Burden-Sharing or Free-Riding? Explaining Variations in States' Acceptance of Unwanted Migration

Eiko R. Thielemann

London School of Economics, Department of Government & European Institute, London WC2A 2AE, E.Thielemann@lse.ac.uk

Forthcoming in: Thielemann, E. (2003) (ed.) 'Forced Migration and European Burden-Sharing, Journal of Refugee Studies (special issue), Vol.16, No. 3.

Abstract:

International burden-sharing, i.e. the question how costs of common initiatives or the provision of public goods should be shared between states raises two important questions. First, the question of motivation. How do we explain calls for burden-sharing beyond the state? Second, the question of patterns. How do we explain patterns of burden-sharing in the international arena? In order to address these questions, the first part of the paper proposes and analyses two burden-sharing approaches—one based on a 'cost-benefit'-, the other one based on a 'norm-based' logic—that offer partly competing and partly complementary hypotheses for answering these two questions. The second part, analyses EU attempts to share burdens in the area of forced migration in order to test these hypotheses empirically. The results suggest that although there is still little evidence for inter-state solidarity in the EU, norm-based approaches can nonetheless offer some powerful explanations of European burden-sharing in this area.

Keywords: burden-sharing; European Union; public goods; private goods; norms; forced migration; Kosovo

Paper prepared for presentation at the 8th European Union Studies Association (EUSA) Biennial International Conference, March 27-29, 2003, Nashville, Tennessee

1. Introduction[1]

International burden-sharing, i.e. the question how costs of common initiatives or the provision of public goods should be shared between states, has long been a concern for academics interested in the workings of international organisations such as NATO (Olsen and Zeckhauser 1966; Boyer 1989; Oneal 1990).[2] In the context of European integration, such concerns have long been limited to debates about the net-budgetary positions of the EU Member States (Wallace 1980; Shackleton 1990; Tsoukalis 1993; Begg and Grimwade 1998). Although these questions have remained highly relevant, more recently there has been an increasing interest in questions that move beyond a concern with purely budgetary issues and take account of less quantifiable costs (e.g. social 'costs' of accepting refugees) and benefits (e.g. market access) of integration (Noll 1997; Harvey 1998; Chalmers 2000).

Despite frequent references to burden-sharing by the Commission and the Member States, as with the recent moves to create a European Refugee Fund (ERF), the term has not yet been properly conceptualised in the EU context. In particular two crucial questions have not yet been sufficiently addressed: First, there is the question of motivation. How do we explain calls for burden-sharing beyond the state. In what ways, and to what extent, are the motivations for burden-sharing beyond the state different to those one can witness within a nation state context? Are burden-sharing arrangements the result of instrumental co-operation to overcome collective action problems? Or can we observe norm-guided actions based on emerging notions of cross-border solidarity? Second, the question of patterns. How do we explain patterns of burden-sharing in the international arena? What are the reasons for the often unequal distribution of burdens in the international system? Have past burden-sharing initiatives led to a more equitable distribution of burdens?

In order to address these questions, the first part of the paper proposes and analyses two burden-sharing approaches—one based on a 'cost-benefit'-, the other one based on a 'norm-based' logic—that offer partly competing and partly complementary hypotheses for answering the above questions. The second part, analyses the some of the available evidence of EU attempts to share burdens in the area of forced migration and discusses the respective usefulness of the two approaches in providing explanations for the motivation and patterns of burden-sharing in this area.

2. Two Approaches to Burden-Burden-Sharing

Political sociology identifies two principal logics of social action, a 'logic of expected consequences' (informed by a cost-benefit rationale) and a 'logic of appropriateness' (informed by a norm-based rationale) (March and Olsen 1998: 7-10). The former sees action as being driven by a logic of rational and strategic behaviour that anticipates consequences and is based on given preferences. Actors chose among alternatives by evaluating expected consequences of their actions for the achievement of certain objectives, expecting other actors to do the same. In this rational choice informed model, actors assess their goals, interests and desires independently of institutions.[3] In other words, it is assumed that actors' preference formation is external to the institutional context in which actors find themselves. Institutions affect only the strategic opportunities for achieving certain objectives (Immergut 1997: 231).

The latter 'logic of approprieteness', views action as being guided by notions of identity and roles shaped by the institutional context in which actors operate. According to this logic, action is based on rules, practices and norms that are socially constructed, publicly known and anticipated. Norm-based approaches emphasise that the motivations, choices and strategic calculations of political actors are framed by institutional context, which shapes opportunities for action. Behaviour often can be associated with what is considered 'appropriate' in a particular socio-cultural context. Such a perspective raises the question, to what extent an actor's broader institutional environment can lead to norm-guided behaviour that may supplant strategic calculation. These approaches regard institutions as a political environment or cultural context which shape an individual's interests, i.e. actors are conditioned (as to their identity, priorities and interpretations of reality) by institutions over time. Decisions are often taken according to what is considered 'appropriate' behaviour, with institutional norms being the main shapers of such notions of 'appropriateness' (Knill and Lenschow 1998). This norm-guided account therefore suggests that a calculus of identity and appropriateness is sometimes more important to actors than a calculus of political costs and benefits (March and Olsen 1989).[4]

Two possible explanations for co-operation and burden-sharing emerge. On the one hand, one can point out the irrationality of egoism and what the actors would gain by co-operation: it may be rational to sacrifice opportunities for individual action and co-operate to achieve collective goods instead (cost-benefit approach). On the other hand, a reflection on the conflict between individual and collective action may denounce egoism as undesirable and seek to attain joint goals by appeals to normative notions such as that of solidarity (norm-based approach). These two burden-sharing approaches and the insights they provide as to the questions of 'motivation' and 'patterns' of international burden-sharing will be outlined in more detail below.

