Clara Portela_volume3_issue2 Review of European and Russian Affairs vol. 3 issue 2/2007 © RERA 2007 all rights reserved 38 ………………………………………………………………… Aid Suspensions as Coercive Tools? The European Union’s Experience in the African- Caribbean-Pacific (ACP) Context Clara Portela 1 Abstract Since the signing of the Cotonou Agreement in 2000, the European Union (EU) has suspended development aid towards a number of African Caribbean and Pacific (ACP) countries in response to breaches of Human Rights and democratic principles by activating the so-called Human Rights clause (article 96). The present article analyses the use by the EU of aid suspensions as political tools and their efficacy in achieving the desired policy goals, in an attempt to identify and explain the determinants leading to the success of these measures. The investigation finds that the use of development aid suspensions is frequently effective. Classical sanctions theory appears to account largely for their success, given that most targets display a significant degree of dependence on the EU as a donor or a trading partner. However, and without refuting the explanatory power of that approach, a closer look at this practice unveils a number of factors that contribute to facilitate success. One of them is the selective use of the tool: suspensions are applied predominantly in cases of interruptions of the democratic process, while they are rarely used in situations of violent conflict. The specificities of the consultations mechanism, and especially the attitude of ACP neighbouring countries- often openly supportive-, largely determine the final outcome. Keywords: sanctions, EU development aid, ACP aid suspensions, ACP-EU relations 1 Clara Portela is a PhD Researcher at the European University Institute. The author wishes to thank Dr Anna Schmidt, Jean-Monet Fellow at the Robert Schuman Centre for Advanced Studies (Florence), for her useful comments. Review of European and Russian Affairs vol. 3 issue 2/2007 © RERA 2007 all rights reserved 39 1. Introduction Over the past decade, the European Union (EU) has suspended development co- operation with a number of African-Caribbean–Pacific (ACP) countries. These suspensions have taken place within a specific legal-institutional regime: the Partnership Agreement between the EU and ACP countries, the Cotonou Convention of 2000, which succeeded a series of Lomé Conventions embodying a privileged relationship between the EU and the former colonies of some of its Member States. Article 96 of the Cotonou Convention provides for a consultation procedure which can be invoked in cases of serious breaches of Human Rights and democratic principles, and which empowers the Council of the EU to suspend development aid. The use of this tool has provoked some controversy: the EU has sometimes been accused of inconsistency in invoking article 96, and in particular in suspending aid. However, little is known about the efficacy of these measures: are development aid cut-offs effective in compelling ACP leaders to comply with EU demands? EU officials have pointed to the limited knowledge the EU has about the use of sanctions (Niño-Pérez 2004). The Commissioner for External Relations and Neighbourhood Policy recognises that the use of sanctions “needs to be optimised” (Ferrero-Waldner 2006). This paper sets itself the objective of assessing the efficacy of development aid cut-offs in the framework of the Cotonou Agreement as a coercive instrument of EU foreign policy. It does so by reviewing the cases in which it has been wielded since the entry into force of the Convention. The standard of success used here is that set by the EU itself: An episode is deemed “successful” here when the EU considered progress sufficiently satisfactory to resume aid. Departing from general sanctions theory, the paper attempts to isolate and evaluate the factors accounting for their success and failure before drawing some conclusions for EU policy in this domain. 2. Sanctions and their Evaluation: The Need for a Fresh Approach Since the end of the Cold War, the instrument of sanctions has been subject to a profound transformation. The classical general trade embargo has been replaced by so-called targeted or “smart” sanctions. They are designed to affect directly the leadership responsible for the objectionable behaviour, avoiding the adverse humanitarian consequences of comprehensive economic sanctions which provoked their unpopularity. The concept of targeted sanctions encompasses a diverse range of measures: travel bans, financial sanctions such as the freezing of assets or investment bans, arms embargoes, flight bans and embargoes on specific commodities (Hufbauer and Oegg 2000). The EU was at the forefront of the promotion of targeted sanctions at UN level. It has officially embraced this notion (Council 2004), and it has consistently applied only smart sanctions in its autonomous sanctions practice in the context of its Common Foreign and Security Policy (CFSP) since the early nineties. However, in parallel to its CFSP sanctions, the EU continues to impose some measures whose characteristics approximate those of economic sanctions. The suspensions of development aid as part of the “appropriate measures” foreseen in Article 96 of the Cotonou Convention are measures of economic nature taken on political grounds and with a coercive intention. Yet, the EU does not label these measures “sanctions”, and keeps this practice legally separate from CFSP measures. Review of European and Russian Affairs vol. 3 issue 2/2007 © RERA 2007 all rights reserved 40 The EU designs its suspensions in such a way as to spare the population of the country concerned from suffering deprivation: they only affect budget support provided directly to governments. Nevertheless, the suspension of aid under article 96 remains an influence attempt in which a benefit (development aid) is withdrawn that would otherwise be available, and