Does national-level democratization benefit from external aid? While studies confirm a positive relationship between the two, open questions remain. For example, what are the defining features of a successful democracy promotion campaign? Why do similar countries reap different results? Read on for Ashley Moran’s answers.
By Ashley Moran for Climate Change and African Political Stability (CCAPS) Program
17 May 2016
Alongside the Third Wave of democratization in the latter half of the Twentieth Century, significant scholarly attention has been devoted to understanding the contexts in which democracy may grow or atrophy. This research has reached a degree of consensus about a range of factors that impact democratic consolidation, agreeing that on average democracy grows in countries experiencing economic growth[1] and in those buttressed by democratic neighbors,[2] while democracy is slower to take root in environments with conflict[3] or social cleavages.[4]
An additional line of inquiry began in recent years to assess whether international aid to promote these democratic changes is indeed effective in doing so. Landmark recent studies on U.S. democracy assistance worldwide found that democracy aid overall does have a positive impact on democratic development—and it has the most impact in Africa.[5]
Yet while research to date has made strides in identifying the contextual factors that impact democratic development, and in affirming that democracy aid contributes to that development, it has not identified empirically how democracy aid has brought about this success. The questions remain: what are the crucial defining features of a successful democracy promotion effort? Why does the same level of investment in similarly situated countries not produce comparable democratic gains? This CCAPS study thus seeks to identify the causal mechanisms through which democracy aid positively impacts democratic development in a set of African countries.
Democracy Aid and Security
Understanding what drives the success of democracy promotion programs has significant potential implications for U.S. and international investments in promoting good governance in Africa. The U.S. administration’s FY 2017 budget request for governance assistance in Africa is $343 million—a 96 percent increase from FY 2015 actual figures.[6] And the rising level of U.S. governance assistance seen in Africa is seen globally as well, with the administration requesting $2.7 billion globally for governance assistance in FY 2017, compared to $1.9 billion spent in FY 2015.[7] Having empirical evidence on the causal mechanism through which democracy aid most directly impacts governance quality could help policy planners allocate democracy aid funds for maximum impact.
Moreover, effective governance reform in Africa could reduce the need for external military or humanitarian interventions in fragile African states. The international community often provides humanitarian relief or conflict response in states that lack the capacity to execute these efforts on their own. Nascent democratic governments in Africa face considerable—and growing—challenges from climate change, violent non-state actors, demographic shifts, and other global trends that stress state capacity and resources. Already 19 African countries are labeled as “extreme” or “high” fragility states, with an additional 18 earning a lower but still unstable warning of “moderate” fragility.[8] Democracy aid could provide critical support to build the capacity of these states to handle the diverse challenges they face, bolstering institutional and societal resilience, while reducing the need for U.S. military or humanitarian intervention in the future.
At its core, this study investigates the origin of change in a political system. There is a vigorous academic debate over whether political change is shaped and determined by the formal institutions or the informal norms and practices at work in a political system. This study seeks to inform this debate by assembling empirical evidence on democracy aid programs over time in a range of contexts to examine the causal mechanisms through which democracy aid impacts the political system.
The Democracy Debate
Prior studies and theories on democratic development put forward two competing arguments regarding what shapes change in a political system. These theories, broadly defined, differ in whether they see democratic development as being driven by changes to formal government institutions or by changes to informal practices and norms in the broader political context. The logic behind these theories reflects two distinct assumptions about the causal mechanism through which change occurs in a political system and thus through which democracy aid could potentially impact democratic development.
The first set of theories on democratic development highlights the role of formal institutions in democratic transformation. This literature argues that institutions provide the decisive parameters for guiding political change and overcoming challenges to democratic transition and consolidation. Advocates for this line of reasoning advance institutional solutions to alleviate societal conflict,[9] manage ethnic divisions,[10] construct political identities and promote social cohesion,[11] and generally guide society toward a stable democracy.[12] In this thinking, institutions fill an irreplaceable role in structuring relations in society and offering a credible way to engage the diverse actors that must be involved if the democratic project is to succeed.[13] In this view, formal institutions are thus the central mechanism by which democratic change takes root.[14]
This line of reasoning about the cause of political change could likewise inform assessments of the mechanisms by which democracy aid contributes to this change. In prioritizing the role of institutions, this literature implies that the crucial factor in the success of democracy promotion efforts would be the legal and policy changes promoted in formal institutions, which in turn structure the choices and relations of societal actors, thereby moving them toward democracy. Thus, if this approach is correct, similar countries experiencing successful versus unsuccessful democratic progress should show a divergence in the type of institutional reforms promoted in those countries.