2.1 The 'Cost-Benefit' Approach

Much of the existing literature on international burden-sharing is based on assumptions that are informed by a cost-benefit logic of social action.[5]

Explaining motives for burden-sharing

Two potential motives are highlighted in the literature. First, according to the theory of public goods (Olsen 1965; Hartley and Sandler 1999; Bolks, Stoll, et al. 2000), which remains the most influential approach to the analysis of burden-sharing, co-operation produces positive-sum benefits which in turn creates the will to share burdens/costs among actors as the benefits of the contribution exceed the costs of the contributions. At the international level, co-operation is thought to produce a level in the provision of valued public goods which an individual state cannot attain on their own. Public goods theory suggests that we should expect a significant under-provision of international public goods in the absence of central taxation and enforcement authorities.[6] Burden-sharing can thus be seen as a rational response to the problem of underprovision.

A second potential motive for burden-sharing based on cost-benefit considerations, is the insurance rationale.[7] A suitable burden-sharing regime can provide a degree of mutual insurance again the occurrence of a particular external shock that might put pressures on one or few countries. Burden-sharing schemes allow states to set off today's contributions against the expected reduced costs in a future crisis situations. From a cost-benefit perspective, however, such scheme can only be expected to include those who perceive of a particular risk to be shared and will only be agreed upon when contributions reflect the differences in the risk faced by each participant.

Explaining patterns of burden-sharing

From a public goods perspective, Mancur Olsen suggests that in any burden-sharing system, the distribution of costs and benefits will be skewed against the larger participants. According to Olsen's argument it is, ceteris paribus, the larger states whose action will make more of a difference to the total common effort than the actions of small states. As a result, larger states will have an incentive to contribute a disproportionate share to the overall effort. We will therefore observe a systematic tendency for "exploitation" of the big by the small (Olsen 1965: 29). In confirmation of this hypothesis a number of studies have analysed the correlation of defence spending and GNP in the NATO context. These studies have shown that the United States, as NATO's largest Member State, has contributed a disproportionate large share of the burden for common defence (Olson and Zeckhauser 1966; Oneal 1990).[8]

More recently, economists have developed a refined version of the public goods approach based on the so-called 'joint product' model (Sandler and Forbes 1980; Sandler 1992). A public good is defined by its properties of non-excludability and non rivalry. It is these properties which set it apart from a private good. The provision of a public good such as deterrence from a hostile attach resulting from collective defence benefits all members of a community and these benefits are extended to other actors at no marginal cost. However, as Sander and Forbes (1980) rightly point out that defence provides a spectrum of outputs ranging from purely public to private or country-specific defence outputs. This means that 'defence provides more than the single output of deterrence implied by the pure public goods model: it also privies protection and damage limitation when deterrence fails, as well as national or private benefits' (Hartley and Sandler 1999). In other words what is often regarded as a public good has in fact excludable private benefits (e.g. maintaining domestic order) and is 'rival' in character (e.g. conventional forces being subject to effects of 'force thinning'). From this revised model what we would expect that contributions to the provision of the particular good in question (which has both public and private characteristics) will be positively related to the proportion of excludable benefits to a particular state (in the defence case: the proportion of excludable defence outputs supplied by an alliance).

2.2 The 'Norm-Based' Approach

Explaining motives of burden-sharing

The prisoner dilemma points to a constellation where actors who act solely with the aim of maximizing their own utility will produce a result which is contrary to their interest (Luce and Raiffa 1957; Rapoport and Chammah 1965). Acting in solidarity under conditions of this kind is dependent on actors not acting according to the principle of utility maximisation but according to a the principle of universalisation, i.e. if they act in such a way they would wish all others to act as well. The principle of universalisation thus forbids placing the costs for providing mutually desired goods only on the shoulders of others. Action on this basis is driven by the norm of fairness (Baurmann 1999: 253).

The existence of solidarity can also be seen as providing a way out of situations which have the structure of a 'prisoner's dilemma'. In the original prisoner's dilemma (Rappapart and Chammah 1965), two prisoners who are accused of a crime are interviewed separately and offered a deal. If only one confesses, that prisoner alone will be released, whereas the other will receive a harsh sentence. If both confess, each will receive an intermediate sentence. If both refuse to confess, they will get a milder sentence for some different crime which can be proven independently of either confessing. If each prisoner is concerned solely with their own interest, it is rational for them to confess. But if each is unwilling to receive a benefit when this will harm the other, then each will refuse to confess, and both will be better off.

Approaches that emphasise norm-guided behaviour and highlight notions of solidarity offer an explanation to the question ‘why share costs’ that is an complementary or alternative account to more prominent cost-benefit models.[9] Solidarity, as a motivation for burden-sharing can be seen in two ways (Mason 2000). First, as a commitment to other members of a group to abide by the outcome of their collective decision-making. Even in many-actor cases, where it is often the case that one's own actions will make no appreciable difference to the overall outcome, the concept of solidarity still matters when the community can be understood to have made some collective decision and solidarity is understood as a commitment to abide by such a decision. Second, solidarity can be understood as a concern for other members of a group, which may be expressed by an unwillingness to receive a benefit unless the others do, or an unwillingness to receive a benefit when this will harm them. This commitment to the well-being of others is sometimes conceived in terms of the recognition of special obligations between the members of a group, which exist in virtue of their being members of it. Solidarity therefore can be said to exist among a group of actors when they are committed to abide by the outcome of some process of collective decision-making, or to promote the wellbeing of other members of the group, perhaps at significant cost to themselves.

