Our study speaks to theoretical questions about the efficacy of conditionality in promoting human rights and democracy. The timing of the observed effects (i.e., nearly immediate and short‐lived) suggests that positive conditionality is operative in promoting these reforms. Taken together with qualitative evidence on the nature of the interaction between the Commission and recipient countries, our results indicate that the Commission is able to use the increase in aid commitments as leverage to incentivize reforms in the areas of human rights and democracy. Recipient countries then respond to these incentives by rapidly altering their behavior. Their reforms are short‐lived because the aid boost only provides the Commission with greater leverage temporarily, and once the increased ability to incentivize reforms with disbursements is removed, countries revert to prior behavior. Our findings point to important conditions for the efficacy of multilateral aid, informing the debate on the impact of the international community's promotion of democracy and human rights in recipient countries.

We develop a novel approach to deal with the hitherto intractable problem of endogenous aid allocation, helping to unpack how aid works as part of a complex interaction between international institutions and recipient countries. By focusing on a theoretically relevant institutional provider of large amounts of development assistance, we leverage features of the policymaking procedure to recover quasi‐experimental variation in aid allocation. We identify an as‐if random process that drives aid allocation but is otherwise unrelated to rights and democracy in recipient countries: the rotating presidency of the Council of the European Union. Since the country holding the presidency is determined by a known and essentially random rotation principle, so too is the set of countries that happens to include the country holding the presidency's former colonies. We show that when a country's former colonizer holds the presidency during the budget‐making period, that country is allocated considerably more European Union (EU) aid. The EU's aid allocation system is highly bureaucratic; while the Council budgets the aid, a separate entity—the EU Commission—is responsible for its disbursement. We find that when the Commission is given more aid designated for particular countries, recipient countries then respond to these incentives by immediately undergoing visible reforms, but not necessarily altering institutions. Instead, much as the increase in aid is short‐lived, so too are its effects.

The view that good governance is primarily a domestic affair has lost ground to an alternative view—that of an international community actively shaping rights and freedoms in states around the world. 1 A growing body of research points to systems of governance that are not sui generis but are substantially affected by international processes, yet the role of one of the international community's primary tools of influence, foreign aid, is a subject of considerable controversy. A central issue is that aid allocations are not randomly assigned; donors give aid for reasons that are not independent of rights and democracy in recipient countries. Many motivations associated with foreign aid provision are likely unobservable, which can lead to bias in estimating the efficacy of aid in the absence of a clearly defined source of identification. 2

The European Union and Positive Conditionality

Our study reconsiders a fundamental question in the foreign aid literature: Does aid serve as an effective incentive for democratization and respect for human rights in recipient countries?3 The efficacy of foreign aid, in general, is widely contested. While some work suggests a positive relationship between aid and political liberalization, particularly in the post–Cold War period (Bermeo 2016; Dunning 2004; Fearon, Humphreys, and Weinstein 2009),4 other scholars argue that foreign aid has either no effect (Knack 2004) or a negative effect on rights and democracy (Alesina and Dollar 2000; Burnside and Dollar 2000; Dreher et al. 2013; Schraeder, Hook, and Taylor 1998). Still others suggest that aid has a conditional effect, which depends on the incentives of external and domestic actors (Krasner and Weinstein 2014, 126).

This article focuses on EU aid, asking whether countries receiving more of this aid are more likely to democratize and to respect the rights of their citizens.5 As a multilateral institution dedicated to the promotion of democracy and human rights by virtue of its founding and subsequent treaties, the EU sits at the intersection of debates on democratization, democratic conditionality, and delegation. Further, as an important actor in world affairs, the EU provides large amounts of aid to a wide array of countries, which is often conditioned on a recipient's respect for human rights and level of democracy (Dunning 2004).

Indeed, since the end of the Cold War, the EU has emphasized positive conditionality in its foreign relations, demanding respect for human rights and democracy in return for foreign aid and other benefits. The idea is that a bigger pot of money can provide ruling elites with a greater incentive to accommodate at least some of the Commission's demands for reform as the price of receiving it. Governments often rely on aid to maintain patronage and services, and thus additional aid can help them to stay in office. Or, governments may desire aid simply because they value having more money to spend. Offering greater amounts of aid as a reward for liberalization might thus allow the Commission to better extract changes, or can even permit it to request additional reforms; states may be willing to pay a greater cost to reap a greater benefit. Further, even if recipient governments do not implement reforms sincerely, making policy changes can alter domestic expectations, creating pressure to open up further (Dai 2007). Alternatively, the public may not favor reforms, but might tolerate them in order to receive the services that the aid can bring. Thus, just as the EU successfully used the large prize of EU membership as a carrot to induce democratic changes (Pop‐Eleches 2007), it may be able to hold out aid to a similar effect.