whose provision is made dependent on the fulfilment of a series of conditions defined by the sender (the EU) and the target on the basis of a mutual agreement. The objective of the present paper is to ascertain their coercive power, i.e. whether they managed to compel the desired behavioural change in the targeted leadership, and what conditions facilitate efficacy. Efficacy is defined here as the capacity of the measures to produce the intended results. The standard of success is adopted from the EU itself: it is the imposing organisation that measures compliance internally and decides what degree of compliance suffices to resume aid. At first sight, an exploration of the efficacy of development aid cut-offs might appear to bear little connection to the current sanctions debate. However, the findings can reveal precisely how effective economic instruments remain in the era of targeted sanctions. Determining the efficacy of aid cut-offs will enable us to ascertain how economic pressure fares in the overall sanctions toolbox. Are targeted sanctions delivering better results in terms of target compliance than the traditional economic tool? The present paper looks into cases of EU suspension of aid to ACP countries from 2000 to the present. The signing of the Cotonou Agreement in 2000 marks the starting year of the series of cases examined. Political conditionality and development aid cut- offs predated the signing of the Cotonou Agreement. The suspension clause was first introduced with article 336a into the Lomé IV Convention as a provision to enhance the protection of Human Rights and democracy. Previous to that date, development aid to ACP countries had been suspended on several occasions; however, the unavailability of legal bases at the time makes the process leading to the decisions on these instances hard to trace (Hazelzet 2005: 4). 3. Development Aid Suspension: What Kind of (EU) Sanction? Which place do development cut-offs occupy in the EU sanctions tool-box? From an EU institutional point of view, the suspension of development co-operation is not properly a “sanction”. In EU terminology, the term sanction is reserved for those measures decided under the Common Foreign and Security Policy (CFSP), which typically encompass arms embargoes, visa bans and the freezing of financial assets. In the present paper, I draw on the following definition of the notion: “Sanctions” can be defined as the interruption of normal relations or the withdrawal of a benefit by a state (or group of states) in response to an objectionable action by another state or entity. The “sender” is the entity imposing the sanction and the “target” is the entity at the receiving end. The aim of the sanction is to compel the target to put an end to the undesirable behaviour. They are meant to withhold a benefit from the target until it complies with a demand by the sender that it would not otherwise have agreed to. Therefore, the suspension of development aid fits into the academic definition of sanctions, even though in EU parlance it is not referred to as such. Review of European and Russian Affairs vol. 3 issue 2/2007 © RERA 2007 all rights reserved 41 The procedure leading to the suspension of development aid is highly institutionalised. The respect for Human Rights, democratic principles and the rule of law constitute essential elements of the ACP Partnership Agreement. It foresees a consultation procedure under its Article 96 in order to address violations of these principles by any of its parties. Once a serious breach has been observed, the Council of the EU invites the ACP country concerned to participate in consultations under Article 96. The decision is taken by the Council on a Commission’s proposal. Consultations have to start no later than 15 days after the invitation is issued, and shall last for no longer than 60 days (120 days after the 2005 review of the agreement). The objective of the consultations is to agree on measures to rectify the situation in the country in question. The EU and the ACP country shall adopt a list of commitments with a timetable to be fulfilled by the responsible government. When consultations are closed, the EU evaluates the progress made by the government in fulfilling the commitments entered into and might then decide on “appropriate measures”. These might entail, among others, the suspension of development cooperation. Appropriate measures can also consist of positive measures, such as the release of new funds. Suspension can be total, for which unanimity in the Council is required, or partial, for which a qualified majority suffices. Subsequently, the EU conducts regular monitoring, including the dispatch of missions to the country to assess progress in implementation. Not even in cases where the suspension is total is the entire flow of aid interrupted. The suspension only affects one strand of development aid, namely the budget support administered directly by the state, mostly affecting infrastructure projects. Moreover, the suspension is only meant for projects that are not yet in place, leaving ongoing programmes unaffected. The suspension often takes the form of the non-notification of the upcoming European Development Fund (EDF) programme. Humanitarian aid is expressly exempted, and development aid directly in support of the population and channelled through NGOs is often left in place. This might include health and education programmes, food security, programmes in support of civil society, human rights and democracy, and rural development projects. Regional projects involving several countries are not interrupted as a result of the suspension of development aid against one of the participants. Finally, bilateral co-operation programmes remain unaffected by EU development aid suspensions, which only concern Community funds. Member States are free to continue bilateral aid despite EU suspension, and some of them often do. A tactic often used by the EU in order to signalise discontent with the policies of one of the beneficiaries of its development aid is the redirection of aid towards aims different to that it had been originally allocated to. The EU has taken this approach also in the context of article 96 suspensions – this has been the case e.g. with Zimbabwe, where funds foreseen for budget support have been redirected towards humanitarian aid and projects in direct support of the population. 