The second set of theories on democratic development has a wider lens, underscoring the influential role played by the broader context within which institutions operate in a political system. This literature moves beyond assessing formal structural and legal changes in institutions to consider how informal practices and norms impact the behavior of institutions and actors in that society. This literature asserts that formal institutions are not as independent as they may appear, with their actions and impact instead being shaped by the broader context within which they are organized in relation to each other. Hall and Taylor, for example, note that the political system, and thus political change, involves both formal and informal processes that link diverse sectors of government and society within a unified structure; this complex combination of factors thus generates ‘distinctive’ outcomes in each location.[15]
A central facet of this second set of literature, then, is its assertion that the same institutional reforms will have different outcomes in different contexts. This underscores the unique context and trajectory of each country—a uniqueness determined by the particular series of historical events and current conditions in that country.[16] Lipset and Rokkan go so far as to argue that formal institutions are inconsequential and that it is instead these key moments in history that set the course and determine future outcomes for each country.[17]
This second line of reasoning about the cause of political change would also frame our understanding of how democracy aid may contribute to such change. In asserting that change arises through the complex set of formal and informal processes that make up a state’s institutional landscape, this literature implies that democratization efforts cannot focus on reforming formal institutions alone, but must also change the informal patterns of behaviour that could impact democratic consolidation. If this literature is correct, similar countries experiencing successful versus unsuccessful democratic progress should likewise diverge in how democracy aid in each country sought to develop those informal democratic norms and domestic support for democratic change.
So which is it: Are formal institutions the lynchpin to democratic development? Or are informal practices and norms equally if not more important? In terms of how democracy aid may influence the process of democratic development, the answer may lie somewhere in between. Yet to understand when and how democracy aid successfully impacts democratic progress—through formal institutions, informal norms, or both—requires an analysis of the strategies that democracy aid programs have employed to advance change in these systems through various formal and informal channels.
Importantly, prior studies on democracy aid have found that the environment in which democratic development advances most consistently is not always the environment in which democracy aid has the most impact. Initial studies found, for example, that democracy aid has a positive impact in countries with high levels of ethnic fragmentation and in countries with low levels of human development[18]—both of which are contexts known to challenge democratic development more generally. This underscores the need to better understand not only what contexts are ripe for democratic development generally, but what contributes to that development at different points in democratic consolidation. The factors that spur democratic advancements in the early days of a democracy—after conflict or under new political and economic institutions—may not be the same dynamics that drive democratic development in the later stages of a democracy. Specific democracy aid interventions, then, may also have greater or lesser effect at these different stages of a democracy.
This research project considers two broad mechanisms theorized to impact democratic development—formal institutions and informal norms—and within this framework examines how democracy aid leverages these mechanisms in varied contexts of democratic development.
Methodology
Parsing out which of these theorized causal mechanisms is at work in successful democracy promotion requires an examination of empirical evidence from democracy promotion efforts in varied contexts and at varied stages of democratic development. This project thus examines the impact of democracy aid in paired countries with similar socioeconomic and political conditions and dissimilar democratic outcomes. The case studies seek to identify which democracy and governance programs—under what conditions—had the most impact in selected African countries and to examine the causal mechanisms at work.
The case studies examine democracy aid programs and strategies implemented for two decades after 1990 by four major donors working in all case study countries—the United States, the African Development Bank, the United Nations Development Programme, and the World Bank—as well as other major donors of democracy aid in each country. This aims to capture the majority of democracy aid in each country while also allowing comparisons of several donors across all countries under study. Most prior studies assessing democracy promotion have considered only a single donor. This study’s consideration of democracy aid from multiple donors aims to provide a more complete picture of democracy aid in each country and thus better prospects at capturing the potential causal mechanism linking democracy aid to changes in a country’s level of democratic development.
Data on democracy aid programs and impacts has been collected through donor program planning and evaluation documents; indicators of democracy aid and democratic development; and indicators of socioeconomic conditions that could impact democratic development.