Explaining patterns of burden-sharing

From a norm-based perspective, patterns of burden-sharing can be explained in two ways. First, burden-sharing bargains can be guided by notions of equity, basing the distribution of burdens on some key that is linked to the actual capacity of the different participants of the burden-sharing regime. A second way of explaining patterns from a norm-based perspective is to look at variations of the participating states commitment to norms that are related to the burden to be shared. From this perspective, a state's commitment to norms such as international human rights or distributive justice is positively related to the burdens that a state is prepared to accept.

The two above approaches thus offer alternative explanations for questions regarding the often interdependent dynamics of motivations for, and resulting patterns of, European burden-sharing. While the question of motivation ultimately offers only limited opportunity for empirical testing by political scientists, the following competing hypotheses for explanations of patterns of international burden-sharing can be formulated:

H1: The greater the difference that a state can make to the provision of a valued public good, the (disproportionately) greater will be its contribution to the burden-sharing regime as other states will have an incentive to free-ride (the exploitation of the big by the small hypothesis)

H2: The greater the excludable benefits that a state received from a burden-sharing scheme, the greater its willingness to contribute to the scheme (the 'joint product hypothesis)

H3: The greater the capacity of a state to contribute to the achievement of a valued good, the greater will be its willingness to contribute (the equity/solidarity hypothesis)

H4: The greater the commitment of a state to a particular norm (right) that is safeguarded by a particular burden-sharing scheme, the greater its willingness to accept costs resulting from the establishment of such a scheme (the protection of 'rights' hypothesis).

In the following section, we will use evidence from recent European burden-sharing initiatives in the area of forced migration, in order to put these four hypothesis to an initial empirical test.

3. Burden-Sharing in the EU: The Case of Forced Migration

Some of the most influential accounts of European integration (Moravcsik 1998) emphasise the opportunity for gain from cooperation as the principal driving force behind the integration project and have generally been sceptical about the potential for burden-sharing arrangements. In the absence of win-win situations (possibility of pareto improvements), the use of side-payments to compensate the actual or potential losers of further integration is often regarded as a necessary condition for reaching an agreement. From this perspective, action is above all driven by cost-benefit calculations and references to solidarity in the EU Treaties are regarded as window dressing and cheap talk.

The developments in the area of Justice and Home Affairs at Maastricht and Amsterdam have constituted an important step toward further integration in Europe. The extent to which these development must also be seen as a watershed in burden-sharing terms—one that potentially breaks with the strong underlying cost-benefit approach to distributive and redistributive issues at the European level—will be analysed after the following sketch of recent initiatives in this area.

3.1 Steps undertaken to achieve a 'balance of efforts' in the area of refugee protection[10]

Since the late 1980s, the EU burden-sharing debate has moved beyond the issue of net-budget contributions in the Common/Single Market. The principal area of such new burden-sharing dynamics has been the field of Justice and Home Affairs. Member State governments reacted to the termination of immigration controls at internal Community borders with a call for enhanced co-operation aimed at strengthened security at the Community's external border. In particular the question of how the challenge of international migration should be met and how its costs should be shared has moved to the top of the political agenda. It was the massive influx of the displaced persons during the first Balkan crisis in the early 1990s which brought burden-sharing questions into the limelight. Against the background of this crisis, a number of proposals and initiatives that envisaged the development of a comprehensive EU burden-sharing system in this policy area were made. These proposals encompassed suggestions for physical/ financial burden-sharing, both with regard to asylum seekers and temporary displaced persons. The first explicit references to such burden-sharing ambitions were made by EU ministers responsible for asylum and immigration at their meeting of 30 November and 1 December 1992.[11] These deliberations lead to a German Presidency Draft Council Resolution on Burden-sharing in July 1994.[12] This proposal foresaw the reception of refugees according to a key which was based on three criteria which were given equal weight (population size, size of Member State territory and GDP).[13] The centrepiece of the German draft foresaw the introduction of a compulsory resettlement mechanism that relied on the above distributive key and reads as follows:

‘Where the numbers admitted by a Member State exceeds its indicative figure [...], other Member States which have not yet reached their indicative figure [...] will accept persons from the first State’.

This proposal, however, did not find the necessary support in the Council.[14] In particular, the UK was strongly opposed to such a scheme.[15] Consensus emerged only much later during the French Presidency for the much watered down Council resolution of 25 September 1995 'on burden-sharing with regard to the admission and residence of displaced persons on a temporary basis'.[16] In it, no more mention is made about compulsory redistributive mechanism or indicative figures. Instead, it offers a number of 'soft' principles to guide states in the event of a mass influx of refugees, including 'the spirit of solidarity', 'equity of distribution' and 'harmonisation of response'. It takes account of British and French wishes in recognising that participation in peace-keeping operations can be counted against refugee admissions when assessing equity of carried burdens. This non-binding resolution was complemented by a Council decision of 4 March 'on alert and emergency procedure on burden-sharing with regard to the admission and residence of displaced persons on a temporary basis'.[17] Together, they poured cold water on the idea of facilitating a quick and automatic burden-sharing response in the case of a mass influx of displaced persons. One has to agree with Suhrke who states that 'evidently, a voluntary and ad hoc commitment to share in the spirit of solidarity represented the limits of the possible' (1998: 411). The ineffectiveness of these instruments became evident during the Kosovo crisis when neither of the two instruments were called upon by the Member States.