4. Article 96 Aid Suspensions: Actual Usage An overview of the cases of article 96 consultations reveals that it is mostly imposed in cases of coups d’état (Bradley 2005). Most cases of suspension were officially justified as reactions to both Human Rights violations and to the interruption of democratic processes. However, there is a clear predominance of the democracy element: article 96 was invoked whenever the democratic process was interrupted – Review of European and Russian Affairs vol. 3 issue 2/2007 © RERA 2007 all rights reserved 42 such as in cases of flawed elections – even if no Human Rights breaches took place. By contrast, no cases can be found where Human Rights violations constituted the only motive for holding consultations. The predominance of interruptions of the democratic process might be ascribed to an implicit assumption that Human Rights are better protected under democratic regimes. Also, the proposal to hold consultations is likely to prosper following a clear-cut violation such as a coup d’état or an illegal constitutional reform, while in the case of Human Rights, the nature and extent of the breaches are often more difficult to document. This type of violations might allow some members of the Council to withhold their consent to the call for consultations, and the government at fault might contest the occurrence of the breaches be more easily. The fact that the suspension of development co-operation to ACP countries takes place in the specific institutional framework of the “appropriate measures” that follow consultations invoked under article 96 bears important consequences. Firstly, consultations provide a framework in which the country in question has an opportunity to present its plan to remedy the breach by entering into specific commitments and drafting a timetable jointly with the EU. The suspension of development cooperation is a measure of last resort: Only in cases where the consultations fail to produce satisfactory results does the EU consider it. Finally, the fact that the government in question maintains effective control over most of the territory of the country seems to be a pre-condition for the opening of consultations. Hazelzet notes that the EU refrains from invoking article 96 in situations of violent conflict, such as “during a civil war, or when a country was on the verge of a peace agreement” (Hazelzet 2005:12). In addition to the volatile character of conflict situations, the involvement of the UN or other organisations in many of the countries in conflict makes article 96 consultations redundant (García- Pérez 2007:4). In most cases, the EU has become involved in these countries by participating in broader political processes, such as in the Democratic Republic of Congo. In these situations, the EU has typically wielded CFSP tools like arms embargoes and the interruption of military co-operation, such as Sudan in 1996 and Ethiopia/Eritrea in 1999. In sum, the EU invokes article 96 whenever it considers that it stands a reasonable chance of influencing the leadership responsible for the breaches. 5. Sanctions Operation: What Does Theory Say? The operation of sanctions was originally formulated by Galtung in what he labelled the “naïve” theory of sanctions. According to the naïve theory, the economic disruption caused by sanctions is expected to translate into political pressure that will eventually compel the leadership to change its policies, or lead to its overthrow (Galtung 1967). Sanctions are therefore expected to work in a similar way to military coercion. Their aim is the “political disintegration of the enemy so that he gives up the pursuit of his goals. The method used is value-deprivation” (Galtung 1967:386). The theory foresees a roughly proportionate relation between value-deprivation and political disintegration: “The idea is that there is a limit to how much value deprivation the system can stand, and that once this limit is reached (resulting in a split in leadership or between leadership and people), then political disintegration will proceed very rapidly and lead to surrender or willingness to negotiate” (Galtung Review of European and Russian Affairs vol. 3 issue 2/2007 © RERA 2007 all rights reserved 43 1967:388). It presupposes that a “societal transmission belt” produces political pressure eventually leading to the ousting of the leaders, thereby turning economic damage into political pressure. Subsequent scholarship has devised a broader formulation capable of accommodating the operation of non-economic sanctions, which are absent in the naïve theory’s account. Pressure exerted by sanctions compels decision makers “to change their calculation of costs and benefits” (Lindsay 1985:155; Cortright and Lopez 2000:22). The aim is to exert a sufficiently strong pressure on the offending state so that continuing to suffer the measures applied against it represents a higher cost than putting an end to its wrongful behaviour. In sum, two different ways of operation of sanctions can be discerned in sanctions theory: One foresees that the targeted leadership will comply with the demands of the sender as a result of a calculation in which the disutility caused to it outweighs the benefits of persisting in the objectionable policies. This disutility can either take the form of personal damage inflicted on policy-makers (as a consequence of blacklisting for a visa ban or the freezing of financial assets), or that of political pressure transmitted through the “societal conveyor belt” as a result of other measures. Sanctions such as trade embargos can be geared at any of these two possible outcomes: the leadership is expected to be compelled to yield by the prospect of destabilisation, or else be overthrown. In many cases, it cannot be discerned whether the sender has only one of these outcomes in mind when levying the sanctions. Due to the decline in use of multilateral comprehensive trade embargos, the scenario of “success by overthrow” has become less frequent. Targeted measures generally cause less abrupt disruption to the target society, so that they are more likely to achieve “success by compliance”. The case of development aid suspension under Article 96 of the Cotonou Agreement fits better with the mechanism of success by compliance than with the overthrow- scenario. EU aid sanctions operate differently from the classical trade embargo. The EU does not intend to extort the population – on the contrary, it makes a dedicated effort to spare it from deprivation. Humanitarian aid is maintained, and often even increased, notably through the redirection of the funds withheld to projects in support of the population. The disutility that the EU expects to inflict on the target arises from a series of interlinked considerations: the absence of new infrastructure projects creates a negative business environment in the country, thereby failing to attract foreign investment. Moreover, given that donors co-ordinate their policies, the EU’s withdrawal is often accompanied by that of the international financial institutions and other EU Member States. In the absence of donor support and private foreign investment, economic performance worsens. As a result, the government becomes increasingly unpopular domestic circles. While a population which is not suffering extreme deprivation is unlikely to conduct a revolt of the type foreseen by the naïve theory, the multiplying effects of aid suspension provides the concerned government with an incentive to avert it. 6. How are Sanctions Evaluated? A Brief Outline The landmark work evaluating sanctions efficacy, Economic Sanctions Reconsidered (Hufbauer et al. 1985; HSE in the following) inspired subsequent scholarship, while it Review of European and Russian Affairs vol. 3 issue 2/2007 © RERA 2007 all rights reserved 44 also faced criticism on various accounts (Hovi et al. 2005). In order to ascertain the determinants for the success of EU development aid suspensions, the present article has adapted HSE’s variables. HSE analyse two types of variables: economic and political. Most of the economic variables are designed to ascertain the economic cost imposed on the target country and on the sender. In the case of ACP countries, the importance of the EU to the target country in political and economic terms is closely interlinked. Thus, HSE’s distinction between the headings “Political” and “Economic Variables” becomes blurred. The scope of the suspension of aid, or type of sanction (i) and the importance of the EU as a trading partner (ii) and as a donor (iii) will be taken into account, as they indicate not only the level of disutility created to the target, but also its vulnerability. HSE’s political variables encompass a factor which is now obsolete: the presence of international assistance as an effort to undermine the sanctions regime – a phenomenon that belongs to the Cold War confrontation. The motive triggering the suspension (iv) and the numbers of years the sanctions were in force (v) are also included. The neighbours’ attitude (vi), which has been highlighted as a factor of central importance in the literature, is also considered. The methodology outlined by HSE contemplates two separate investigations: it first ascertains the “policy result” and then the “sanctions contribution” to the desired objective. The present investigation simplifies the analysis by creating a single table, in an attempt to identify which conditions are conducive to a successful conclusion of consultations. The selection of cases follows three criteria: the table features partial or total suspensions of aid resulting from consultations under article 96 of the Cotonou Agreement in which no mandatory sanctions were imposed simultaneously by the UNSC. The case of Liberia is therefore excluded. The following list features cases of consultations in which a suspension of aid was decided, leaving out all those instances in which progress was so smooth that the EU did not deem it necessary to use coercion in the first place. This is e.g. the case of consultations with Guinea-Bissau or the recent consultations with Mauritania. Review of European and Russian Affairs vol. 3 issue 2/2007 © RERA 2007 all rights reserved 45 7. Cases of Article 96 Aid Suspensions Measure years EU as partner EU as donor reason Neighbour’s attitude Resumption of aid Central African Republic Partial suspension 2003-2005 (2) Very high Very high Coup d’état none + Ivory Coast Partial suspension 2000-2002 (2) Very high High Flawed elections None + Fiji Partial suspension 2001-2003 (2) high Low Coup d’état None + Haiti (Almost) total suspension 2001 (1) low high Flawed elections OAS/ Consultative Group + Guinea- Conakry Partial suspension 2003- ongoing (4) high high Coup d’état none +/- (partial resumption) Togo (Almost) total suspension (1993-94); 1998 -2006 (13) - - Illegal constitutional change ACP supportive + Zimbabwe Partial suspension 2002 – ongoing (5) medium high Flawed elections Neighbours opposed - 7.1 Central African Republic The Central African Republic has not been favoured by the donor community in spite of its heavy dependence on foreign aid. The EU has a prominent role among donors, especially since the IMF cut budgetary support in 2001 (Laakso et al. 2006). Article 96 was invoked following a coup d’état by General Bozizé in 2003, which overthrew President Patassé. Consultations, starting in June 2003, were not limited to the discussion of democracy and Human Rights, but addressed also economic governance and corruption, and even respect for the Kimberly process concerning the export of diamonds. A monitoring mission in October 2003 concluded that, while dialogue on reconciliation had been successful, progress made regarding the freedom of the opposition, the timetable for elections and the reform of the military was unsatisfactory. Therefore, the Council decided a partial suspension of development co-operation encompassing some road projects, macro-economic support, and a progressive reduction of the 9th EDF funds by 20% per year. Cooperation was resumed following the flawless presidential and legislative elections in 2005. 