Can an economic case be made for investing in health?BMJ 1998; 316 doi: https://doi.org/10.1136/bmj.316.7147.1762 (Published 13 June 1998) Cite this as: BMJ 1998;316:1762
No, but it's the wrong question
- Charles Normand, Professor of health policy
Social insurance for health services grew out of voluntary schemes to ensure access to care for workers, especially to provide treatment for industrial injuries.1 At least part of the reason was a desire to help workers back to work. Implicit in some of the pronouncements from the World Bank is a belief that health services in developing countries should be used to treat and thus restore the productivity of workers. 2 3 Increasing output can be used as one measure of benefits in cost benefit analysis,4 but this gets no mention in some recent texts on health economics,5 and it is now rare to value benefits in economic evaluation in terms of improved productivity. As government services are increasingly being judged on their ability to contribute to economic growth—for example, crudely testable skills seem to be the main objectives of primary education—it is interesting to consider if a similar case can (or should) be made for health services.
For health services to contribute to economic growth certain conditions must apply. Firstly, treatment must contribute to health status in such a way that the person's productivity is increased (or restored). Secondly, that person working must increase overall output in the economy. The first is not guaranteed, since the treatment may improve quality of life or survival without increasing productivity, and the person treated may not be in the workforce. The second depends either on there being no unemployment (and in Britain there are up to three million reasons for questioning that) or the person having exclusive skills (only sometimes the case). If someone being sick allows a previously unemployed person to work then there may be no loss in production and no gain when the person returns to work. In practice there is likely to be some loss owing to lower productivity of temporary or inexperienced staff.
In a recent Office of Health Economics lecture, Baumol pointed out that health service provision is a handicraft industry, which will always be labour intensive. This means that relatively little scope exists for improved productivity, and costs are likely to rise more quickly than in manufacturing. He pointed out that this is a normal and desirable state of affairs. Of course, this does not mean that potential efficiency improvements should not be pursued,6 but expansion of handicraft industries tends to slow economic growth.
Clearly, productivity and growth (as conventionally measured) are unlikely to be sufficiently increased by health services for this to be the main justification of healthcare expenditure. In any case, this is not really the right question. The proper objective should not be higher growth in itself but more generally increased welfare of the population. The economic case for health services is made when this is the use of resources that has the greatest impact on welfare. An important lesson of the 1970s and 1980s is that economic indicators can show a large improvement without an equivalent improvement being felt widely in the population. All economists agree that gross domestic product is at best a crude measure of welfare, and its continued use reflects the difficulty of agreeing to use anything else.
In assessing priorities for development the United Nations Development Programme has advocated a wider definition of welfare that has a strong focus on health.7 Since the measures of national income currently used do not include health directly, and since the indirect effects of better health on measured gross domestic product are probably weak, it is difficult to see how this kind of economic case for investing in health and health care can be made. Equally, since health (along with other important indicators of welfare) is not part of the broad measure of national income, we can question the pursuit of this narrow objective. There is no point in increasing measured national income unless this helps to improve national welfare. Indeed, growth that lowers health may lower national welfare. Perhaps the question should not be, Can we justify investment in health on the basis of contribution to growth? but rather, Can we justify growth on the basis of its contribution to health?