Table 11A Correlation Coefficients between Under 5 Mortality and per-Capita Health Expenditure ................................................................. 13 Table 11B Correlation Coefficients between IMR and per-Capita Health Expenditure.................................................................. 13 Table 11C Correlation Coefficients between LEB (Female) and per-Capita Health Expenditure ................................................................. 14

3

APPENDICES

Appendix 1 Utilisation of Government Hospital Services for In-Patient Treatments (in %) ............................................................ 16 Appendix 2 Constituents of Primary, Secondary and Tertiary Care Services .......... 16 Appendix 3 Central Grants in Total Public Spending on Health (in %) .................... 17 Appendix 4A Percentage of Allocation to Primary, Secondary and Tertiary Care Services in Total Public Health Spending: 1985-86 ...... 18 Appendix 4B Percentage of Allocation to Primary, Secondary and Tertiary Care Services in Total Public Health Spending: 1991-92 ...... 18 Appendix 4C Percentage of Allocation to Primary, Secondary and Tertiary Care Services in Total Public Health Spending: 1995-96 ...... 19 Appendix 4D Percentage of Allocation to Primary, Secondary and Tertiary Care Services in Total Public Health Spending: 1998-99 ...... 19 Appendix 5A Real Per-Capita Allocation to Primary, Secondary, Tertiary and Total Health Care Services: 1985-86 (in Rs.) ................. 20 Appendix 5B Real Per-Capita Allocation to Primary, Secondary, Tertiary and Total Health Care Services: 1991-92 (in Rs.) ................. 20 Appendix 5C Real Per-Capita Allocation to Primary, Secondary, Tertiary and Total Health Care Services: 1995-96 (in Rs.) ................. 21 Appendix 5D Real Per-Capita Allocation to Primary, Secondary, Tertiary and Total Health Care Services: 1998-99 (in Rs.) ................. 21

CHARTS Chart 1 Share of Health in State and Central Budget (in %)...................................... 22 Chart 2 Share of Health in Revenue Budget of Selected States (in %) ...................... 22

4

TRENDS IN PUBLIC SPENDING ON HEALTH IN INDIA 1. Introduction The public investment in health and health outcomes often do not exhibit any relationship. In spite of this, analysis of differences in health spending assumes significance and eventually is the starting point in any attempt to explain the differences in the health outcomes as observed by the World Bank1. Resource allocation to the health sector is often found to be inadequate in many countries like poor funding of primary care and the rural sector where a majority of the population lives with poorer health status. Also, an in-built urban bias in the government spending on health in most countries as observed in Morocco that less than 20 per cent of the Health Ministry’s budget is allocated to rural areas2. While the percentage of health expenditures devoted to urban sector was around 80 per cent in Ghana during 1985 and around 89 per cent in Cote d’Ivoire during 19843.

The World Development Report 1993 observed that recurrent expenditure for primary care inputs other than salaries is particularly vulnerable to budget cuts. This is applicable in the case of India also, as the expenditure on salaries alone account for more than 60 per cent of the total public health budget, leaving very little for all other components. Since the expenditure on salaries can not be downsized immediately, any reduction in the budget directly affects the expenditures on items like, drugs, medicines, maintenance, etc. As a result, the quality of the services rendered deteriorated in the public facilities. This is evident from the fact that the utilisation of public facilities declined considerably among the poorer sections in both rural and urban India during 1986-87 and 1995-96 (Appendix 1).

The reforms initiated during 1980s in countries like, Germany, Ireland, Italy, etc., have slowed down the rate of growth of health expenditures subsequently4. The experience of these countries repeated in India when the fiscal adjustment programmes were initiated during 1990s. The Central Government in India faced a severe fiscal strain due to the targeted fiscal deficit. 1 2

3

World Bank (1993), World Development Report, Oxford University Press, p.53. Hotchkiss D.R. and A.Gordillo (1999), Household Health Expenditures in Morocco: Implications for Health Care Reform, International Journal of Health Planning and Management, Vol.14 (3), July-August, pp.201-217. Vogel, Ronald J., (1988), Cost Recovery in Health Care Sector: Selected Country Studies in West Africa, World Bank Technical Paper No.82, The World Bank, Washington DC, p.20 & 25.

5

As a result the Central Government reduced its allocation to many states in India5. The cut in the central transfers forced the state governments to reduce the budget allocation to soft targets like health, which were already under-funded. The continuous decline in the budget allocation to health sector as seen from Chart 1, suggests that health sector is loosing its priority and in effect can not be called as a State Subject any more. This raises the fundamental question regarding the rationale and role of public spending on health. The role of public sector in health was considered essential on account of positive externalities, equity and market failure. In this background, this study proposes to analyse the structure of spending on health by government and its impact on health outcomes over a period of about 15 years.

The present study attempts to estimate the extent of spending on health by selected state governments in India and their trends over time. This would also provide an assessment of the inter-state differentials in health care spending among the selected states. An examination of the level and structure of spending and also inter-state differentials in comparison with utilisation of health facilities and the health status outcomes would help us to draw some policy inferences.

2. Methodology and Data

The term health care expenditure is defined to include different components of health services by researchers and hence the term health expenditure varies from one study to another. Many studies in the Indian context confined to medical, public health and family welfare to define health expenditure. Few other studies included nutrition, water supply, sanitation, etc. A detailed discussion on this issue is presented in Reddy and Selvaraju6. In the present study, it is defined to include medical, public health and family welfare as reported in the respective state budget documents since the main thrust of the study is to analyse the public spending on health.