While such pressure aims to alter behavior, target countries may respond either by adjusting their actions or by changing the institutions producing such behavior. The latter would yield more durable changes, but it would be costlier to implement than the former. Hence, larger bribes, and greater tenacity in outside cajoling and monitoring, are often required to facilitate institutional change, without which such behavioral changes may be short‐lived. It remains an open question whether EU aid is able to spark any changes, and, if so, whether it represents a large enough incentive to alter institutions.

Of course, recipients may differ in their propensity to make such changes, along with the types of reforms they undergo. Put differently, increased available aid may not impact all potential recipients equally. While there exist a variety of potential ways in which aid could impact states differentially, we highlight the impact of alternative revenue sources as a particularly interesting factor. When states have funding outside of EU aid, they may value aid from the EU less and thus be less willing to respond to EU pressure. Instead, states may simply rely on other donors or revenue streams if the EU does not provide the promised aid as a result. Knowing that these states have such an option, the EU may impose less stringent conditions in the first place.6

Generally speaking, however, EU aid may be more effective at inducing reforms than aid from other Western donors. Indeed, the EU's multilateral nature and delegation of nontrivial powers to a network of supranational institutions may make its conditional commitments more credible, as delegating aid provision to international organizations can result in aid flows that are relatively free of short‐term domestic political pressures (Milner 2006; Rodrik 1995). Further, the EU designates much of its aid for specific projects designed to strengthen political reforms in recipient countries, which can affect rights and democracy directly by building civil society or strengthening institutions.

However, the EU's efforts to promote democracy and human rights through its aid giving face well‐known problems. For instance, while the EU spends the largest amount of aid in its neighborhood—where positive conditionality has been the key approach (Del Sarto and Schumacher 2011)—qualitative accounts have argued that this strategy has not led to many noticeable improvements (Holden 2005). These accounts attribute the purported lack of progress to two main factors: First, authoritarian resilience to reforms stands in the way of EU attempts to promote political liberalization (Teti, Thompson, and Noble 2013). Highly restrictive atmospheres that work against change and other adverse domestic conditions may thus represent inopportune conditions for EU funds (Çelenk 2009). Second, policy ambiguity and inconsistency complicate efforts to promote clear objectives. In particular, the EU does not clearly define democracy or human rights, due to the contested nature of these concepts (Del Sarto and Schumacher 2011). The long chain of command subverts the EU's goals as the numerous agents involved in the policy process carry differing interpretations of these objectives (Bicchi 2010). Moreover, the EU's prioritization of its security interests, which often contradict its aid priorities, contributes to policy ambiguity as well (Youngs 2010). Thus, while a large literature traces the success of liberalization in Eastern Europe to the powerful forces of European integration (Kelley 2004; Levitz and Pop‐Eleches 2010), the question remains whether the EU can foster liberalization even when the prize at stake falls well short of the substantial benefits of full EU membership.

Our research design allows us to adjudicate this important debate using a novel approach. Moreover, our concentration on multilateral EU aid also permits us to follow a particular aid‐giving procedure in detail, facilitating a nuanced interpretation of the role of international institutions in promoting human rights through the incentives provided by foreign aid. In particular, we show that the rotation principle of the Council produces short‐term increases in aid commitments directed toward countries favored by the state holding the presidency. Importantly, the Commission is then given great discretion in disbursing these commitments. Our results are consistent with a mechanism under which the Commission is able to leverage these short‐term increases to aid allocations to promote reforms in recipient countries. To flesh out this process, we consider the role of each of these bodies in turn.