Research Questions
The research questions aim to assess whether approaches to democracy promotion varied between similar countries that have experienced successful and unsuccessful democratic change. Any variations are then assessed as to whether they provide evidence for different theoretical explanations of what causes political change. In doing so, the study examines the causal mechanisms argued by theories on formal institutions (i.e. whether promoting particular types of institutional changes proved critical) and theories on informal norms (i.e. whether accounting for informal patterns of behavior and interacting contextual factors proved critical).
To achieve this, the study considers three related but distinct arguments found in democracy literature on the role of democratic institutions and norms.[19] This study explores whether there is empirical evidence for any or all of the following arguments:
1. Democracy aid programs that increase the representativeness of formal government institutions will lead to improvements in a country’s democratic development.
2. Democracy aid programs that increase checks and balances across formal government institutions will lead to improvements in a country’s democratic development.
3. Democracy aid programs that foster democratic norms and democratic mobilization will lead to improvements in a country’s democratic development.
The case study design allows comparative analysis of these questions in several ways. It assesses differences in each donor’s approach over time in each country, differences in the donor’s approach across countries sharing similar contexts but different democratic outcomes, and differences in the donor’s approach across country pairs in different contexts. The study also allows comparison of democracy aid strategies across the four donors under study in all case countries.
Case Matching
Because the case studies explore how the design of democracy aid impacts its effectiveness in contributing to the overall level of democracy in a country, it is critical that other factors that could potentially explain variations in democracy levels across countries are as equal as possible between paired cases.
The case matching process thus identified pairs of countries in Africa with similar socioeconomic and political conditions and dissimilar democratic outcomes. Countries were paired using a three-stage statistical matching process.[20] Figures 1 and 2 show a comparison of countries on the variables used to match countries for this study, and the appendix describes the indicators and data used for each variable.
In the first stage, all countries receiving democracy aid in Africa were matched on the similarity of their democracy trends prior to the start of the study in 1990. This included measures of short- and long-term democratic trends. Figure 1[21] shows a sample result from the Guinea-Benin case pairing. In all figures, the selected cases are shown in their respective colors, and the cases that were not chosen are shown in gray to convey the range of possible country values for each indicator in each year. The time window used for matching is shown in the black box, and cases were selected to be similar in these time periods only. The closer the country lines are to each other, the better the match.
In the second stage, countries with the most similar democratic starting points in 1990 were then paired according to their similarity on socioeconomic and political dynamics during the study period. This sought to account for alternative explanations for democratic development and key predictors of democracy aid allocation and effectiveness, including regional democratic diffusion,[22] economic growth,[23] ethnic fractionalization,[24] conflict,[25] trade, and military alliances.[26] This statistical matching process produced a list of prospective case study pairs that were most similar on the matching criteria from these first two stages. Figure 2 shows sample results from the Guinea-Benin case pairing.
Figure 1. First Stage of Matching Process for Case Selection
All countries receiving democracy aid in Africa were matched on the similarity of their short- and long-term democracy trends prior to the start of the study in 1990. Below is a sample result showing the Guinea-Benin case pairing.
Click to enlargeFigure 2. Second Stage of Matching Process for Case Selection
Countries with the most similar democratic starting points in 1990 were then paired based on their similarity on six, equally weighted indicators of socioeconomic and political dynamics during the study period. Below is a sample result showing the Guinea-Benin case pairing.
Click to enlargeSource: All graphs included here were produced by Rich Nielsen based on indicators selected by the author.
In the third stage, the author selected final pairs that also had similar levels of total democracy aid during the two-decade study period (1990-2010) and dissimilar levels of democratic development by 2012.[27] To the extent possible, final pairs were selected such that they also had similar colonial history, Cold War alliances, natural resource wealth, population density, and land size.
The country pairs selected for study represent two different contextual settings shown in scholarly literature to have important influences on democratic development: first, countries recovering from conflict (Burundi and Rwanda) and, second, countries with low levels of ‘human development’ (Benin and Guinea). The study is thus able to explore how democracy promotion efforts have responded to varying degrees of conflict and socioeconomic constraints.
The results of these case studies will be released as a full-length report and set of concise briefs following this initial brief on the study’s methodology.
Prospective Impact
The human and financial stakes are high in democracy promotion. It is not enough to know which contexts are most ripe or hostile for democratic development. Effective democracy promotion also requires knowing how to best design democracy aid programs to work within these external constraints.