4 5 6

Tanzi, V and L.Schuknecht (2000), Public Spending in the 20th Century: Global Perspective, Cambridge University Press, Cambridge, p.38. Tulasidhar V.B. (1993), Expenditure Compression and Health Sector Outlays, Economic and Political Weekly, Vol.28 (45), November 6, pp.2473-2477. Reddy K.N. and V.Selvaraju (1994), Health Care Expenditure by Government in India, Seven Hills Publications, New Delhi. 6

The classification of health expenditure into purpose oriented categories also varies from one study to another but the variations are not so wide as in the case of health expenditure. In the literature, preventive and curative health services have been discussed, distinguished and analysed more often than primary, secondary and tertiary level health services. In the latter categorisation, analyses on primary health care services have emerged more predominantly ever since the Alma Ata declaration on Health for All was made. This left the analysts debating over the acceptable definition of primary, secondary and tertiary health services. Alma Ata conference defined primary health care as “essential health care made universally accessible to individuals and families by means acceptable to them, through their full participation, and at a cost that the community and country could afford”7. These services were to include as a minimum, health education, maternal and child health, family planning, immunisation, prevention and control of endemic diseases, treatment of common diseases and injuries, and provision of essential drugs. A more practical definition of this as defined by Mach and AbelSmith includes the following8. 1. 2. 3.

All health care from the village or urban community level upto the health centre or firstline hospital, Vertical health programmes, such as communicable disease control or health education, and Water supply, sanitation, nutrition and other activities regarded as primary health care.

Apart from the discussions on the definition of primary health care services, discussions on secondary and tertiary care services are not often made. It is perceived that all the health services of specialised care are generally tertiary services. And, the services other than primary and tertiary are treated as secondary care services that involve the provision of general health care. On the basis of this, the present study classified the health expenditure into various purpose categories by assessing the nature of each of the services as listed in the budget documents of the states. Various health services constituting primary, secondary and tertiary care are presented in Appendix 2 for reference.

The scope of the present study is restricted to ten selected states namely, Andhra Pradesh, Tamil Nadu, Kerala, Madhya Pradesh, Rajasthan, Uttar Pradesh, West Bengal, Orissa, Maharashtra and Gujarat. Haryana has also been added to this list during the course of this 7

Mach EP and B.Abel-Smith, (1983), Planning The Finances of The Health Sector, World Health Organisation, Geneva, p.10.

7

study. These states represent varying levels of development in terms of health indicators as well as various socio-economic indicators. The analysis is carried out for four time points viz. 198586, 1991-92, 1995-96 and 1998-99 which effectively help us capture the trends in expenditures over a period of about 15 years covering pre and post- reform periods. An attempt is made to examine the extent of spending on primary, secondary and tertiary care services and the salary and non-salary components of the total health spending. Also, the relationship between the structure of public spending on health - in terms of allocation to primary, secondary and tertiary care services - and the utilisation of public health services is examined. For this purpose, the information available from 42nd and 52nd Rounds of NSSO on health have been utilised. Expenditure incurred by the Ministry of Health and Family Welfare on medical, public health and family welfare has been defined in this study as the public spending on health by the respective state governments. Detailed information on public spending have been culled out from the budget documents of the respective state governments.

In the present study, only revenue expenditure has been considered and the expenditures incurred under capital account of the budgets have not been included. Because, simple addition of capital expenditures to revenue expenditure in order to arrive at the total spending on health would be misleading and exaggerate the levels of spending. Ideally, annualised capital expenditure rather than total spending under capital account would be appropriate to assess the extent of public spending in any given year. However, the present analysis does not underestimate the health spending significantly by not considering the annualised capital expenditure because the annualised capital expenditure accounts for less than 10 per cent of the revenue expenditure9.

The rest of the paper is organised as follows. Public spending on health by the 11 major states selected for this study in terms of total budget and in real per-capita have been discussed for the four selected time points in Section 3. Proportion of salaries in total health budget has also been worked out and presented in this section. In Section 4, percentage of spending and percapita spending on health by the Central Government has been analysed. Trends in the public spending on primary, secondary and tertiary care services have been analysed in Section 5.

8 9

Mach EP and B.Abel-Smith, (1983), Planning the Finances of The Health Sector, World Health Organisation, Geneva, pp.71-72. Selvaraju V., (2001), Budgetary Subsidies to Health Sector among Selected States in India, paper for the forthcoming HELPONET Workshop scheduled in July 2001, IIM, Ahmedabad. 8

Nexus between sectoral allocations to health and health outcomes has been examined in Section 6. Summary and policy implications are presented in Section 7.

3. Analysis of Public Spending on Health

Allocations to social sectors in the revenue budgets of the state governments have been declining ever since the introduction of the fiscal adjustment policies. Studies carried out even during the early 90s found declining allocation to health sector in many states10. For instance, health budget accounts for 7.19 per cent of the total revenue budget of the eleven major states under review in this study in 1985-86 (Table 1). Subsequently the share of health budget sharply declined to 5.76 per cent in 1991-92 and continues to be at the same level (Chart 2). The budget allocation to health of all the states recorded a sharp decline from 1984-85 to 1991-92. The decline in the allocation continued during 1991-92 and 1995-96 as well and shown an improvement in 1998-99. However the increase in the allocation during 1995-96 and 1998-99 is negligible.