The Council and Former Colonies In principle, policies such as foreign aid are delegated to the EU because, as a multilateral institution, the EU is perceived as providing aid in a more objective, nonpartisan manner (Milner and Tingley 2013). However, the EU is composed of individual member states whose policy preferences may differ from those of the EU as a whole. As a result, we expect individual EU member states to attempt to move policies closer toward their own preferred positions. The Council is often best positioned to do so. It is composed of high‐level political representatives of the member governments, and because the ministers represent individual member states, they have national, in addition to European, interests. One particularly salient national interest with respect to foreign policy is their preference to assist countries they favor. In particular, we expect that the ministers seek to reward their respective nations' former colonies, since former colonizers tend to have political, cultural, and economic ties to these states that stem from shared history, duty, economic benefits for their companies, political legitimacy, symbolic representations of international power, and familial connections (Arts and Dickson 2004). One primary means through which the states may assist their former colonies is through EU aid. In fact, although the literature investigating donor motivations for providing aid disagrees on many points, a consistent, large, and uncontroversial finding is that former colonizers give disproportionate amounts of aid to their former colonies (Alesina and Dollar 2000; Holland 2002). This may be especially true within the EU, since EU aid represents a considerable supplement to members' own bilateral aid giving and allows them to share the costs associated with aid provision. This favoritism can be readily observed in the Council ministers' actions. France, for example, shaped the construction of a special aid mechanism, the Yaoundé convention, which was composed primarily of its former colonies, to allocate them more aid. Similarly, the UK joined the European Community under the condition that EU aid be expanded dramatically to benefit former British dominions. When Spain and Portugal then entered in 1986, they demanded that Latin American and Mediterranean states, particularly their former colonies, receive increased development assistance (Arts and Dickson 2004). Indeed, opinions within the EU are still divided over which states to focus assistance on, with former colonial powers seeking to direct more aid to their former colonies, and other states such as Germany, the Netherlands, Finland, and Sweden advocating broader aid policies (Arts and Dickson 2004).

The Presidency and the Foreign Aid Budget The power to affect aid policy is not distributed equally among the ministers, however. Instead, the Council aims to avoid some of the difficulties associated with decision making through the institution of the Council presidency, which rotates every 6 months among the governments of the member states. When a particular state holds the presidency, the associated Council ministers adopt a leadership role. Given the ministers' favoritism toward their respective nations' former colonies, we expect that when former colonizers hold the presidency, they use this leadership position to support these states. Further, because the presidency's role is circumscribed in many respects, one particularly salient remaining means through which the presidency can provide such support is through increased foreign aid.7 Indeed, while the presidency exerts influence over the foreign aid budget, its influence in other areas is often severely limited for several reasons. Outside of the budget preparation, the presidency primarily performs administrative tasks that do not confer much influence, leaving little time for other activities. The presidency is also responsible for agenda items that are inherited or that arise for external reasons, curbing its discretion. Additionally, the short period of time that a particular country holds the presidency limits that country's potential impact. Further, as described previously, the presidency possesses few formal powers and operates under a “culture of consensus,” such that it must obtain agreement from the other Council ministers on most issues (Westlake 1995).8 Due to these constraints on its influence, Council presidencies thus turn to aid allocation to favor their former colonies, as exemplified by many anecdotal accounts. For example, the Belgian presidency in the second half of 2001 placed former colonies in Africa high on the European Community's aid agenda and channeled additional aid to one of its own former territories, the Democratic Republic of the Congo.9 Similarly, the 2002 Spanish presidency made a point of aiding its former colonies in Latin America (Bengtsson 2003; Manners 2003).10 Below, we demonstrate that this association holds systematically both across colonizers and over time. How does the presidency direct this aid to its former colonies? Though the presidency is tasked with many jobs, of particular interest here is the presidency's influence over the adoption of the EU budget. The Council has the last word on compulsory expenditures, which include aid, and the presidency can control budgetary meetings (Tallberg 2003b), shape the agenda (Tallberg 2003a), mediate disputes that arise, and use its budgetary veto power to force concessions (Manners 2003). For instance, during the Finnish presidency in the second half of 1999, a budgetary dispute arose regarding the appropriate amount of funds to allocate to Kosovo. Finland mediated the dispute and negotiated a compromise that it favored. Additionally, the British presidency of 1998 presided over and mediated a number of budgetary reforms, including shifting regional boundaries and regional aid. It also lobbied to maintain regional aid to its preferred locations (Manners 2003). Increasing the funds allocated to specific countries and programs therefore allows the presidency to boost the aid budgeted to its former colonies. We thus expect that holding the Council presidency causes former colonizers to budget extra aid for their former colonies.