Coups d’État and Foreign Aid

Summary

Do international donors penalize coups d’état by reducing aid? How significant is the impact of coups on aid flows? These questions have become increasingly important over the past three decades as the concept of political conditionality has gradually permeated the donor community, pushing for stringent actions to be taken against democratic transgressions like coups. I argue that the end of the Cold War was a historical juncture that reshaped the international donor community’s aid-based sanctioning policy toward coups. However, I also posit that the U.S. does not comply with the growing international norm of political conditionality due to its geopolitical interests trumping its rhetorical commitment to penalizing coups. This paper exploits exogenous variation in the success and failure of coups to estimate the causal effect of coup-led regime change on aid flows. My empirical evidence supports the preposition that since the end of the Cold War, the donor community on average has reduced the amounts of aid disbursements in response to coups d’état although the U.S. has been inconsistent in applying aid sanctions against coups both during the Cold War and post-Cold War periods. While demonstrating a genuine shift in the international community’s collective responses toward coups since the end of the Cold War, my findings also attest to potential heterogeneity across major bilateral donors, which may undermine the overall effectiveness of aid and political conditionality.

Introduction

On March 21, 2012, military junior officers staged a coup d’état to oust then President Amadou Toumani Toure in Mali. Before the coup, Mali had been hailed as one of the most stable democracies in sub-Saharan Africa. This sudden transgression against democratic rule in Mali shocked the international community, which unequivocally condemned the coup d’état and called for the immediate restoration of civilian rule. The country’s two largest bilateral donors, the U.S. and France froze their development assistance to Mali immediately after the coup (Djau, 2012). Accordingly, major multilateral donors, including the European Union (EU), World Bank, and African Development Bank, all followed suit by announcing the suspension of their aid to the country (Hicks, 2012).

These immediate, harsh responses by international donors suggest that political conditionality is genuinely at work.¹ However, such coordinated actions by donors are often thought to be exceptional. Anecdotal evidence paints a bleak picture of donors’ weak commitment to enforcing political conditionality. The most recent case comes from the July 2013 coup in Egypt, which led to only a temporary freeze on the U.S. economic and military aid to the Egyptian government. Only six months after the coup, the U.S. Congress passed legislation that lifted restrictions on its development and military assistance to Egypt (Londono, 2014).

Do international donors penalize coups d’état by reducing their aid? How significant is the impact of coups on aid flows? Do donors uniformly punish coup perpetrators? These questions are particularly relevant to policymakers who use aid as a means to promote democracy and good governance because the effectiveness of aid sanctions and political conditionality rests on the assumption that donors can act uniformly to exert collective pressures on coup perpetrators to restore civilian/democratic rule. This paper seeks to answer these empirical questions by analyzing the global sample of coups and foreign aid disbursements for the period of 1960–2012. I argue that the end of the Cold War was a historical juncture that reshaped the donor community’s aid-based sanctioning policy against coups. The end of Cold War politics coincided with the beginning of donors’ collective efforts to reward democracy and good governance through their aid policy, which called for most stringent actions to be taken against the incidence of coups. However, I posit that geopolitical interests have trumped the U.S. rhetorical commitment to penalizing coups, thus making American aid sanctions against coups inconsistent.

My empirical analysis shows that since the end of the Cold War, the donor community on average has collectively penalized coups d’état by reducing aid. In contrast, I find little evidence of such aid sanctions undertaken against coup-led regime change during the Cold War period. These findings run counter to the prevailing notion in the literature that there has been no notable change in donors’ actual aid allocation patterns over time (e.g., Clist, 2011, Easterly, 2002, Easterly, 2007, Kanbur, 2000, Svensson, 2003). In particular, Easterly (2007, p. 649) finds “no evidence that the political opportunity created by the end of the Cold War led to a de-emphasis of strategic considerations and more emphasis on the need of the recipient.” My findings indeed reflect a fundamental change in the donor community’s collective response to coups d’état around the late 1980s and the early 1990s, a period when Cold War politics ceased to dictate donors’ aid allocation decisions, and concerns related to the domestic politics of recipient countries carried more weight in shaping their aid policy.

