Party Organization and Women’s Empowerment: A Field Experiment in Ghana

• In Event: Experimental Studies on the Grassroots Participation and Representation of Women

Abstract

Women are currently under-represented in political parties in many new democracies, in part because of social norms that inhibit women from becoming active in local politics. What are the effects of reducing this gender gap in grassroots participation? We examine this question by presenting results from a randomized field experiment conducted around the December 2016 elections in Ghana. The treatment in our experiment is a large community meeting presided over by the traditional chief (a “durbar” in local parlance), held in rural villages by Ghana’s National Commission for Civic Education (NCCE), an independent, non-partisan government agency that regularly conducts civic education programs. The goal of the civic education treatment is to address social norms against the participation of women in political parties and encourage greater women’s participation in each community before the election. The treatment allows us to study the downstream effects of increased women’s political participation on (a) political party strategies, including mobilization and outreach to women before the election; (b) the subsequent economic welfare of women; and (c) social norms about the participation of women in politics and local leadership.

Our paper reports the results from this randomized field experiment, a unique opportunity to understand how the resources available to political parties affect their electoral and linkage strategies in new democracies, a long-standing question of interest to scholars seeking to explain parties’ use of clientelism, fraud, and intimidation, and the forces shaping the emergence of more programmatic political competition (e.g., Aldrich 1995; Hale 2005; Kitschelt 1999, 2000; Kitschelt & Wilkinson 2007; Shefter 1977; Stokes et al. 2013; Tavits 2013). Particularly important in new democracies in developing countries are labor resources – the grassroots members and “brokers” capable of mobilizing or persuading voters through networks of personal relationships. It has been difficult to assess how these resources affect party behavior, however, as parties often simultaneously differ on many dimensions, and political competition affects both the resources available to a party and the mix of strategies it selects. We address this issue by randomizing treatment over rural communities to exogenously increase the labor supply of women in local party branches that begin with very similar initial resource endowments.

By exogenously increasing the number of women active in party organizations at the local level, we also study how political empowerment affects the well-being of a marginalized population. Recent “ancillary studies” (Baldwin & Bhavnani 2015) of field experiments in economics on education, credit, and conditional cash transfers have expanded our understanding of the impacts of economic empowerment on political attitudes and behavior (e.g., De La O 2013, Friedman et al. 2016). But there have been relatively few experimental studies of the reverse relationship: the economic effects of political empowerment, and of women’s political empowerment, in particular. We explore whether women’s political participation reduces their economic vulnerability and improves their access to resources.

Moreover, our paper extends the literature on gender and politics in the developing world. Existing research has focused on the impacts of increased representation of women in leadership positions, particularly through reserved seats and quotas (e.g., Chattopadhyay and Duflo 2004, Duflo 2005, 2012, Tripp and Kang 2008, Krook 2009, Clayton 2015). These studies assume that political empowerment affects women’s well-being by better aligning the preferences of women and policymakers. Rather than focus on leadership positions, our study focuses on the much less-studied empowerment of women at the grassroots level.

The field experiment takes place in 44 communities in 5 rural districts in Ghana and draws on data collected from a three-wave panel survey conducted in 2016 and 2017, before and after the country’s 2016 presidential and parliamentary elections.

Authors

• Noah Nathan, University of Michigan

• Nahomi Ichino, University of Michigan, Ann Arbor