Table 1 Share of Health in Revenue Budget of Major States (in %) States

1985-86

1991-92

1995-96

1998-99

1 Andhra Pradesh

6.43

5.77

5.64

6.07

2 Gujarat

7.61

5.42

5.34

5.39

3 Haryana

7.06

4.19

2.99

4.16

4 Kerala

8.40

6.92

6.81

5.86

5 Maharashtra

6.24

5.25

5.18

4.52

6 Madhya Pradesh

6.43

5.66

5.07

5.77

7 Orissa

7.59

5.94

5.42

5.88

8 Rajasthan

8.17

6.85

6.18

7.08

9 Tamil Nadu

7.23

4.82

6.28

6.13

10 Uttar Pradesh

7.67

6.00

5.72

4.73

11 West Bengal

8.02

7.31

7.16

7.95

7.19

5.76

5.66

5.67

Average of States

In terms of per-capita spending on health, it increased from Rs.21.28 in 1985-86 to Rs.25.74 in 1991-92 and remained at the same level at Rs.25.66 during 1995-96 in 1980-81 prices (Table 2). This implies that health sector was assigned a low priority during 1985-86 to 199192 and during 1990s even that low priority was lost in order to achieve the fiscal deficit targets. 10

Tulasidhar V.B. (1993), Expenditure Compression and Health Sector Outlays, Economic and Political Weekly, Vol.28 (45), November 6, pp.2473-2477. 9

As a result household expenditures even to avail the free ward treatments in public hospitals increased significantly as evident from Table 3.

Table 2 Real Per-Capita Public Spending on Health (in Rs.) States

1985-86

1991-92

1995-96

1998-99

1 Andhra Pradesh

20.44

21.03

21.92

31.88

2 Gujarat

24.32

30.51

28.77

45.44

3 Haryana

26.79

26.65

24.39

33.78

4 Kerala

25.97

32.15

30.98

35.05

5 Maharashtra

27.46

30.87

30.73

33.67

6 Madhya Pradesh

16.19

19.17

17.89

25.49

7 Orissa

16.95

23.26

19.54

28.28

8 Rajasthan

21.85

29.07

31.02

37.70

9 Tamil Nadu

15.38

21.61

32.09

42.42

10 Uttar Pradesh

16.12

20.38

19.01

18.10

11 West Bengal

22.65

28.49

25.96

41.24

Average Spending

21.28

25.74

25.66

33.91

Note: The figures are in constant prices 1980-81=100

Table 3 Average Household Expenditure Per Hospitalised Case in Government Hospitals (in Rs.) Sector/Period

Free Ward

General Ward

Special Ward

Rural 1986-87

582

1421

1268

1995-96

1781

3241

10540

Urban 1986-87

630

1040

1483

1995-96

1521

3350

12474

Source:

1.NSSO, Sarvekshana, Department of Statistics, CSO, Government of India, April-June 1992, p.S-437. 2.NSSO, Morbidity and Treatment of Ailments, Report No.441, Department of Statistics, CSO, Government of India, November 1998, p.A-94 and A-199.

The level of per-capita public spending by the states suggests that those states, which spend relatively higher levels also, are the states with better health status indicators, excepting Rajasthan. Per-capita spending in Rajasthan is much higher compared to states like Kerala, West Bengal, etc., but the health status indicators place the state in the bottom of the list. One of the reasons for this dichotomy could be due to the low density of population coupled with a larger proportion of tribal population and desert region leading to higher unit cost of services.

10

Analysts of public expenditures often find that the major portion of the public spending on health is incurred on salaries and wages leaving a small proportion for supplies, materials and maintenance. Break up of public spending of the states reveal that the proportion of salaries and wages in total spending recorded an increase from 60.43 per cent in 1985-86 to 74.04 per cent in 1998-99 (Table 4). Salaries accounted for as high as 80 per cent of the total health budget in BIMARU states in 1998-99. Low levels of proportions reported for Andhra Pradesh, Gujarat and Maharashtra are mainly due to the fact that local governments in these states are entrusted with the running and maintenance of primary health centres and dispensaries for which they receive grants and assistance from the respective state governments. Some of these grants and assistance provided to local governments towards salaries and wages are not duly recorded under the head of salaries and wages in the state budgets.

Table 4 Percentage Share of Salaries and Wages in Total Public Spending on Health States

1985-86

1991-92

1995-96

1998-99

1

Andhra Pradesh

55.98

51.80

63.22

50.28

2

Gujarat

31.76

35.13

42.13

38.82

3

Haryana

53.75

70.63

72.73

63.83

4

Kerala

63.54

70.90

66.57

70.06

5

Maharashtra

41.79

36.71

39.17

46.58

6

Madhya Pradesh

65.53

70.68

76.75

79.58

7

Orissa

65.14

72.08

73.22

82.39

8

Rajasthan

52.42

70.54

70.39

79.02

9

Tamil Nadu

48.23

65.82

64.80

74.00

10

Uttar Pradesh

50.66

68.15

69.06

80.10

11

West Bengal

60.43

66.76

72.53

77.04

The huge salary components in public health spending and its increasing trend is a cause of serious concern. A recent study reveals that health care spending by governments and households are complementary and not substitutes in the sense that governments spend on largely manpower by way of salaries and households spend a large proportion on medicines and treatment facilities11. Given the fact that households even in the rural sector devote more than 50 per cent of health expenditure on medicines, any further pressure on the households will drive the poor and rural households away from utilising the health services. Evidences from the household surveys of NSSO suggest that there is a sharp decline in the utilisation of

11

Selvaraju, V. (2000), Health Care Expenditure in Rural India: A Micro Perspective, paper presented at the International Seminar on Human Development in India, organised by the National Council of Applied Economic Research, New Delhi, November. 11

hospital services for in-patient treatments by the poorer sections in both rural and urban India in 1995-96 as compared to 1986-87 (Table 5). Therefore, a sharp increase in the total volume of public spending on health is essential to improve the health care utilisation of households and also to keep the salary and non-salary components of public spending at a desired level.