While these specific burden-sharing initiatives in the mid 1990s thus had very limited effects, other EU activities in this area were more path-breaking and had some quite paradoxical consequences on the distribution of burdens across Member States. Two areas are particularly worth mentioning. First, the flagship of the EU asylum aquis—the Dublin Convention—not only affects the individual asylum-seeker but also the protection responsibilities of the each Member State. The Dublin Convention provides the rule that the 'Member State of first entry' is the one responsible for dealing with a particular asylum claim. Asylum seekers who move to another Member State as a secondary movement can be send back to the 'state of first entry'.[18] Second, EU institutions and Member States have developed elaborate mechanism for deflecting migration movements from the Union's territory. Such mechanisms comprise pre-entry measures—such as carrier sanctions and technical assistance to third countries' exit control—as well as post-entry measures, such as rules for sending back asylum seekers to safe third countries.

However, the ambitions for a comprehensive burden-sharing system in this area were not shelved and were reiterated in the text of the Amsterdam Treaty of October 1997, Article 63 (ex 73k) of states that the Council

shall, within a period of five years after the entry force of the Treaty of Amsterdam, adopt […] measures on refugees and displaced persons within the following areas:
(a) minimum standards for giving temporary protection to displaced persons from third countries who cannot return to their country of origin and for persons who otherwise need international protection,
(b) promoting a balance of effort between Member states in receiving and bearing the consequences of receiving refugees and displaced persons

The special European Council in Tampere on Justice and Home Affairs on 15 and 16 October 1999, called for the establishment of a financial reserve for the implementation of emergency measures to provide temporary protection in the event of a mass influx of refugees.[19] Against this background, the Commission launched a number of experimental instruments during the 1997 to 1999 period.[20] The aim of these 'Joint Actions' was not only to test new methods of burden-sharing but also Member States' willingness for further steps towards more comprehensive solutions. These initiatives therefore formed the basis for the Council Decision of 28 September 2000 establishing the European Refugee Fund (ERF).[21]

Created on the basis of Article 63(2)(b) of the Treaty establishing the European Community, the ERF is to allocate resources proportionately to the burden on each Member State by reason of their efforts in receiving refugees and displaced persons. Its rationale is ‘to demonstrate solidarity between Member States by achieving a balance in the efforts made by those Member States in receiving refugees and displaced persons and bearing the consequences of so doing’ (para 21).[22] The ERF will initially operate from 1 January 2000 to 31 December 2004 and disperse total volume of EUR 216 million according to two elements, a fixed and a proportional one.[23] First, the Fund will disperse an equal flat rate amount to each participating Member State[24] irrespective of the number of displaced persons in its territory . This element which runs counter the redistributive objectives of the Fund will be scaled down over the five year period of the Fund’s initial operational period. The remaining resources shall be distributed in proportion to the number of displaced persons in each Member States.

3.2 Explaining the motives behind burden-sharing

With regard to the question of why states initiated steps towards a burden-sharing regime in the area of forced migration, cost-benefit approaches appear to offer the most convincing account of developments. From this perspective, burden-sharing offers some countries the prospect of reducing their own costs. It is therefore not at all surprising that the first substantial burden-sharing proposals in this area in the early 1990s were initiated by Germany, the western country most affected by the war in former Yugoslavia.

Having said this, it is important to note that in the official text establishing the EU's burden-sharing instruments in this area put heavy emphasis on notions of solidarity and fairness. The text of the ERF decision is only one of many examples. It states that the

‘implementation of such a policy [common policy on asylum] should be based on solidarity between member States and requires the existence of mechanisms intended to promote a balance in the efforts made by the Member States in receiving and bearing the consequences of receiving refugees and displaced persons’ (para 2)

It goes on to say that

'it is fair to allocate resources [from the ERF] proportionately to the burden on each Member State by reason of its efforts in receiving refugees and displaced persons'.[25]

While many would dismiss these pledges as non-binding and therefore inconsequential, there can be little doubt that since the start of the integration process, some of the EC’s most prominent political leaders, from Schuman to Kohl—while clearly being committed to pursuing what they saw as their country’s national interest—have viewed the integration process not merely in cost-benefit terms. They have seen the Community not just as an economic venture but also as an emerging political community.

A second argument in support of the claim that references to solidarity might not just constitute non-committal flowery statements can be seen in the fact that most, if not all, Member States have a long tradition of upholding constitutionally codified principles on the desirability for solidarity between regions within their state. The constitutions of all EU Member States contain provisions which foresee burden-sharing on the basis of some notion of solidarity between the different territorial entities and regions in cases of economic, financial or infrastructural imbalances.[26] Some lawyers have suggested that one should therefore regard solidarity as one of the Union's general principles of law as it constitutes an accepted norm in the domestic constitutions of the Member States (Schieffer 1998: 208-212). At least in part, such an interpretation appears to have been accepted by the European Court of Justice.[27]

Nonetheless, on balance it appears that while reference to solidarity and fairness appear to have played a part in selling these initiatives as part of the process towards an 'ever closer union', the timing of the proposals and the hard bargaining that characterised the establishment of the ERF (Noll 2000: ??) and the inability to agree on a distribution key in the case of temporary protection measures (in the case of mass influx)[28] lend strong support to explanations that are based on cost-benefit approaches. One question, however, remains and that is how the agreement of the likely net-losers from the ERF was obtained. The compromise regarding fixed lump-sum payments, irrespective of the number of refugees received, appears to be insufficient for an explanation based on a side-payment logic (in particular as the this concession will be phased out over time). The insurance rationale discussed above might point towards a more convincing explanation for why the immediate net-losers supported the establishment of the ERF [EVIDENCE?]. The relatively small size of the Fund, as well as it initially limited timeframe, can be assumed to have also facilitated the reaching of the agreement on the ERF.