7.2 Ivory Coast A comparison of the case of Ivory Coast with that of the Central African Republic shows that different former colonies often receive different treatment by the former metropolis. Ivory Coast has strong trade ties with France, partly maintained by a significant French business community. Following the flawed elections in 2000, from which Guei emerged as a successful candidate over his opponent Gbagbo, the EU called for consultations pointing to unfulfilled commitments that had been agreed in previous rounds of consultations under article 366a of the Lomé Convention. However, Guei was soon thrown away from office by a popular uprising and replaced by Gbagbo. Following the closure of consultations in July 2001, the EU decided to resume aid through a gradual and conditional approach, with an emphasis on funding Review of European and Russian Affairs vol. 3 issue 2/2007 © RERA 2007 all rights reserved 46 projects in support of the restoration of democracy, rule of law and good governance. Co-operation was fully resumed in January 2002 in view of satisfactory progress by the Ivorian government. However, the outbreak of violence compelled the EU not to sign the indicative programme agreed under the 9th EDF, so that most of the substantial funds it had originally committed were never disbursed. Only a few projects on rural development and good governance were financed. Nevertheless, the Commission expanded the scope of the humanitarian aid to encompass elections preparations and DDR programmes. Consultations were again proposed by the Commission in 2004 in view of the Human Rights breaches that followed the intensification of fighting in the north of the country that started in 2002. However, the Council rejected the proposal arguing that the threat of civil war would render Article 96 consultations ineffective. This decision was justified on the grounds that “the problem lay not in the nature of the reconciliation government per se but in the fact that this…not exist in any tangible sense as an interlocutor for the Union” (Youngs 2006:346). What followed was a period in which French bilateral activity gained pre-eminence, while EU assistance remained de facto suspended and other Member States gradually winded down their bilateral programmes. 7.3 Fiji Fiji is one of the ACP countries which is less dependent on European aid; indeed, it is barely eligible for aid. Consultations under article 96 were initiated in October 2000, following a coup d’état in May of that year. During the consultations, power shifted from the original putschistes to the military, and eventually to a nominated President. The EU made the notification of 9th EDF allocation and the financing of new programmes under the previous EDF (6th to 8th) conditional upon the holding of elections and the appointment of a legitimate government. Yet, regional projects, humanitarian aid and trade related preferences were left in place. Once the conditions were fulfilled, the EU decided on a gradual resumption of cooperation in November 2001, consisting of projects in support of the rural population and on Human Rights and democracy. However, an irregularity observed in the allocation of seats in the parliament - a number of elected MPs of the Labour Party had been excluded - left the EU unsatisfied. Full co-operation was only resumed in November 2003 after the nomination of Labour cabinet ministers. A new round of consultations was held in early 2007 following the military take-over that took place in Fiji in December 2007. The leadership undertook a number of commitments, including the holding of elections no later than the March 2009. 7.4 Haiti As the poorest country in the Western hemisphere, Haiti is heavily dependent on foreign aid. However, the importance of the EU as a trading partner is low due to the overwhelming commercial weight of the US. Article 96 was invoked as a consequence of the electoral fraud observed in the May 2000 legislative elections. As a result of the consultations held in September 2000, the Haitian government promised to arrange new elections for senators. Review of European and Russian Affairs vol. 3 issue 2/2007 © RERA 2007 all rights reserved 47 Following the Haitian failure to abide by its commitment, the EU decided the suspension of budget aid and food security and of the second instalment of the 8th EDF, redirection of the remaining funds to projects in direct support of the population, civil society and private sector and the non-notification of the 9th EDF. Only humanitarian aid was left in place. In the following years, the EU modified its measures several times as a result of its participation in the Consultative Group of Haiti, whose members encouraged the EU to lift the suspension. Development aid was resumed in October 2005 as a result of the Haitian commitment to hold elections, which eventually took place in February 2006. 