Table 5 Utilisation of Hospital Services for In-Patient Treatments (in %) Fractile Groups

1986-87 (42nd Round)

1995-96 (52nd Round)

Rural

Rural

Urban

Urban

0-10

11.20

10.99

2.50

5.90

10-20

8.40

11.00

3.90

6.50

20-40

20.42

21.42

11.30

15.50

40-60

18.93

20.21

14.10

18.90

60-80

20.77

24.32

23.30

20.30

80-90

13.02

6.80

15.70

13.40

90-100

7.26

5.26

29.20

19.50

All Groups 100.00 100.00 100.00 100.00 nd nd Note: Services availed from all types hospitals covered by NSSO in its 42 and 52 rounds. Source: 1.NSSO, Morbidity and Utilisation of Medical Services, Report No.364, Department of Statistics, CSO, Government of India, September 1989, pp.A-8-13. 2.NSSO, Morbidity and Treatment of Ailments, Report No.441, Department of Statistics, CSO, Government of India, November 1998, p.A-65 and p.A-170.

One of the reasons for the larger salary component in public spending on health in BIMARU states is due to the fact that government hospitals account for considerably a larger share in the total number of hospitals in these states. Further, the ratio of population per bed is also found to be higher in BIMARU states as seen in Table 6. Dependency on public health facilities is much higher due to inadequate private health facilities in these states. Efforts are needed through public investments for building necessary health infrastructure and to attract private sector health facilities.

4. Role of Central Government in Public Spending on Health The Central Government plays a significant role in the health sector of the states through centrally sponsored programmes, loans, transfers and grants. It is often observed that health priorities and also the health outcomes of states are influenced by the Central grants. Central government intervenes primarily in the areas of disease control programmes under public health and family welfare programmes of the states in order to accomplish national health priorities. Some of the disease control programmes are partially funded on matching grant basis, while others are fully funded. The Central share in total public spending on health in the state of Orissa was nearly 30 per cent in 1985-86 and also in 1991-92 and declined to 21.44 per cent 12

in 1998-99 (Table 7). Both in Orissa and Rajasthan, the Central share in the health budget remained relatively higher among other states throughout the 15 years period. The health budget of Maharashtra contained the lowest Central share of 8.75 per cent followed by West Bengal with 12.90 per cent of Central share in 1998-99.

Table 6 Health Sector Infrastructure in Selected States States

Total Number of Government Hospitals

Proportion of Government Hospitals

Population per Bed

1

Andhra Pradesh

2950

5.02

1526

2

Gujarat

2528

12.34

706

3

Haryana

79

74.68

2584

4

Kerala

2040

6.91

391

5

Madhya Pradesh

363

100.00

3770

6

Maharashtra

3115

14.29

1023

7

Orissa

430

95.58

2314

8

Rajasthan

218

100.00

2347

9

Tamil Nadu

408

69.12

1120

10 Uttar Pradesh

735

72.65

2593

11 West Bengal

399

60.90

1351

15097

29.63

1498

All India Note: Source:

Total number of hospitals and beds include those existing under government, local bodies, private and voluntary organisations in the state. Government of India, Health Information of India - 1995 & 96, Central Bureau of Health Intelligence, DGHS, Ministry of Health and Family Welfare, New Delhi, pp.131-133.

Table 7 Percentage of Central Spending in Total Public Spending on Health States

1985-86

1991-92

1998-99

1 Andhra Pradesh

23.54

20.94

0.00

2 Haryana

16.61

17.72

18.54

1.46

19.61

17.17

4 Maharashtra

10.21

5.70

8.75

5 Orissa

29.90

29.90

21.44

6 Rajasthan

23.81

19.66

19.54

7 Tamil Nadu

13.91

22.53

17.11

8 West Bengal

14.25

12.70

12.90

3 Kerala

In terms of real per-capita expenditure also the backward states of Orissa and Rajasthan received substantially higher central allocation during 1985-86 to 1998-99 (Table 8). Central allocation to Tamil Nadu in real per-capita terms increased significantly from Rs.3.79 in 198586 to Rs.7.54 in 1991-92 and remained the second highest among the states in 1998-99 at 13

Rs.7.26. The central allocation for the state of Maharashtra remained substantially low at Rs.1.76 in 1991-92 and at Rs.2.94 in 1998-99. It can be seen from Appendix 3 that in states like Haryana, Orissa and Rajasthan more then 90 per cent of the spending on family welfare is shared by the Central government since 1991-92. On the other hand, on the total spending on public health in Kerala, the Central share increased from 43.98 per cent in 1991-92 to 61.28 per cent in 1998-99. At the same time, the share in family welfare declined from 97.08 per cent in 1991-92 to 79.86 per cent in 1998-99. Compared to other states, the Central share in public health in Kerala is substantially high.

Table 8 Real Per-Capita Public Spending on Health By Central Government (in Rs.) States

1985-86

1991-92

1998-99

1 Andhra Pradesh

4.81

4.40

0.00

2 Haryana

4.45

4.72

6.26

3 Kerala

0.38

6.31

6.02

4 Maharashtra

2.80

1.76

2.94

5 Orissa

5.07

6.95

6.06

6 Rajasthan

5.20

5.71

7.37

7 Tamil Nadu

3.79

7.54

7.26

8 West Bengal

3.16

3.61

5.32

Note: The figures are in constant prices 1980-81=100

5. Analysis of Sectoral Spending on Health Allocation to different levels of health services assumed significance with the Alma Atta declaration that emphasised a substantial allocation to primary care services. Investments in primary care services were found to be the key to achieve better health outcomes. The share of government health spending on primary care services was estimated to be around 43 per cent in India12. The estimates of the present study reveal that the allocation to primary health services recorded a marginal increase from 37.80 per cent in 1985-86 to 38.63 per cent in 1990-91 and exhibited a declining trend thereafter in the states under review (Table 9). In 1998-99, it declined to 35.81 per cent of the total health spending. State specific sectoral allocations are presented in Appendices 4A to 4D respectively for 1985-86, 1991-92, 1995-96 and 1998-99.