3.3 Explaining patterns of burden-sharing

When looking at the number of refugees received by European countries over the last ten years, one cannot fail to notice the very large differences in reception figures. To what extent do these figures offer support to one or more of the earlier proposed hypotheses?

Based on a cost-benefit perspective, a number of scholars, most prominently Suhrke (1998), have employed public goods theory to analyse international cooperation and burden-sharing in the area of forced migration. Suhrke argues that refugee reception is an international public good from which all state benefit, irrespective of which country receives the refugees. Increased security can be regarded as the principal benefit, as an accommodation of refugees can be expected to reduce the risk of refugees fuelling and spreading the conflict they are fleeing from.

The Kosovo crisis of 1997 provides for an excellent opportunity to study the willingness of European states to share refugees. With neighbouring countries, in particular Macedonia, heavily overburdened by refugees, the UNHCR appealed to Western European states to accept refugees under the UNHCR's Humanitarian Evacuation Programme (HEP) under which the UNHCR would transport refugees to states which agreed to accept a particular number of Kosovo refugees. This meant that in this particular case, refugee reception rates were relatively free from the preferences of refugees themselves with regard to their preferred country of destination. The Kosovo case therefore allows a less distorted comparison of states' willingness to accept refugees. However, it can be seen in Table 1 that the HEP figures show a very similar pattern to overall rate of asylum applications across Europe.

Table 1: Evidence relating to cost-benefit hypothesis 1 (exploitation of the big by the small hypothesis)

R A N K		Number of Kosovo refugees accepted in per 100,000 inhabitants		Number of Kosovo refugees accepted per $1billion of GNP		Total Asylum Applications (per 1000 inhabitants) (averages 1990-99)		Total Asylum Applications per GNP ($ billion)		Gross National Product ($ billion) (1999)		Population (1999)	R A N K
1	Austria	627.94	Austria	24.78	Sweden	27.71	Sweden	1049.32	Germany	2095.00	Germany	82090000.00	1
2	Denmark	530.08	Denmark	16.40	Germany	22.90	Germany	897.18	United Kingdom	1451.00	United Kingdom	59500000.00	2
3	Sweden	420.99	Sweden	15.94	Denmark	21.14	Netherlands	811.97	France	1432.00	France	59100000.00	3
4	Ireland	274.67	Ireland	13.04	Netherlands	20.34	Belgium	721.60	Italy	1171.00	Italy	57080000.00	4
5	Netherlands	256.80	Portugal	11.76	Belgium	17.62	Denmark	653.95	Spain	592.00	Spain	39420000.00	5
6	Luxembourg	232.56	Netherlands	10.25	Austria	16.03	Austria	632.73	Netherlands	396.00	Netherlands	15810000.00	6
7	Finland	191.49	Finland	7.80	Luxembourg	13.26	Luxembourg	316.67	Belgium	250.00	Greece	10530000.00	7
8	Germany	178.95	Germany	7.01	United Kingdom	6.29	United Kingdom	257.85	Sweden	234.00	Belgium	10240000.00	8
9	Portugal	127.13	Luxembourg	5.56	France	5.02	Ireland	232.28	Austria	205.00	Portugal	9990000.00	9
10	Belgium	119.14	Italy	4.98	Ireland	4.89	France	207.30	Denmark	172.00	Sweden	8860000.00	10
11	France	106.60	Belgium	4.88	Finland	3.53	Greece	196.88	Finland	127.00	Austria	8090000.00	11
12	Italy	102.14	France	4.40	Greece	2.34	Finland	143.86	Greece	125.00	Denmark	5320000.00	12
13	United Kingdom	72.44	United Kingdom	2.97	Spain	2.12	Spain	141.15	Portugal	108.00	Finland	5170000.00	13
14	Spain	36.28	Spain	2.42	Italy	1.57	Italy	76.46	Ireland	79.00	Ireland	3750000.00	14
15	Greece	0.00	Greece	0.00	Portugal	0.56	Portugal	51.76	Luxembourg	18.00	Luxembourg	430000.00	15

The figures in Table 1 make it clear that the size of a Member State (in terms of population or GNP) is not a particularly strong indicator for relative refugee burdens in the EU. Although absolute numbers, in particular those of the early 1990s appear lend some support for Olson's hypothesis, if one puts the number of annual asylum applications in relation with the size of Member States' populations (or GNP), one finds that the top positions in the Table and thus the highest 'burdens' are predominantly taken by smaller states. In other words, the public goods informed hypothesis regarding the expected 'exploitation of the big by the small' cannot be confirmed.

