The International Monetary Fund, Structural Adjustment, and Women's Health: A Cross-National Analysis of Maternal Mortality in Sub-Saharan Africa

AbstractWe conduct a cross-national analysis to test the dependency theory hypothesis that International Monetary Fund structural adjustment adversely impacts maternal mortality in sub-Saharan Africa. We use generalized least square random effects regression models and modified two-step Heckman models that correct for endogeneity using data on 37 African nations with up to four time points (1990, 1995, 2000, and 2005). We find support for our hypothesis, which indicates that sub-Saharan African nations that receive an International Monetary Fund structural adjustment loan tend to have higher levels of maternal mortality than sub-Saharan African nations that do not receive such a loan. This finding remains stable when controlling for endogeneity related to whether or not a sub-Saharan African nation receives a structural adjustment loan. We conclude by discussing the theoretical implications, methodological implications, policy suggestions, and possible directions for future research. NOTESNotes1 See CitationBuchman (1996) for a notable exception. However, the author does not find support for the dependency theory hypothesis that IMF structural adjustment is related to higher levels of maternal mortality in her analysis of a sample of up to 58 low- and middle-income nations for the early 1990s. She attributes this finding to the availability of the maternal mortality data for one point in time.2 The following 36 sub-Saharan African nations are included in the analysis. The years highlighted in bold font denote a country being under structural adjustment. They include Benin (1995, 2000, and 2005), Botswana (1990, 1995, 2000, and 2005), Burkina Faso (1990, 2000, and 2005), Burundi (1990 and 2005), Cameroon (1990, 1995, and 2005), Côte d'Ivoire (1990, 1995, and 2000), Central African Republic (1990), Chad (1990, 2000, and 2005), Comoros (1990 and 1995), Congo (1990, 1995, and 2000), Eritrea (2000), Ethiopia (1995, 2000, and 2005), Gabon (1995), Gambia (1995, 2000, and 2005), Ghana (1990, 1995, 2000, and 2005), Guinea (1990, 1995, 2000, and 2005), Guinea Bissau (2000), Madagascar (2005), Malawi (1990, 1995, 2000, and 2005), Mali (1990, 1995, 2000, and 2005), Mauritania (1990, 1995, 2000, 2005), Mauritius (1990, 1995, 2000, and 2005), Mozambique (1990, 1995, 2000, and 2005), Niger (1990, 1995, 2000, and 2005), Nigeria (1990, 2000, and 2005), Rwanda (1990, 2000, and 2005), Senegal (1990, 1995, 2000, and 2005), Sierra Leone (1990), South Africa (1995, 2000, and 2005), Sudan (1990, 1995, and 2005), Swaziland (1990, 1995, 2000, and 2005), Tanzania (1990 and 1995), Togo (1990, 1995, 2000, and 2005), Uganda (1990, 1995, 2000, and 2005), Zambia (2000 and 2005), and Zimbabwe (1990).3 We do not report the first stage probit models for the sake of space. However, we find that higher levels of multilateral debt service are associated with a greater likelihood of receiving a structural adjustment loan. We also find that higher levels of gross domestic product per capita are associated with a lower likelihood of receiving a structural adjustment loan. The level of democracy, level of conflict within a nation, and dummy variable for whether a sub-Saharan African nation is a U.S. ally do not explain significant variation in the likelihood of receiving a structural adjustment loan. The findings are available from the authors upon request.