14

The share of tertiary care services in total health spending remained stagnant around 17 per cent during 1985-86 to 1991-92 and declined sharply to 10.59 per cent during the initial phase of the adjustment programmes. But in 1998-99, the share of tertiary services shot up to 22.10 per cent and as a result, the share of secondary care services declined from 40.19 per cent in 1995-96 to 29.46 per cent in 1998-99. The trends in the allocation to primary, secondary and tertiary services suggest that even the public spending on health is tending to promote tertiary care services at the cost of secondary and primary care services.

Table 9 Trends in Sectoral Spending on Health (in %) Year

Primary

Secondary

Tertiary

Others

Total

1985-86

37.80

32.57

17.45

12.18

100.00

1990-91

38.63

30.54

17.79

13.03

100.00

1995-96

37.41

40.19

10.59

11.81

100.00

1998-99

35.81

29.45

22.10

12.64

100.00

Note: Figures presented refer to average spending by eleven major states reviewed in this study

Analysis of real per-capita spending reveals that the spending on primary care services recorded a marginal increase from Rs.8.15 in 1985-86 to Rs.9.89 in 1991-92 and remained near stagnant till 1995-96 in the states under review (Table 10). The per-capita spending on primary care registered a considerable increase in 1998-99 and reached a level of Rs.12.34. State specific estimates of per-capita spending is presented in Appendices 5A to 5D respectively for 1985-86, 1991-92, 1995-96 and 1998-99. Spending on secondary care services witnessed an increasing trend from Rs.7.21 in 1985-86 to Rs.10.45 in 1995-96 and declined marginally to Rs.10.22 in 1998-99. Spending on tertiary care services shot up in per-capita terms from Rs.2.74 in 1995-96 to Rs.7.15 in 1998-99. Allocation to tertiary care services as a proportion of total spending also recorded a sharp increase in 1998-99 as seen in Table 9.

The figures presented in Tables 9 and 10 suggest that both primary and secondary care services remained a priority area in terms of per-capita spending and also as a share in total spending among the eleven states under review. However, the faster growth recorded by tertiary care services in 1998-99, if left unchecked would negate the underlying principles of equity and efficiency in health sector. Given the fact that tertiary care services tend to promote and escalate the cost of treatments, faster growth of tertiary care services evidenced in the late 90s will have serious financial implications both on households and governments. 12

World Bank, (1994), India: Policy and Finance Strategies for Strengthening Primary Health Care Services, Population and Human Resources Operations Division, October 1994, p.104. 15

Table 10 Trends in Per-Capita Real Public Spending on Health by Selected Major States (in Rs.) Year

Primary

Secondary

Tertiary

Others

Total

1985-86

8.15

7.21

3.43

2.50

21.28

1990-91

9.89

8.09

4.53

3.23

25.74

1995-96

9.66

10.45

2.74

2.81

25.66

1998-99

12.34

10.22

7.15

4.21

33.91

Note: Figures presented refer to average spending by eleven major states reviewed in this study

6. Nexus between Health Expenditures and Health Outcomes It is often argued that the relationship between public spending health and health outcomes measured in terms of life expectancy is weak. Studies have shown that factors like income, literacy level, and other social and economic factors influence the level of health status indicators than health expenditure perse. A detailed discussion on this issue is presented in Reddy and Selvaraju13. A recent study on poor countries by Filmer, Hammer and Pritchett reveals that the weak links between the health expenditure and health services found in cross section analyses is mainly due to low elasticity of demand for health services and relatively larger private sector14.

In the light of this discussion, an attempt was made in this section to examine the nexus between the health status indicators and health expenditures to find out, how efficient is the public spending on health in India. For examining this, per-capita public spending on primary, secondary, tertiary care services and total per capita spending have been linked with health outcome indicators like under 5 mortality, IMR and life expectancy at birth. Studies in the past have proved that regression analysis of any type to find out the impact of public spending on health outcomes is ineffective. Therefore, the analysis in this section attempts to find out whether there exists any relation, and not the cause and effect relation, between health spending and outcomes by estimating simple correlation coefficients.

The matrix of estimated correlation coefficients linking per-capita health spending with under 5 mortality, IMR and life expectancy are presented respectively in Tables 11A., 11B and 11C. 13

Reddy, K.N.and V.Selvaraju, (1998), Determinants of Health Status in India: An Empirical Verification", in D.K.Das (ed) Indian Economy After 50 Years of Independence: Social Sector and Development, Volume 4, Deep and Deep Publications, New Delhi, pp.110-117.

16

The coefficients presented in Tables 11A and 11B suggest that total per capita spending and the per-capita spending on secondary care services are highly correlated with under 5 mortality and IMR in all the four years. Also the signs of the coefficients are negative as expected. Even though, the coefficients for primary care and tertiary care services also assumed a negative sign (except one in 1995-96), the degree of association is found to be weak in all the four years under review. One of the reasons for the weak association is the lag between the spending and outcomes. The extent of lag varies widely among states depending on their capacity to absorb the spending and reflect in terms of health outcomes. Table 11A Correlation Coefficients between Under 5 Mortality and Per-Capita Health Expenditure Year

Primary Care Expenditure

Secondary Care Expenditure

Tertiary Care Expenditure

General Expenditure

Per-Capita Health Expenditure

1985-86

-0.0358

-0.6158*

-0.0138

-0.3295

-0.6409*

1991-92

-0.2793

-0.4848

-0.0716

-0.0960

-0.6243*

1995-96

-0.2736

-0.7011*

0.0049

-0.2569

-0.6751*

1998-99

-0.1460

-0.5752

-0.1422

-0.1896

-0.5086

Note: 1. N=11 (states). 2. * indicates the coefficients are significant at 5 per cent.