7.5 Guinea-Conakry The democratic standards in the Republic of Guinea had been deteriorating for some time before the flawed presidential elections of December 2003 gave rise to consultations with the EU. These opened in July 2004 and concentrated on improvements to be made in the electoral process, such as freedom of the media and non-harassment of the opposition. In April 2005, the EU decided to redirect the unexpended balances of previous EDF towards decentralisation, liberalisation of the media and economic governance, while 9th EDF funds continued for programmes in direct support of the population. Slight progress by the Guinean authorities, for example with respect to the operation of independent radio senders has been reciprocated with a partial lifting of measures. However, as the current administration and the opposition have not yet agreed a programme for electoral reform, cooperation has not resumed fully. 7.6 Togo The EU suspended part of its aid to Togo already in 1993 as a consequence of an illegal change to the Constitution made by President Eyadema. Cooperation was resumed after the Constitution was restored in 1994, but it was interrupted again in 1998 due to flawed presidential elections. The EU supported financially the establishment of an independent national electoral commission for the 2003 election; however, Eyadema dissolved it and the elections were again reported to be flawed. Togo did not receive any EU funds of the 8th and 9th EDF – it was only a recipient of humanitarian aid. Following consultations in 2004, co-operation is now in place again. The case of Togo is unusual in that the consultations that took place in 2004 were called at the request of its government. Consultations started in March 2004 and were closed in November, with an agreement by the Togolese authorities to implement a list of commitments, mainly related to the re-establishment of democratic rule. The resumption of full co-operation was made conditional on the holding of free elections. A number of problems surfaced at the implementation phase, so that the monitoring mission in June reported that despite the good will shown by the government, it was not possible to make a conclusive evaluation. However, the EU decided to release funds to support the preparation of elections, human rights programmes and justice reforms. Following the signing of the Global Political Agreement by government and the opposition and the announcement of elections in the following year, the EU fully resumed aid in 2006. 7.7 Zimbabwe Review of European and Russian Affairs vol. 3 issue 2/2007 © RERA 2007 all rights reserved 48 The case of Zimbabwe is exceptional in a number of respects: It is the only example in which a suspension of aid under article 96 has been complemented by CFSP sanctions. Previous cases of development aid suspension combined with visa bans or arms embargoes (Nigeria 1993 or Sudan 1990) were handled outside the EU-ACP framework, given the unavailability of the consultations procedure before 1995. Zimbabwe has also been described as a deviant case in studies of governance, human rights and democracy in ACP countries. It has been singled out as the only case in which, according to the relevant indicators, freedom of participation had improved in the year previous to consultations under article 96 (Laakso et al. 2006). Consultations were called following the refusal by the Zimbabwean authorities to allow the entry into the country of EU electoral observers on the eve of the 2002 elections, and their harassment of the opposition. A central déclencheur of the crisis was President Mugabe’s policy of illegal occupation of large farms and eviction of their white owners. The EU moved very quickly from the political dialogue under article 8 of the Cotonou Agreement to article 96 consultations, and these were apparently exceptionally short (interview: 2007). The suspension of co-operation that ensued left not only humanitarian aid in place, but also health and food security programmes. Equally, programmes under the European Initiative for Democracy and Human Rights (EIDHR) continued – over the past few years, more than seventeen projects have been conducted. Immediately after the suspension, the EU imposed CFSP sanctions: an arms embargo along with a visa ban and a freezing of assets against one of the longest blacklists of the EU. This order was strictly respected: the EU only proceeded to impose CFSP sanctions once the consultations had been closed. At the time of writing, no progress has been made. Zimbabwe’s neighbours continue to offer their political support to Mugabe. On the contrary, the deep economic crisis has compelled the EU to increase the amount of humanitarian aid provided. The presence of CFSP sanctions does not seem to have made any noticeable impact on the situation (interviews 2007). If anything, the imposition of CFSP measures might have a negative effect on the behaviour of the leadership: Zimbabwe is the only case of consultations under article 96 Cotonou where no progress has been observed over a period of five years. 8. When Do Development Aid Suspensions ‘Work’? Five out of the seven cases display an unequivocally positive result. The large number of successes hardly comes as a surprise, given that consultations are called only in situations where the EU believes to stand a reasonable chance of influencing the leadership in breach. Article 96 is not invoked in every case of violation of Human Rights and democratic principles. This selective approach partly accounts for the high level of success. Nevertheless, the positive record stands in sharp contrast with the overall performance of CFSP sanctions, which features few successes. Yet, the positive record of Article 96 has to be qualified by a number of considerations. To start with, the eventually positive outcomes are often preceded by tortuous paths to resumption. The uncooperative Haitian authorities, as opposed to the forthcoming Central African Republic, are a case in point. Only in Guinea-Conakry, where results are still considered suboptimal despite some progresses, and Zimbabwe aid has not Review of European and Russian Affairs vol. 3 issue 2/2007 © RERA 2007 all rights reserved 49 totally resumed. A number of failed consultations undertaken under Lomé IV