Donors must know not only that democracy aid and democratic progress are related in a particular context but also how democracy aid is most effectively designed to meet the particular needs of that context. The hope here is that an empirical analysis of the factors driving political change in diverse contexts can inform the design of democracy aid programs and in turn potentially help maximize their effectiveness in building governance capacity and societal resilience.
Appendix
This study’s case matching process aimed to account for factors outside democracy and governance assistance that could impact variations in countries’ democratic development. Below is a summary of the indicators and data used in each stage of the matching process.
Stage 1: All countries were matched first on the similarity of their democratic trends prior to the start of the study in 1990. This included measures of short- and long-term democratic trends.
Click to enlargeStage 2: Countries with the most similar democratic starting points in 1990 were then paired based on six, equally weighted indicators. These sought to control for alternative explanations for democratic development and key predictors of democracy aid allocation and effectiveness. Data were used for the most complete range of years available for each indicator.
Click to enlarge
Stage 3: The computer-assisted matching done in Stages 1 and 2 produced a list of prospective case study pairs. Final pairs were then selected that have similar levels of total democratic assistance and similar conditions on a broader range of pertinent socioeconomic and political factors but dissimilar democratic outcomes (see below).
Click to enlarge
[1] Studies using a range of methods have produced general agreement that economic development strengthens democracy. Przeworski and Limongi find that, while economic growth is not related to transitions to democracy, it does strengthen existing democracies; see Adam Przeworski and Fernando Limongi, “Modernization: Theories and Facts,” World Politics 49 (January 1997): 155-183. These results have been reinforced by subsequent studies, for example by Adam Przeworski et al., Democracy and Development: Political Institutions and Well-Being in the World, 1950-1990 (New York: Cambridge University Press, 2000); and Kristian Skrede Gleditsch and Michael D. Ward, “Diffusion and the International Context of Democratization,” International Organization 60, 4 (2006): 911-933.
[2] Prior studies using a variety of methods and datasets have provided consistent findings that democratic diffusion from neighboring countries has a positive impact on a country’s democratic development. For example, Gleditsch and Ward, 917-921, 924-925; and Daniel Brinks and Michael Coppedge, “Diffusion Is No Illusion: Neighbor Emulation in the Third Wave of Democracy,” Comparative Political Studies 39, 4 (2006): 463-489.
[3] Gleditsch and Ward find that democracies are more likely to survive in contexts with lower conflict, 924, 926.
[4] Varied studies assert that a high level of ethnic division in a country can present challenges for democratic consolidation. For an overview of the areas of agreement on this relationship, see Arend Lijphart, “The Wave of Power-Sharing Democracy,” in The Architecture of Democracy: Constitutional Design, Conflict Management and Democracy, ed. Andrew Reynolds (Oxford: Oxford University Press, 2002), 38-9. For a challenge to studies asserting a relationship between democratic consolidation and ethnic fragmentation, see Steven M. Fish and Robin S. Brooks, “Does Diversity Hurt Democracy?” Journal of Democracy 15, 1 (2004): 154-166.
[5] Two quantitative studies analyzed democracy assistance from the U.S. Agency for International Development from 1990 through 2003 and 2004, respectively. See Steven E. Finkel, Aníbal Pérez Liñan, and Mitchell A. Seligson, “The Effects of U.S. Foreign Assistance on Democracy Building, 1990-2003,” World Politics 59, 3 (2007): 404-440; and Steven E. Finkel et al., Deepening Our Understanding of the Effects of U.S. Foreign Assistance on Democracy Building, Final Report (Washington: U.S. Agency for International Development, 2008).
[6] FY 2017 Congressional Budget Justification, Foreign Operations Appendix 3 (Washington: U.S. Department of State, 2016), 541-558.
[7] Ibid., 518.
[8] Monty G. Marshall and Benjamin R. Cole, Table 1: State Fragility Index and Matrix 2014 (Vienna: Center for Systemic Peace, 2014).
[9] Torben Iverson and David Soskice, “Distribution and Redistribution: The Shadow of the Nineteenth Century,” World Politics 61, 3 (2009): 438-86.
[10] Donald L. Horowitz, “Democracy in Divided Societies,” Journal of democracy 4, 4 (1993): 18-38; Lijphart 2002; and Arend Lijphart, “Constitutional Design for Divided Societies,” Journal of Democracy 15, 2 (2004): 96-109.