Table 11B Correlation Coefficients between IMR and Per-Capita Health Expenditure Year

Primary Care Expenditure

Secondary Care Expenditure

Tertiary Care Expenditure

General Expenditure

Per-Capita Health Expenditure

1985-86

-0.0127

-0.6378*

-0.1165

-0.2851

-0.6761*

1991-92

-0.2352

-0.5623

-0.1132

-0.0062

-0.6507*

1995-96

-0.2326

-0.7909**

-0.0723

-0.1647

-0.6993*

1998-99

-0.0590

-0.6508*

-0.1408

-0.1038

-0.4736

Note: 1. N=11 (states). 2. * indicates the coefficients are significant at 5 per cent. 3. ** indicates the coefficients are significant at 10 per cent.

Life expectancy at birth (LEB) is also found to be highly correlated with the total per-capita spending on health as shown in Table 11C. Analysis of health expenditure into primary, secondary and tertiary care suggests that LEB is strongly associated with secondary care services with expected positive sign. Its association with primary and tertiary care services are again found to be weak. The level and the structure of spending on primary, secondary and tertiary care services for individual states in Appendices 4 and 5 corroborate the above findings. For instance, Kerala has been spending consistently on primarily care services while at the same time increased the allocation to secondary care services. On the contrary, the level and structure of spending of other states moved in favour of tertiary care services by 14

Filmer D, J.S.Hammer and L.H.pritchett (2000), Weak Links in the Chain: A Diagnosis of Health Policy in Poor Countries, World Bank Research Observer, Vol.15 (2), August, pp.199-224. 17

undermining the secondary care services. The trends in the spending pattern of the states reveal that the states, which spend relatively higher on secondary care services, achieve better health outcomes. Table 11C Correlation Coefficients between LEB (Female) and Per-Capita Health Expenditure Year

Primary Care Expenditure

Secondary Care Expenditure

Tertiary Care Expenditure

General Expenditure

Per-Capita Health Expenditure

1991-92

a

0.1035

0.6180*

0.0043

0.1652

0.6590*

1995-96

b

0.1871

0.7453**

0.1197

0.2060

0.6820*

Note: 1. N=11 (states). 2. * indicates the coefficients are significant at 5 per cent. 3. ** indicates the coefficients are significant at 10 per cent. 4. a LEB (female) for 1991-96 related with expenditure during 1991-92 5. b LEB (female) for 1996-01 related with expenditure during 1995-96

7. Summary and Policy Implications

Allocation to health sector in India in the total budget has been declining and the decline is more pronounced during the initial years of the fiscal adjustment programme. Of the total public spending on health, the component of salaries and wages witnessed an increasing trend and reached a level of about 70 to 80 per cent especially in BIMARU states. The household expenditure for treatments have registered a substantial increase as evidenced from two rounds of NSSO. The results of these surveys also suggest that the utilisation of health facilities have declined sharply among the poorer sections. A substantial public investment is needed to reverse the existing trends so that a fair proportion is spent on non-salary components to make the services more effective. A move in this direction would also reduce the burden of morbidity on households.

The share of Central Government in the total spending of the states suggests that Central Government spends a considerable amount on family welfare programmes of many states and to some extent on public health programmes. But the analysis suggests that the states with a larger share of Central spending in public health programmes are also the states with better health outcomes. The state level analysis suggests that the extent of spending does not really have any significant bearing on health outcomes but it is the structure of spending which would impact the outcomes significantly. Analysis of correlation coefficients also strengthens this view by exhibiting a strong association between health outcomes and spending on secondary care services. This would mean the prime focus should be on secondary care services rather than the hitherto emphasis on primary care services. At the same time, the primary care 18

services can not be undermined. Therefore, given the emphasis on primary care services, efforts should be made to augment the capacity of secondary care services.

19

Appendix 1 Utilisation of Government Hospital Services for In-Patient Treatments (in %) Sl. No.

Fractile Groups

1

1986-87 (42

nd

Round)

1995-96 (52

nd

Round)

Rural

Urban

Rural

Urban

0-10

12.94

11.59

3.13

12.87

2

10-20

10.59

11.42

6.30

7.40

3

20-40

22.94

25.66

16.70

20.90

4

40-60

18.69

23.91

17.20

18.17

5

60-80

19.73

17.28

24.30

19.77

6

80-90

8.82

3.63

14.07

10.23

7

90-100

6.30

6.52

18.30

10.67

100.00

100.00

100.00

100.00

All Groups

Note: Government hospitals refer to public hospitals, PHCs and public dispensaries. Source:

1.NSSO, Morbidity and Utilisation of Medical Services, Report No.364, Department of Statistics, CSO, Government of India, September 1989, pp.A-8-13. 2.NSSO, Morbidity and Treatment of Ailments, Report No.441, Department of Statistics, CSO, Government of India, November 1998, p.A-65 and p.A-170.