The joint product hypothesis in contrast looks somewhat more promising. It suggests that the amount of free-riding in the provision of a particular good will be inversely related to the proportion of excludable outputs that the regime provides. In other words, the more country specific the benefits from accommodating refugees, the fewer opportunities will exist to free-ride. According to this model we might therefore expect countries 'closer' to a particular conflict from which refugees emanate of having the greatest interest in the stabilisation of the crisis, and thus a greater willingness to accommodate refugees. By accepting displaced persons, a neighbouring country might expect to help in diffusing a crisis, thus reducing the danger of the conflict spilling over into its own territory. In Table 2, 'closeness' in the Kosovo case is measured by geographic proximity between Member States' capitals and the capital of Kosovo, Pristina.[29]

Table 2: Evidence relating to cost-benefit hypothesis 2 (the 'joint product hypothesis)

	Number of Kosovo refugees accepted in per 100,000 inhabitants		Distance (national capital to Pristina, in miles)
Austria	627.94	Greece	424.00
Denmark	530.08	Austria	527.00
Sweden	420.99	Italy	555.00
Ireland	274.67	Germany	886.00
Netherlands	256.80	Luxembourg	1034.00
Luxembourg	232.56	Denmark	1128.00
Finland	191.49	Belgium	1161.00
Germany	178.95	Netherlands	1199.00
Portugal	127.13	France	1202.00
Belgium	119.14	Sweden	1327.00
France	106.60	United Kingdom	1368.00
Italy	102.14	Finland	1403.00
United Kingdom	72.44	Spain	1607.00
Spain	36.28	Ireland	1693.00
Greece	0.00	Portugal	1966.00

While the Austrian case appears to support this hypothesis, the other results are more ambiguous.[30] Instead of close by countries such as Italy, Germany and Greece, it was more 'remote' countries such as Denmark and Sweden which accepted the highest number of Kosovo refugees in per capita terms. However, one can reasonably suspect (CHECK UNHCR FIGURES??) that Germany and Italy did take a large number of Kosovo refugees even before the start of the UNHCR's Humanitarian Evacuation Programme which might therefore explain their relatively low share. In case of Greece, Greek sensibilities concerning the countries Albanian minority in the north of the country meant that Greece was concerned about heightened ethnic tensions on its own territory, concerns which appeared to have outweighed broader international security considerations.

The explanatory strength of public goods approaches that are based on cost-benefit considerations in the area of forced migration therefore appears to be weaker than in other areas of collective burden-sharing (such as defence) and less useful than has sometimes been suggested in the literature. Do norm-based explanations fare any better?

When looking at the evolution of refugee burdens in Europe over the period of the last ten years (Table 3), the figures presented suggest that despite the frequent calls for, and references to, European solidarity and the principle of fairness when tackling common challenges, there so far appears to be little evidence that such norms have played a significant role in guiding the behaviour of states.

Table 3: Evidence relating to norm-based hypothesis 1 (equity hypothesis)

Evolution of Asylum Burden (per 100,000 inhabitants)

Country	1990	1991	1992	1993	1994	1995	1996	1997	1998	1999
Austria	282	338	201	59	63	73	86	83	171	248
Belgium	127	148	172	263	140	112	121	115	215	349
Denmark1	357	243	377	310	150	189	139	105	114	131
Finland	102	89	268	278	129	99	114	98	110	125
France	93	80	49	47	44	34	29	36	38	52
Germany3	235	312	534	393	155	156	142	127	120	116
Greece	59	25	18	8	12	12	16	42	28	15
Ireland	0	1	1	2	10	11	31	103	123	206
Italy5	8	46	11	3	3	3	1	3	19	58
Luxembourg	0	0	0	0	0	91	60	100	398	677
Netherlands	134	137	129	224	333	185	140	218	286	249
Portugal	1	3	7	21	8	5	3	3	4	3
Spain	22	21	30	32	30	14	12	13	17	21
Sweden	332	309	948	424	210	102	65	109	145	127
United Kingdom7	44	75	41	38	55	74	50	55	77	120
Standard Deviation	124	122	264	156	96	64	53	57	110	172

Notes

1 Applications lodged by asylum-seekers arriving spontaneously in Denmark and those lodged at embassies abroad.

3 Excluding applications which have been "re-opened" (since 1997).

5 The 1999 figure, provided by the Government, includes applications not yet officially recorded by the Eligibility Commission. During 1999, the Eligibility Commission received 12,146 asylum applicants.

7 Figures refer to the number of applications. On average, there are some 1.3 persons per asylum application in the UK.

Like in the early 1990s, when the Bosnian refugee crises led to the first calls for a EU wide refugee burden-sharing regime, it is still true today that a number of mainly smaller states carry a disproportionate share of the of the burden/responsibility of receiving refugees. The standard deviation of asylum applications per head in Table 3 shows that the distribution of asylum seekers remains highly inequitable, and that this inequality has steadily increased over the past few years. This supports the conclusion drawn by Noll and others in this volume that the burden-sharing instruments developed by the EU in the 1990s have been rather ineffective (and in some cases, such as that of the Dublin Convention, even counterproductive) in addressing the inequitable distribution of refugees in the EU.

One can therefore conclude that there is so far little evidence that the norms of solidarity and equity have guided recipient states in their relationship with other countries inside and outside the EU, with states, despite their rhetoric being rather innovative in creating instruments that shift the refugee burden to other state and non-state actors (carriers etc.).[31] There appears to be evidence, however, that norm-guided behaviour has played a significant role in the relationship between recipient Member States and protections seekers. Here we find substantive evidence for the claim that a country's willingness to receive refugees is positively related to its more general commitment to norms such as protection and distributive justice. Using overseas development aid, recognition rates and domestic social spending in Table 4 as proxies for a state's commitment to such norms we find quite a good correspondence with the relative willingness of a state to accept Kosovo refugees under the HEP.[32] While the correlation is not statistically significant at conventional levels in the case of social protection expenditure, there is quite strong, and statistically significant, correlation between the intake of Kosovo refugees and states' foreign aid payments (tau_b=325, p<0.05, one-tailed test); and similarly vis-à-vis recognition rates (tau_b=333, p<0.05, one-tailed test). In other words, the hypothesis that a state's willingness to accept burdens (costs) will be related to its commitment to particular norms and the protection of certain rights is confirmed by this analysis.