However, I also find that there is some degree of heterogeneity across major bilateral donors in their individual responses to coups. In particular, my findings show that the U.S.—the largest donor in the world—has failed to impose strict and consistent aid sanctions against coup perpetrators even during the post-Cold War period. As Marinov and Goemans (2013) suggest, foreign aid can become a catalyst for expediting the timing of elections that follow coups. However, if donors fail to uniformly act in punishing coups d’état, coup perpetrators can potentially evade international pressures, which could otherwise force them to relinquish power and return to democratic/civilian rule. The main findings of this paper thus attest to the possibility that the uncoordinated nature of aid sanctions can potentially jeopardize the overall effectiveness of political conditionality.

This paper seeks to contribute to the literature on foreign aid and political conditionality in three ways. First, this article examines how donors respond to a particular type of democratic transgression—coups d’état. Coups instantly shatter the confidence donors have in the democratic institutions of their recipient countries and thus serve as a critical test for donors’ real commitment to political conditionality. Whereas studies on the relationship between democracy and aid allocation abound (e.g., Alesina and Dollar, 2000, Clist, 2011, Dollar and Levin, 2006, Easterly, 2007, Hyde and Boulding, 2008, Neumayer, 2003a, Wright and Winters, 2010), there have been few scholarly efforts to empirically test the relationship between coups and foreign aid. Most of the existing studies on aid sanctions examine only a few instances of coups to illustrate how the concept of political conditionality has been enforced in certain cases (e.g., Barracca, 2007, Crawford, 2001, Santiso, 2002, Stokke, 1995, Youngs, 2010).² They offer little leverage on the question of whether coups actually cause any significant aid cuts from the donor community. This article seeks to fill this gap in the literature by quantitatively testing donors’ responses to coups d’état.

Second, I estimate the causal effect of coup-led regime change on foreign aid. The occurrence of a coup is most likely to be correlated with various other potential confounders—such as regime legitimacy, state capacity, political stability, or the presence of a politicized military—which may also be a strong predictor of whether a given country receives aid or how much it receives. Many of these potential confounders cannot be easily measured, and the issues of endogeneity are most likely lurking in studies that seek to uncover the relationship between coups and foreign aid. Following Jones and Olken (2009), I will address this problem by focusing on the sample of observations in which coup attempts have already been made. I argue—and provide extensive evidence—that conditional on the occurrence of coup attempts, whether a coup succeeds in unseating the government is exogenous to aid flows. I exploit this conditional exogenous variation in the success (or failure) of attempted coups to estimate the causal effect of coup-led regime change on foreign aid.

Lastly, I consider potential heterogeneity across major bilateral donors in their responses to coups. As existing studies show, donors are far from homogeneous in terms of the way they allocate their aid (Alesina and Dollar, 2000, Dollar and Levin, 2006, Neumayer, 2003a). Various strategic considerations influence their aid allocation decisions. Reflecting this theme, do donors all respond differently to coups? Or do they uniformly reduce their aid to coup perpetrators and penalize them? As Faust (2013, p. 1) points out, for political conditionality to be effective, “donor harmonization is key, because only a coherent incentive system sets credible signals and has a chance to contribute to institutional reform.” Thus, I carefully evaluate how each of the five largest bilateral donors—namely, the U.S., UK, Germany, Japan, and France—responds to coups d’état.

The remainder of the paper proceeds as follows. In the next section (Section 2), I review the literature on coups and foreign aid and provide a theoretical overview of how the concept of political conditionality has gained currency in the donor community since the end of the Cold War. Section 3 describes the data used in my empirical evaluation of the relationship between coups d’état and foreign aid. Section 4 presents the main findings of my econometric analysis. Section 5 concludes with a description of several policy implications that can be drawn from the study’s core findings.