[11] Robert Bates, Rui de Figueiredo, and Barry Weingast, “The Politics of Interpretation: Rationality, Culture, and Transition,” Politics and Society 26, 4 (1998): 603-642.
[12] Kenneth Shepsle, “Studying Institutions: Some Lessons from the Rational Choice Approach,” Journal of Theoretical Politics 1, 2 (1989): 131-47.
[13] Daron Acemoglu and James Robinson, Economic Origins of Dictatorship and Democracy (Cambridge: Cambridge University Press, 2005); and Adam Przeworski, “Democracy as a Contingent Outcome of Conflicts,” in Constitutionalism and Democracy, ed. Jon Elster and Rune Slagstad (Cambridge: Cambridge University Press, 1988), 59-80.
[14] Gary Cox, “Electoral Rules and Electoral Coordination,” Annual Review of Political Science 2 (1999): 145-161.
[15] Peter Hall and Rosemary Taylor, “Political Science and the Three New Institutionalisms,” Political Studies 44, 4 (1996), 937. Hall and Taylor describe the institutional organization of a state as the “formal or informal procedures, routines, norms and conventions embedded in the organizational structure of the polity” that structure “relations among legislators, organized interests, the electorate and the judiciary,” 938.
[16] Hall and Taylor; and Paul Pierson, “Increasing Returns, Path Dependence, and the Study of Politics,” American Political Science Review 94, 2 (2000): 251-267.
[17] Seymour Martin Lipset and Stein Rokkan, “Cleavage Structures, Party Systems, and Voter Alignments,” in Party Systems and Voter Alignments, ed. Seymour Martin Lipset and Stein Rokkan (New York: Free Press, 1967).
[18] Finkel, Liñan, and Seligson; and Finkel et al.
[19] These hypotheses were developed by the full research team comprised of Ashley Moran, Brooke Russell, Dan Robles-Olsen, Matt Deal, and Rob Wieczorek.
[20] Case matching was implemented in collaboration with Rich Nielsen, based on a computer-assisted case matching methodology he developed to leverage large amounts of data in identifying the best statistical matches for paired comparisons. For the methodology underlying his computer-assisted matching, see Rich Nielsen, “Case Selection via Matching,” Sociological Methods & Research, online release (October 27, 2014): 1-29. For the present study, the author selected indicators for use in case matching based on a review of relevant literature, then Nielsen applied his model in the first two stages of case matching described in this brief. The author alone, however, is responsible for any shortcomings or omissions in the data or research design.
[21] For the methodology underlying this computer-assisted matching, see Rich Nielsen, “Case Selection via Matching,” Sociological Methods & Research, online release (October 27, 2014): 1-29.
[22] Prior studies have shown that democratic diffusion is associated with democratic development. See Gleditsch and Ward; and Brinks and Coppedge.
[23] Prior studies have found consistent links between economic growth and democratic development; see Przeworski and Limongi, Przeworski at al., and Gleditsch and Ward. Further, studies have found that democracy aid is more effective in countries with low GDP (see Finkel et al.) and that low GDP is a predictor of democracy aid allocation; for the latter, see Richard Nielsen and Daniel Nielson, “Triage for Democracy: Selection Effects in Governance Aid,” unpublished manuscript, August 2010; and Mike Findley et al., “Preliminary Results: Finding the Bright Spots in USAID Democracy and Governance Assistance to Africa,” unpublished manuscript developed for the Robert Strauss Center, April 2011.
[24] A prior study found that democracy aid is more effective in countries with high ethnic fractionalization; see Finkel et al.
[25] A prior study found that democracy aid is less effective in conditions of political conflict; see Finkel et al.
[26] Prior studies found that trade and military alliances are predictors of democracy aid allocation (Nielsen and Nielson) across many types of democracy aid (Findley et al.), with trade partners receiving more democracy aid and military alliance partners receiving less democracy aid.
[27] “Total democracy aid” was based on the amount of democracy and good governance aid committed from 1990 to 2010, as reported by AidData, AidData 2.2 Provisional Dataset, accessed September 12, 2013. The “level of democratic development” was assessed using Polity IV and Freedom House scores.
For more information on issues and events that shape our world, please visit the ISN Blog or browse our resources.
Ashley Moran is associate director of the Robert S Strauss Center for International Security and Law and leads the democratic governance research team under the Center’s CCAPS program.
Copyright