Table 4: Evidence relating to norm-based hypothesis 2 (the 'rights' hypothesis)

R A N K		Number of Kosovo refugees accepted (per 100,000 inhabitants) (under HEP)**		Total Asylum Applications (per 1000 inhabitants) (averages 1990-99)		Grant equivalent of total Overseas Development Aid as % of GNP (1998-99 average)		Expenditure on social protection (as % of GDP) (1998)		Total recognition of asylum-seekers (average 1990-1999) (in %)*		Stock of foreign population from former Yugoslavia (1995) (percentage of total foreign stock)	R A N K
1	Austria	627.94	Sweden	27.71	Denmark	1.02	Sweden	33.30	Austria	73.50	Austria	45.55	1
2	Denmark	530.08	Germany	22.90	Netherlands	0.85	France	30.50	Denmark	50.80	Germany	15.52	2
3	Sweden	420.99	Denmark	21.14	Sweden	0.66	Denmark	30.00	Sweden	49.70	Denmark	12.61	3
4	Ireland	274.67	Netherlands	20.34	Luxembourg	0.65	Germany	29.30	Ireland	43.40	Sweden	7.22	4
5	Netherlands	256.80	Belgium	17.62	France	0.45	Netherlands	28.50	Portugal	38.80	Italy	5.66	5
6	Luxembourg	232.56	Austria	16.03	Finland	0.35	Austria	28.40	Netherlands	29.40	Belgium	4.708	6
7	Finland	191.49	Luxembourg	13.26	Belgium	0.34	Belgium	27.50	Finland	24.50	Netherlands	4.62	7
8	Germany	178.95	United Kingdom	6.29	Greece	0.30	Finland	27.20	Germany	20.00	Greece	3.66	8
9	Portugal	127.13	France	5.02	Germany	0.30	United Kingdom	26.80	Luxembourg	17.90	Finland	3.50	9
10	Belgium	119.14	Ireland	4.89	Portugal	0.28	Italy	25.20	Italy	15.60	France	1.50	10
11	France	106.60	Finland	3.53	United Kingdom	0.27	Greece	24.50	Belgium	13.10	Ireland		11
12	Italy	102.14	Greece	2.34	Austria	0.26	Luxembourg	24.10	France	12.30	Luxembourg		12
13	United Kingdom	72.44	Spain	2.12	Spain	0.25	Portugal	23.40	United Kingdom	11.50	Portugal		13
14	Spain	36.28	Italy	1.57	Italy	0.19	Spain	21.60	Spain	9.90	Spain		14
15	Greece	0.00	Portugal	0.56	Ireland	0.15	Ireland	16.10	Greece	8.10	United Kingdom		15

* Convention status and subsidiary forms of protection status granted

** Departures of Kosovar refugees from FYR of Macedonia up to 31 October 1999; Source: UNHCR/IOM.

4. Conclusion

Recent EU initiatives in the area of forced migration (although clearly significant in institutional terms) do not appear to constitute a radical departure for the EU in burden-sharing terms. While it is remarkable that Member States managed to agree at all on the European Refugee Fund, a new (albeit small) EU redistributive instrument, without the provision of any significant side-payments, the negotiations leading to the creation of the ERF and the other burden-sharing instruments that have been created, once again have shown that interstate solidarity in the EU remains very low. The motivation for the creation of burden-sharing instruments, however, should nonetheless not exclusively be seen in exclusively cost-benefit terms. The move towards burden-sharing appear to have originated not out of a concern with other Member States (and their disproportionate share of the burden) but out of a concern over potential threats to the European integration project and the system of international refugee protection. It was feared that in the absence of a common European approach unequal migration pressures would pose not only a threat to the Single Market (and in particular the achievement of the principle of free movement within it), but that such uncoordinated action might also lead to an unravelling of central international human rights/refugee norms. In other words, a failure to agree on a some common approach would have not only increased pressures for a re-establishment of border controls in the Schengen area, thus threatening the operation of the Single Market but would have also accelerated the drive towards burden-shifting and moves towards the lowest common denominator in reception standards which would have had the potential of undermining some of the core principles of the international refugee/human rights regime, which, despite its shortcomings,[33] has clearly shaped interests and identities of policy makers over the last fifty years (Risse and Sikkink 1999) and continues to do so. The motivation for the attempt to find a common approach to the sharing of legislation, resources and refugees therefore appears to have been driven by both a 'cost-benefit'- as well as a 'norm-driven' logic.

When seeking to explain the pattern of refugee burden-sharing in Europe, one cannot fail to notice the shortcomings of public goods approaches that were developed in the NATO burden-sharing context. An analysis of relative refugee reception rates in relation to population figures or GNP leads us to reject a standard public goods hypothesis. With attempts to assess the 'private'/'excludable', country specific benefits of refugee protection (joint product model) producing at only ambiguous results, looking at countries' commitment to distributive and humanitarian norms as an alternative determinant for explaining the varying willingness of states to accept refugees can offer a plausible supplementary set of norm-based explanations. For these, while keeping in mind the limited number of variables included in the above analysis, [34] the paper provides some first supportive evidence. Most importantly therefore, the above analysis therefore support the claim that the two logics of social action outlined above are not mutually exclusive (March and Olsen 1998: 10). Political actors are constituted both by their interests, through which they evaluate anticipated consequences, and by the norms embedded in political institutions and their own identities. They calculate consequences and follow norms, and the relation between the two is often subtle.

