This study describes the derivation of health production functions and applies them to the evaluation of the health impact of investments in safe water and sanitation. Also, the study demonstrates the potential use of a health production function for resource allocation decicion-making in the health sector.
Using cross-sectional data from 101 countries it is shown that safe water is a statistically significant determinant for changes in the Infant Mortality Rate (IMR) in an all-countries equation (i.e. 101 countries), and is almost significant at the ten per cent level in an equation estimated for low-income countries. However, safe water is shown to be less important for the IMR than other explanatory variables in the model.
The study then derives cross-sectional health production functions for the province of Central Java, Indonesia. Morbidity of water-borne diseases and morbidity of diarrhoea are chosen as the health output measures; safe water coverage and sanitation (excreta disposal) coverage are the input variables. The functions that best fit the data have a reciprocal form and exhibit constant returns to scale. Safe water coverage and sanitation coverage are significant regressors for the morbidity data. Safe water appears to be more important for morbidity than sanitation. A simultaenous m-fold increase in the coverage of safe water and sanitation produces a potential morbidity reduction of 1-1/m. Safe water and sanitation appear to have a low substitutability.
Both financial and economic cost minimisation procedures are then undertaken. The economic analysis takes account of shadow prices of land, labour, and foreign exchange, and includes an estimate of bureaucratic-induced X-inefficiencies. The social opportunity cost of capital approach is adopted to estimate a discount rate. In deriving isocost curves for safe water and sanitation, the fact that safe water coverage and sanitation coverage are intermediate inputs is taken into account.
Even though investment cost accounts for a high proportion of per capita cost of safe water and sanitation facilities, the proportion of the operational, maintenance and renewal costs to per capita cost cannot be underestimated. Thus, if any safe water and sanitation project is to remain sustainable, sufficient incoming funds must be available to meet these costs. The use of the 'user pays' principle is worth consideration.
At current cost levels, the cost minimising points occur when sanitation coverage is about one-fourth of safe water coverage. If cost minimisation is desirable, only increases in safe water coverage are advisable; increases in sanitation coverage are not. Had the cost minimising points been satisfied, a potential cost saving of more than Rp 100 billion (US$ 54 million) at 1990 prices, or a potential morbidity reduction of about one-fifth of the current levels, would have been achieved.
The most desirable policy is to integrate a health goal, i.e. reductions in morbidity of water-borne diseases and morbidity of diarrhoea, with an economic goal, i.e. minimising the costs of the provision of safe water and sanitation facilities. Any, or a combination, of the following alternatives can be adopted to serve such a policy: to reduce the average cost of sanitation coverage, to maintain the real discount rate at as low a rate as possible, to minimise the X-inefficiencies, market failures, and government interventions, and to improve the average productivity of safe water through sanitary behaviour promotion and water quality improvement.