reveal that the co-operation of countries at breach cannot be taken for granted: Ivory Coast and Togo had undergone consultations under Lomé with poor results. Thirdly, the degree of progress towards the EU demands to be achieved by the target in order to allow for a resumption of aid is decided by the EU itself, rather than by any independent evaluator. In certain cases such as Guinea-Conakry, it is doubtful that any external assessment would have considered that the meagre progress made by the leadership warranted the re-establishment of co-operation. Thirdly, success has to be assessed against limited - albeit not modest - goals: the EU has largely achieved its goal of promoting a return to democratic rule. However, the fact that conditions were met to an extent that the EU considered sufficient to fully resume co-operation could not prevent a recurrence of violence and grave human rights breaches in Ivory Coast. Indeed, none of the countries featured in the list have overcome the structural fragility of its institutions. Countries like Ivory Coast, Guinea-Bissau and Fiji present a pattern of recurrence of democratic crisis. Consultations are highly useful instruments, but they do not go beyond solving the crises at hand. Moreover, EU officials resent the fact that leaderships that proved very co-operative during the consultations were far more reluctant to implement reforms after these were terminated (interview 2007). Lastly, the fact that the development aid suspensions under CFSP proved unsuccessful in compelling a behavioural change in the leaderships at fault – such as Nigeria in 1993, Sudan in 1994, or Zimbabwe in 2002 – suggests that it is the intensive exchange between the parties that takes place during the procedure of consultations what provides for positive outcomes. In conclusion, the consultations constitute a highly useful tool for the reestablishment of democratic rule in emergency situations, but remain ill-suited to provide for lasting solutions to deep-rooted instability in weak states. The most often voiced criticism against the EU decisions to suspend development aid is its alleged lack of consistency. However, what becomes apparent in terms of the motives that compel the EU to invoke article 96 is the predominance of interruptions of the democratic process. Human Rights breaches are also omnipresent (although not explicitly displayed as such by the table). However, these are invariably linked to democratic flaws: a deterioration of the Human Rights situation alone has never triggered article 96 consultations. Also, the Human Rights breaches that the EU reacts to through this mechanism are civil and political rights closely linked to the democratic process: freedom of speech, freedom of association and of peaceful demonstration. Some of them are defined in terms of facilitating the holding of free and fair elections, such as the freedom of the press and the non-harassment of the opposition. Interestingly, in the context of Article 96 consultations ACP leaderships are often encouraged to subscribe to commitments related to good governance such as anti-corruption measures, although they are formally protected by a different provision of the Cotonou Agreement, article 97. This article has only been invoked once, in the case of Liberia, where consultations took place in parallel to article 96 consultations. In terms of efficacy, the most significant finding is that suspension works best as a coercive tool in cases where the EU is either an important trading partner or an important donor (or both), and crucially, whenever neighbouring countries are supportive of the EU. This is an intuitive finding, and it is possible to ascertain how Review of European and Russian Affairs vol. 3 issue 2/2007 © RERA 2007 all rights reserved 50 these factors are interrelated: neighbouring countries which do not are agree with EU suspension might be in a position to supply the country not only with political support, but also with essential supplies, which will reduce the importance of the EU as a donor and trading partner. From this point of view, the analysis tends to confirm the continued relevance of the determinant factors identified by traditional sanctions research: the EU has considerable economic weight as a donor or a trade partner in almost all the cases, and the co-operation of neighbours or like-minded countries proved highly relevant. In any case, the central determinant for the success of EU development aid cut-offs remains the attitude of the concerned leadership. In a number of cases, the coups d’état were conducted to unseat a leader whose own democratic record was poor with the unequivocal intention of restoring civilian democratic rule, therefore providing a favourable terrain for consultations. In these cases, the EU found itself in the role of “accompanying” some putschistes who were from the start ready to co-operate in a democratic transition. This is the case of the Central African Republic, and also of Mauritania and Guinea-Bissau, whose forthcoming attitude prevented the EU from cutting-off aid. Still, the fact that Guinea-Conakry presents the characteristics that would make it amenable to EU leverage has not prevented it from being still a relative failure. For its part, Zimbabwe appears to be a veritable deviant case. It is wealthier than most of the other examples, at least measured by pre-crisis indicators. Apart from the fact that the moves to consultations and subsequently to suspension seem to have taken place in a rush, the almost simultaneous imposition of CFSP sanctions signalise a more confrontational stance than in other cases. It should not be forgotten that the motive for suspension, the irregularities in the 2002 elections, generally garners less support from ACP neighbours than military coups. Support from its neighbours is central in Zimbabwe’s resistance, both from an economic and symbolic point of view. The situation appears