Appendix 2 Constituents of Primary, Secondary and Tertiary Care Services Budget Heads of Health Primary Care Services

Budget Heads of Health Tertiary Care Services

1 Primary Health Centres

1 Attached to Teaching Institutions

2 Health Sub-Centres

2 Major Hospitals

3 Other Health Services

3 Tuberculosis Institutions

4 School Health Schemes

4 Allopathy - Medical Education, Training & Research

5 Public Health 6 Family welfare Secondary Care Services

General

1 Employees State Insurance Scheme

1 Direction and Administration

2 Central Government Health Scheme

2 Health Statatistics, Research, Evaluation & Training

3 Hospital and Dispensaries

3 Medi.Stores Depot & Deptl.Drug Manufactures

4 Community Health Centres

4 Tribal Area Sub-Plan

5 Ayurveda – Other Systerms

5 Other Expenditures

6 Homeopathy – Other Systems

6 Assist.to local bodies, corporates, etc.

7 Unani – Other Systems

7 Machinery and Equipments - Family Welfare

8 Sidha – Other Systems 9 Other Systems

20

Appendix 3 Central Grants in Total Public Spending on Health (in %) States Andhra Pradesh

Haryana

Kerala

Expenditure Head

1985-86

1991-92

1998-99

Medical

0.09

0.00

0.00

Public Health

15.48

10.72

0.00

Family Welfare

90.91

95.20

0.00

Total

23.54

20.94

0.00

Medical

0.00

0.00

1.20

Public Health

63.72

0.00

13.93

Family Welfare

0.00

100.00

100.00

Total

16.61

17.72

18.54

Medical

0.00

0.30

0.00

Public Health

16.11

43.98

61.28

Family Welfare

0.00

97.08

79.86

Total

1.46

19.61

17.17

0.00

0.01

0.07

Public Health

26.91

14.03

8.78

Family Welfare

0.00

0.00

59.10

Total

10.21

5.70

8.75

Medical

4.63

3.99

0.51

Public Health

30.52

17.34

7.03

Family Welfare

99.96

99.15

96.42

Total

29.90

29.90

21.44

Medical

1.07

0.39

0.39

Public Health

36.66

15.15

21.24

Family Welfare

87.79

92.02

90.96

Total

23.81

19.66

19.54

Maharashtra Medical

Orissa

Rajasthan

Source:

Selvaraju V., (2001), Budgetary Subsidies to Health Sector among Selected States in India, paper for the forthcoming HELPONET Workshop scheduled in July 2001, IIM, Ahmedabad.

21

Appendix 4A Percentage of Allocation to Primary, Secondary and Tertiary Care Services in Total Public Health Spending: 1985-86 Sl. No.

States

Primary

Secondary

Tertiary

Others

Total

1 Andhra Pradesh

46.85

23.33

23.49

6.32

100.00

2 Gujarat

46.85

21.53

22.76

8.86

100.00

3 Haryana

50.63

29.17

12.17

8.03

100.00

4 Kerala

24.73

59.74

8.69

6.84

100.00

5 Maharashtra

31.27

36.63

6.23

25.87

100.00

6 Madhya Pradesh

41.97

20.60

23.91

13.53

100.00

7 Orissa

46.37

31.87

5.46

16.30

100.00

8 Rajasthan

27.11

51.22

12.04

9.63

100.00

9 Tamil Nadu

49.61

30.31

10.72

9.36

100.00

10 Uttar Pradesh

44.29

32.87

17.91

4.93

100.00

11 West Bengal

21.23

26.46

36.01

16.30

100.00

Appendix 4B Percentage of Allocation to Primary, Secondary and Tertiary Care Services in Total Public Health Spending: 1991-92 Sl. No.

States

Primary

Secondary

Tertiary

Others

Total

1 Andhra Pradesh

49.17

24.23

22.38

4.22

100.00

2 Gujarat

37.11

22.18

30.29

10.43

100.00

3 Haryana

42.22

35.24

16.28

6.25

100.00

4 Kerala

27.29

53.70

9.31

9.70

100.00

5 Maharashtra

24.56

37.82

6.84

30.79

100.00

6 Madhya Pradesh

39.21

20.33

23.60

16.85

100.00

7 Orissa

41.82

32.38

5.25

20.54

100.00

8 Rajasthan

35.20

36.40

19.80

8.61

100.00

9 Tamil Nadu

67.65

12.31

13.43

6.62

100.00

10 Uttar Pradesh

40.49

39.11

12.65

7.75

100.00

11 West Bengal

32.20

21.66

33.23

12.90

100.00

22

Appendix 4C Percentage of Allocation to Primary, Secondary and Tertiary Care Services in Total Public Health Spending: 1995-96 Sl. No.

States

Primary

Secondary

Tertiary

Others

Total

1 Andhra Pradesh

49.46

23.84

22.94

3.76

100.00

2 Gujarat

40.40

44.13

7.35

8.12

100.00

3 Haryana

39.55

38.88

15.90

5.67

100.00

4 Kerala

28.36

54.05

8.19

9.41

100.00

5 Maharashtra

26.17

36.84

6.60

30.39

100.00

6 Madhya Pradesh

35.70

44.59

5.51

14.20

100.00

7 Orissa

44.69

33.04

5.55

16.72

100.00

8 Rajasthan

39.60

34.65

19.14

6.60

100.00

9 Tamil Nadu

42.86

45.54

7.17

4.43

100.00

10 Uttar Pradesh

37.44

41.40

11.67

9.49

100.00

11 West Bengal

35.01

45.66

7.75

11.58

100.00

Appendix 4D Percentage of Allocation to Primary, Secondary and Tertiary Care Services in Total Public Health Spending: 1998-99 Sl. No.

