The policy was generally expected to be a success. In 1993, the World Bank listed the USD1.25 poverty headcount ratio at 24.3, while the minimum user fee listed by McIntyre, Bloom, Doherty et al. (1995) exceeded USD2.25. Much of the initial evidence supported this expectation. Registrations were higher in many of the facilities examined, for example, and, despite the fact that much of the initial research covered only a small number of facilities, the effects were observed in rural KwaZulu-Natal, urban Johannesburg, and, even in dental facilities. Furthermore, qualitative research by Walker and Gilson (2004) strongly suggested that the policy had offered benefits to, especially, poorer individuals.
However, part of understanding the effectiveness of a policy is to examine what people do, when faced with a choice. Another part is examining whether or not there were externally beneficial effects. For that reason, the policy was re-examined through the lens of recent advancements in policy evaluation. From a policy evaluation viewpoint, we might expect that user fee abolition at public facilities would lead to increases in the use of public facilities Furthermore, we would find the policy to be relatively more effective if user fee abolition changed behaviour in the following way: rather than not making use of any health care, user fee abolition resulted in individuals accessing the public health care system. One final view of policy success would be the observation that, overall, the population, especially those able to access public health care without paying user fees, would be in better health.
In order to test these possibilities, a few colleagues and I examined health care facility choice decisions, primarily. As part of the analysis, we also examined whether or not we could uncover broad health benefits. The research involved the application of a number of different approaches, depending upon whether we were examining only the 1994 policy change or both the 1994 and 1996 changes. The general conclusion from our research is that, on average, elminating user fees at public facilities did not lead more people to use public facilities, at least when they were sick. We were also not able to uncover any convincing evidence that user fee abolition had increased the health of the population. However, there was one bright spot in our analysis. The bright spot arose from the fact that a focus on the average is often too simplistic. Making use of a rather more flexible empirical approach, we were able to uncover evidence in support of earlier qualitative research. It does appear that, amongst those in the absolute worst socioeconomic circumstances, public health care facilities were more frequently accessed after the abolition of user fees. More importantly, our evidence also suggests that user fee abolition led to a switch in behaviour. Those accessing public facilities would have, before user fee abolition, been more likely not to seek any health care, e.g., treat themselves or possibly lie home in bed. However, even these positive effects are very small.
Brink, Anna S., and Steven F. Koch, “User Fee Abolition in South Africa in 1994 and 1996: Differences-in-Differences.” Economic Research Southern Africa Working Paper 430, May 2014.
Koch, Steven F., and Jeffrey S. Racine, “Healthcare Facility Choice and User Fee Abolition: Regression Discontinuity in a Multinomial Choice Setting.” Economic Research Southern Africa Working Paper No. 373, September 2013.
Koch, Steven F., “The Abolition of User Fees and the Demand for Health Care: Re-evaluating the Impact.” Economic Research Southern Africa Working Paper No. 307, July 2012. (More recent revisions available from the author.)
McIntyre, D., Bloom, G. H., Doherty, J. et al. (1995), “Health Expenditure and Finance in South Africa,” Health Systems Trust and the World Bank, Durban, South Africa and Washington, DC, USA.
Walker, L. & Gilson, L. (2004), “’We are bitter, but we are satisfied’: nurses as street- level bureaucrats in South Africa”, Social Science & Medicine 59(6), 1251–1261.