highly atypical also from a sender-target point of view. The unsound economic policies of the Mugabe government have triggered a humanitarian crisis that compelled the EU to increase the amount of aid (interview 2007). Finally, the low proportion of non-African cases in the sample might obscure the role played by the geographic location of the target. Placed in the Caribbean, Haiti belongs to the sphere of influence of the US, which contrasts with the European pre-eminence in most of Sub-Saharan Africa. While the result of the consultations was eventually positive, the Haitian leadership proved less forthcoming than many of the other cases. The heavy involvement of the OAS in the resolution of the Haitian democratic crisis seems to have strongly influenced the behaviour of the EU – the decision to resume aid as been described a being motivated by the desire to appear as a “credible partner” in the Consultative Group on Haiti (Laakso 2006:37). In this case, the involvement of other interlocutors seems to have reduced the EU’s room of manoeuvre. 9. Conclusions The use by the EU of development aid suspensions as a political tool to address breaches of democratic rule in ACP countries is reasonably effective. The success of this instrument can be largely ascribed to the weight that the EU carries for the concerned countries as a donor and a trading partner, but also to the positive attitude of neighbours and fellow members of the ACP group. The institutionalised and transparent nature of the consultation process is another factor contributing to article Review of European and Russian Affairs vol. 3 issue 2/2007 © RERA 2007 all rights reserved 51 96’s satisfactory performance. However, this success should be qualified by the selective use of this tool by the EU, which has generally refrained from calling consultations in cases of violent conflict, or in cases where it simply does not expect to dispose of sufficient leverage. Review of European and Russian Affairs vol. 3 issue 2/2007 © RERA 2007 all rights reserved 52 Bibliography Bradley, A. (2005) An ACP Perspective and Overview of Article 96 Cases. European Centre for Development Policy Management (ECDPM): Discussion Paper No. 64D. ECDPM: Brussels Fierro, E. (2003) The EU’s Approach to Human Rights Conditionality in Practice. M. Nijhoff: The Hague/ London Galtung, J. (1967), “On the effects of international economic sanctions, with examples from the case of Rhodesia”, World Politics 19 García-Pérez, M. (2007) How to deal with Human Rights among Partners: Consultations under the Cotonou Agreement, in: Shared Responsibilities – Human Rights and Development in the 21st Century, Intersentia: Antwerp-Oxford Haass, RN. and O’Sullivan, MN. (eds.) (2000) Honey and Vinegar: Incentives, Sanctions and Foreign Policy. Brookings Institution Press: Washington D.C. Hazelzet, H. (2005) Suspension of Development Co-operation: An Instrument to Promote Human Rights and Democracy? European Centre for Development Policy Management (ECDPM): Discussion Paper No. 64B. ECDPM: Brussels Hovi, J., Huseby, R. and Sprinz, D. (2005) “When Do (imposed) Economic Sanctions Work?”, World Politics 57, July 2005, pp. 479-99 Hufbauer, GC. and Oegg, B. (2000) “Targeted Sanctions: A Policy Alternative?”, Paper submitted to the symposium “Sanctions reform? Evaluating the Economic Weapon in Asia and the World”, 23 February 2000, Washington D.C.: Institute for International Economics Hufbauer, GC., Schott JJ. and Elliot, KA. (1985) Economic Sanctions Reconsidered. History and current policy. Institute for International Economics: Washington, D.C. Laakso, L., Kivimäki, T. and Seppänen, M. (2006) Evaluation of the consultation processes under article 96 in the Cotonou Partnership Agreement. Comparative Overview. Conflict Transformation Service: Albertslund Lindsay, JM. (1986) Trade Sanctions as Policy Instruments: a Re-examination, International Studies Quarterly 30, pp.153-173 Mackie, J. and Zinke, J. (2005) When Agreement Breaks Down, What Next? The Cotonou Agreement’s Article 96 Consultation Procedure. European Centre for Development Policy Management (ECDPM) Discussion Paper No. 64A. ECDPM: Brussels Mbangu, L. (2005) Recent Cases of Article 96 Consultations, European Centre for Development Policy Management (ECDPM) Discussion Paper No. 64C. ECDPM: Brussels Niño-Pérez, J. (2004) “EU instruments for conflict prevention”, in V. Kronenberger and J. Wouters: The European Union and Conflict Prevention. Policy and Legal Aspects. TMC Asser Press: The Hague Portela, C. (2005) “Where and why does the EU impose sanctions?”, Politique Européenne no.17, autumme-hiver 2005/6 Portela, C. and Kreutz, J. (2004) “New EU principles on the use of sanctions: from symbolism to effectiveness?” European Security Review, no.23, June 2004 Youngs, R. (2006) “The EU and Conflict in West Africa”, European Foreign Affairs Review 11: 333- 352 Speeches/ Documents/ Interviews Review of European and Russian Affairs vol. 3 issue 2/2007 © RERA 2007 all rights reserved 53 Ferrero-Waldner, B. (2006) “The EU in the World”, European Policy Centre Breakfast Briefing, Brussels, 2 February 2006 Interviews with EU officials, diplomats from EU Member States and country experts (January/February 2007) Council of the European Union (2007) Opening of Consultations with the Republic of Fiji Islands under Article 96 of the Cotonou Agreement (Brussels, 18 April 2007) Conclusions of the European Union, Brussels, 19 April 2007, 8630/07 (Presse 83) Delegation of the European Commission for the Pacific (2007) “Outcome Statement on Consultations between the EU and Fiji held in Brussels from 18-19 April, 2007, Press Release Council of the European Union (2004) “Basic Principles on the Use of Restrictive Measures (sanctions)”, Council Doc. 10198/1/04 REV, Brussels, 7 June 2004 Council of the European Union (2003) “Guidelines on the Implementation and Evaluation of Sanctions”, Council Doc. 15579/03, Brussels, 3 December 2003