States

Primary

Secondary

Tertiary

Others

Total

1 Andhra Pradesh

46.66

27.98

18.16

7.20

100.00

2 Gujarat

36.26

23.61

28.47

11.66

100.00

3 Haryana

40.97

36.46

16.83

5.74

100.00

4 Kerala

28.20

54.04

8.77

8.99

100.00

5 Maharashtra

20.93

38.68

7.47

32.92

100.00

6 Madhya Pradesh

39.24

21.33

19.54

19.89

100.00

7 Orissa

41.28

31.80

5.71

21.21

100.00

8 Rajasthan

36.75

34.47

20.13

8.65

100.00

9 Tamil Nadu

42.29

14.84

36.84

6.04

100.00

10 Uttar Pradesh

33.73

37.22

25.49

3.57

100.00

11 West Bengal

33.93

19.42

34.58

12.07

100.00

23

Appendix 5A Real Per-Capita Allocation to Primary, Secondary, Tertiary and Total Health Care Services: 1985-86 (in Rs.) Sl.No.

States

Primary

1 Andhra Pradesh

Secondary

Tertiary

Others

Total

9.58

4.77

4.80

1.29

20.44

2 Gujarat

11.39

5.24

5.54

2.15

24.32

3 Haryana

13.57

7.82

3.26

2.15

26.79

4 Kerala

6.42

15.52

2.26

1.78

25.97

5 Maharashtra

8.59

10.06

1.71

7.11

27.46

6 Madhya Pradesh

6.79

3.34

3.87

2.19

16.19

7 Orissa

7.86

5.40

0.93

2.76

16.95

8 Rajasthan

5.92

11.19

2.63

2.10

21.85

9 Tamil Nadu

7.63

4.66

1.65

1.44

15.38

10 Uttar Pradesh

7.14

5.30

2.89

0.79

16.12

11 West Bengal

4.81

5.99

8.16

3.69

22.65

8.15

7.21

3.43

2.50

21.28

Average Spending Note: in Constant Prices 1980-81=100

Appendix 5B Real Per-Capita Allocation to Primary, Secondary, Tertiary and Total Health Care Services: 1991-92 (in Rs.) Sl. No.

States

Primary

Secondary

Tertiary

Others

Total

1 Andhra Pradesh

10.34

5.09

4.71

0.89

21.03

2 Gujarat

11.32

6.77

9.24

3.18

30.51

3 Haryana

11.25

9.39

4.34

1.67

26.65

4 Kerala

8.77

17.27

2.99

3.12

32.15

5 Maharashtra

7.58

11.68

2.11

9.51

30.87

6 Madhya Pradesh

7.52

3.90

4.52

3.23

19.17

7 Orissa

9.73

7.53

1.22

4.78

23.26

8 Rajasthan

10.23

10.58

5.75

2.50

29.07

9 Tamil Nadu

14.62

2.66

2.90

1.43

21.61

10 Uttar Pradesh

8.25

7.97

2.58

1.58

20.38

11 West Bengal

9.17

6.17

9.47

3.68

28.49

9.89

8.09

4.53

3.23

25.74

Average Spending Note: in Constant Prices 1980-81=100

24

Appendix 5C Real Per-Capita Allocation to Primary, Secondary, Tertiary and Total Health Care Services: 1995-96 (in Rs.) Sl. No.

States

Primary

Secondary

Tertiary

Others

Total

1 Andhra Pradesh

10.84

5.23

5.03

0.83

21.92

2 Gujarat

11.62

12.70

2.12

2.34

28.77

3 Haryana

9.65

9.48

3.88

1.38

24.39

4 Kerala

8.79

16.75

2.54

2.91

30.98

5 Maharashtra

8.04

11.32

2.03

9.34

30.73

6 Madhya Pradesh

6.39

7.98

0.99

2.54

17.89

7 Orissa

8.73

6.46

1.09

3.27

19.54

8 Rajasthan

12.28

10.75

5.94

2.05

31.02

9 Tamil Nadu

13.75

14.61

2.30

1.42

32.09

10 Uttar Pradesh

7.12

7.87

2.22

1.80

19.01

11 West Bengal

9.09

11.85

2.01

3.01

25.96

9.66

10.45

2.74

2.81

25.66

Average Spending Note: in Constant Prices 1980-81=100

Appendix 5D Real Per-Capita Allocation to Primary, Secondary, Tertiary and Total Health Care Services: 1998-99 (in Rs.) Sl. No.

States

Primary

Secondary

Tertiary

Others

Total

1 Andhra Pradesh

14.88

8.92

5.79

2.30

31.88

2 Gujarat

16.48

10.73

12.94

5.30

45.44

3 Haryana

13.84

12.32

5.68

1.94

33.78

4 Kerala

9.89

18.94

3.07

3.15

35.05

5 Maharashtra

7.05

13.02

2.52

11.08

33.67

6 Madhya Pradesh

10.00

5.44

4.98

5.07

25.49

7 Orissa

11.67

8.99

1.61

6.00

28.28

8 Rajasthan

13.85

12.99

7.59

3.26

37.70

9 Tamil Nadu

17.94

6.29

15.62

2.56

42.42

10 Uttar Pradesh

6.10

6.74

4.61

0.65

18.10

11 West Bengal

13.99

8.01

14.26

4.98

41.24

12.34

10.22

7.15

4.21

33.91

Average Spending Note: in Constant Prices 1980-81=100

25

Chart 1 Share of Health in State and Central Budget (in %) 8

7

6

5

4

3

2

1

Share of Health Expenditure in States

1997-98

1996-97

1995-96

1994-95

1993-94

1992-93

1991-92

1990-91

1989-90

1988-89

1987-88

1986-87

1985-86

1984-85

1983-84

1982-83

1981-82

1980-81

0

Share of Health Expenditure in Central Govt.

Chart 2 Share of Health in Revenue Budget of Selected States (in %)

8

7.19 7

5.76

5.66

1991-92

1995-96

6

5.67

5

4

3

2

1

0

1984-85

1998-99

26