The Political Economy of Improving Health Care for the Poor in Rural Africa: Institutional Solutions to the Principal–Agent Problem

Abstract

Both governments and private for-profit markets have been disappointing in meeting the needs of the African poor for health care. NGO services provide a much more attractive alternative for this clientele, despite the fees they charge. They do so because they represent an institutional solution to the ‘imperfect information’ problem in health care. Through simulations based on data from Cameroon, we demonstrate that if fee-charging NGOs replace the highly subsidised but poorly managed facilities operated by African governments the poor would be better off. Those NGOs that are decentralised in their financial and personnel management are most effective. The politics of making the recommended changes are assessed.

Notes

Kenneth L. Leonard, Department of Agricultural and Resource Economics, University of Maryland. E-mail: [email protected]. His research was funded by NSF grant 94-22768 and a University of California Rocca Fellowship. David K. Leonard, Department of Political Science, University of California, Berkeley. E-mail: [email protected]. His research was funded by NSF grant 94-22768 and the Ministry of Foreign Affairs of the Netherlands. This paper was prepared for the conference on New Institutional Theory, Institutional Reform and Poverty Reduction, Development Studies Institute, London School of Economics, 7–8 September 2000.

Russell [ 1996 ] goes so far as to suggest that despite patients' persistence in purchasing health care, they cannot actually afford it. That purchases of health care come at the cost of other purchases is not surprising, nor does it present any justification for government subsidy of health costs.

Ndeso-Atanga [ 2000 ] shows that patients in Cameroon are much more willing to use government services for delivery than for other services and that as the outcome is clear before they leave the clinic they always pay an extra ‘appreciation’ to attendants when the outcome is successful. This evidence is not proof but highly suggestive.

Leonard [ 2003 ] shows that traditional healers are able to verify outcomes because patients believe the healers are agents of higher powers and able to punish them if they lie. This allows them to offer outcome based contracts for a variety of services.

We observe that in Tanzania many private practitioners with reputations for quality spent much of their early career in the government service where they were able to develop a reputation while being paid by the government. Thus they can reap the benefits of reputation without paying the full costs. This is a form of serial moonlighting.

For historical reasons, the overwhelming majority of voluntary organisations providing health services in Africa are religious, that is, missions. We refer to them by the more generic term NGO because we see no reason why other value based organisations with substantial reputations at stake, such as Medecins Sans Frontieres, should not perform as well as missions do in Africa. Indeed anecdotal observation suggests that they do equally well.

The distinctions between the different forms of decentralisation – devolution, deconcentration, delegation, and privatisation – are further explained in Leonard and Marshall [ 1982 : Ch.1].

We created an estimated household wealth measure by regressing the total declared annual income of the household on a number of household characteristics, such as means of employment, type of roof or floor, education level etc. and use the estimated household income as our measure of wealth. Thus, for example, if no one in the household had a good source of income, but there was a good outside source of money they were not seen as poor.

Additional information

Notes on contributors

Kenneth L Leonard

Kenneth L. Leonard, Department of Agricultural and Resource Economics, University of Maryland. E-mail: [email protected]. His research was funded by NSF grant 94-22768 and a University of California Rocca Fellowship. David K. Leonard, Department of Political Science, University of California, Berkeley. E-mail: [email protected]. His research was funded by NSF grant 94-22768 and the Ministry of Foreign Affairs of the Netherlands. This paper was prepared for the conference on New Institutional Theory, Institutional Reform and Poverty Reduction, Development Studies Institute, London School of Economics, 7–8 September 2000.