Gender difference in support for Democracy in Sub-Saharan Africa: Do social institutions matter?

ABSTRACT

Several recent papers have noted gender differences in support for democracy in Africa, but the causes of this difference remain unclear. This article investigates whether the observed gender gap is due to the related gender inequality in social institutions, which affects women's daily life and deprives them of social and economic empowerment inside and outside the home. Using Afrobarometer survey data (rounds 2 [2002–3], 3 [2004–5], and 4 [2008–9]), the study finds that the gender difference in support for democracy is no longer significant once gender discrimination is controlled for in the family code, physical integrity, or civil liberties components of the Social Institutions and Gender Index (SIGI). Interaction terms show that women's support for democracy is only lower in places where gender inequality in these social institutions is particularly large. This study thus provides evidence that women who live in countries with favorable institutions toward women are more supportive of democracy than women who do not.

KEYWORDS:

• Support for democracy
• gender difference
• social institutions

JEL Codes:

• J16
• O120
• O38

NOTES ON CONTRIBUTORS

Maty Konte is Research Fellow on Governance and Development Economics at the United Nations University-MERIT in Maastricht, the Netherlands.

Stephan Klasen is Professor of Development Economics at the University of Göttingen, Germany.

ACKNOWLEDGMENTS

We would like to thank Tony Addison, Cecilia Garcia-Peñalosa, Seif Dendir, and James Thurlow for their helpful comments. Maty Konte is grateful for financial support from UNU-MERIT and UNU-WIDER.

Notes

1 The definition of legitimacy attitudes by Lipset (1963) is “Belief that the existing political institutions are the most appropriate ones for the society” (Matthew D. Fails and Heather N. Pierce [2010]): 64). Larry Diamond (1999) argues that the stability of democratic systems requires a belief in the legitimacy of democracy by people. Moreover, Robert Mattes and Michael Bratton (2007) report that “No matter how well or badly international aid donors or academic think tanks rate the extent of democracy in a given country, this form of regime will only consolidate if ordinary people believe that democracy is being supplied” (192).

2 Among the different dimensions of gender equality we can note economic participation, health and well-being, political empowerment, and education attainment.

3 Round 2 data for Zimbabwe is from 2004.

4 For the intertemporal sample we combine versions 2009 and 2012, noting that there are comparability issues across these two verisons of the SIGI. See below for details.

5 Furthermore, Evans and Rose (2007a : iii) have figured out mechanisms through which education affects support for democracy and argued that “the mechanisms through which schooling influences democratic support relate to cognitive elements of political comprehension and involvement that are consistent with an intrinsic model of the effect of education on democratic values and outcomes rather than a view of education as a marker of resource inequalities.”

6 See also M. Najeeb Shafiq (2010) for further investigation of the impact of education on support for democracy in other developing countries. Using the Pew Global Attitudes Project surveys, Shafiq (2010) finds that education has a strong effect on support for democracy in Lebanon, Jordan, and Pakistan.

7 Using the World Value Surveys, Rowley and Smith (2009) find that predominantly Muslim countries have a higher degree of support for democracy than other countries. Maseland and van Hoorn (2011) challenge the paradoxical finding of high support yet little experience of democracy in Muslim countries, arguing that the positive attitudes of citizens in Muslim countries toward democracy are not limited to Muslim countries and can be explained by the theory of decreasing marginal utility, which suggests that people more highly value scarcer goods.

8 For example, one can note the analysis by Michelle J. Hindin (2000) for a case study in Zimbabwe, Anastasia J. Gage (1995) for Togo, Deborah Balk (1994) for a case study in Bangladesh, among others.

9 Further details on the data are available at: http://www.afrobarometer.org/survey.

10 It refers to question number 37 for round 3, and 38 for round 2.

11 Similar results using the alternative proxies for support for democracy are found but not shown here because of space contraints. Readers are invited to refer to the working paper version, Maty Konte (2014) available at the UNU-MERIT Working Paper series.

12 To look at whether the degree of support for democracy is associated with the people's experience, we group individuals into three different age categories: those who are between ages 18 and 25 (27 percent of the sample), people between ages 26 and 35 (29 percent), and people older than 35 (43 percent of the sample). For the place of residence we have distinguished between people living in rural areas (63 percent) versus urban areas (36 percent). Employment status has three categories: inactive, accounting for 31 percent of the sample, and active, sorted into unemployed (34 percent) and employed (33 percent). To measure access to media, we consider separately access to news from radio, from TV, and from newspapers. For each, the variable access to media is a dummy equal to 0 if the individual attests never having had access to media from the given source, and 1 otherwise. In the sample, almost 87 percent have access to news from radio, against 54 percent for TV. Indeed, access to TV remains costly in developing countries, especially for people leaving in rural areas. Finally, only 40.61 percent have access to news from newspapers, a number which is not surprising given the fact that reading newspapers requires some level of education, yet in this dataset 20 percent of the people do not have formal schooling, and 18 percent have not completed their primary degree.

13 Further details on the SIGI index can be found in Branisa, Klasen, and Ziegler (2009).

14 Round 3 contains only eighteen countries since it excludes Burkina Faso and Liberia, while round 2 includes neither these two countries nor Benin or Madagascar. In the combined data, 65.7 percent of the population support democracy against 34.26 percent who do not.

15 For round 3 the gender gap is equal to 12.84 points; it decreases to 7.4 in round 2.