REFERENCES

Baurmann, M. (1999). Solidarity as a Social Norm and as a Constitutional Norm. Solidarity. K. Bayertz. Dordrecht, Kluwer.

Begg, I. and N. Grimwade (1998). Paying for Europe. Sheffield, Sheffield Academic Press.

Böcker, A. and T. Havinga (1998). Asylum Migration to the European Union: Patterns of origin and destination. Luxembourg, Office for Official Publications of the European Communities.

Bolks, S., R. J. Stoll, et al. (2000). Alliance Dynamics and State Capability: Making Security Policy Choices. APSA Conference, Washington.

Boyer, M. A. (1989). “Trading public goods in the Western alliance system.” Journal of Conflict Resolution 33(4).

Chalmers, M. (2000). Sharing Security: The Political Economy of Burden-sharing. London, Macmillan.

Checkel, J. T. (1999), Sanctions, Social Learning and Institutions: Explaining State Compliance with Norms of the European Human Rights Regime. ARENA Working Papers. Oslo, ARENA. 99/11.

Durkheim, É. (1983) De la division du travail social, trans. W.D. Halls, The Division of Labour in Society, London: Macmillan.

Hagen, Roar (2000), Rational Solidarity and Functional Differentiation, Acta Sociologica, Vol. 43, pp. 27-42.

Hartley, K. and T. Sandler (1999). “NATO Burden-Sharing: Past and Future.” Journal of Peace Research 36(6).

Harvey, C. J. (1998). “The European Regulation of Asylum: Constructing a Model of Regional Solidarity?” European Public Law 4(4): 561-92.

Hooghe, L. and Keating, M. (1994), The politics of European Union regional policy, Journal of European Public Policy, Vol. 1, No. 3, pp. 367-393.

Hurwitz, A. (1999). “The 1990 Dublin Convention: A Comprehensive Assessment.” International Journal of Refugee Law 11(4): 646-77.

Immergut, E. M. (1997), The Normative Roots of the New Institutionlism: Historical Institutionalism and Comparative Policy Studies, in A. Benz and W. Seibel (Eds.), Theorieentwicklung in der Politikwissenschaft--eine Zwischenbilanz, Baden Baden: Nomos, pp. 325-355.

Jandl, M. (1995). Structure and Costs of the Refugee and Asylum Systems in Seven European Countries. Vienna, ICMPD.

Keeler, J. (1996). “Agricultural Power in the European Community: Explaining the Fate of CAP and GATT Negotiations.” Comparative Politics 28(1): 127-49.

Kerber, K. (1997), Temporary Protection: as assessment of the harmonisation policies of the European Union Member States, International Journal of Refugee Law, 9

Knill, C. and Lenschow, A. (1998), Coping with Europe: The impact of British and German administrations on the implementation of EU environmental policy, Journal of European Public Policy, Vol. 5, No. 4, pp. 595-614.

Krasner, S. (1983), International Regimes, Ithaca, NY: Cornell University Press.

Lavenex, S. (1999). Safe Third Countries, Central European Press.

Lowndes, V. (1996), Varieties of New Institutionalism: A Critical Appraisal, Public Administration, Vol. 74, No. summer, pp. 181-97.

Luce, D. R. and H. Raiffa (1957). Games and Decisions. New York: Wiley.

Lumsdaine, D. H. (1993). Moral Vision in International Politics: The Foreign Aid Regime 1949-1989. Princeton, Princeton University Press.

March, J. G. and Olsen, J. P. (1989), Rediscovering Institutions: The Organizational Basis of Politics, New York: Free Press.

March, J. G. and Olsen, J. P. (1998), The Institutional Dynamics of International Political Orders, ARENA Working Paper 98/5, ARENA - University of Oslo, Olso.

Mason, A. (2000). Community, solidarity, and belonging : levels of community and their normative significance. Cambridge, Cambridge University Press.

Moravcsik, A. (1998). The choice for Europe: social purpose and state power from Messina to Maastricht. Ithaca, N.Y., Cornell University Press.

Noll, G. (1997). “Prisoner's Dilemma in Fortress Europe: On the Prospects for Equitable Burden-Sharing in the European Union.” German Yearbook of International Law 40: 405-37.

Noll, G. (2000). Negotiating Asylum. The Hague, Nijhoff.

Olsen, M. (1965). The Logic of Collective Action: Public Goods and the Theory of Groups. Cambridge, Harvard University Press.

Olsen, M. and R. Zeckhauser (1966). “An economic theory of alliances.” Review of Economics and Statistics 48: 266-79.

Oneal, J. R. (1990). “The theory of collective action and burden-sharing in NATO.” International Organization 44(3): 379-402.

Paterson, L. A. (1997). “Agricultural Policy Reform in the European Community: a three-level game analysis.” International Organization 51(1): 135-65.

Rappapart, A. and A. M. Chammah (1965). Prisoner's Dilemma. Ann Arbor, University of Michigan Press.

Risse, T. and K. Sikkink (1999). The socialization of international human rights norms into domestic practices: introduction. The Power of Human Rights: International Norms and Domestic Change. T. Risse, S. C. Ropp and K. Sikkink. Cambridge, Cambridge University Press: 1-38.

Sandler, T. (1992). Collective Action--Theory and Applications, Harvester Wheatsheaf.

Sandler, T. and J. F. Forbes (1980). “Burden-Sharing, Strategy and the Design of NATO.” Economic Enquiry 18(3): 425-444.

Schieffer, M. (1998). Die Zusammenarbeit der EU-Mitgliedstaaten in den Bereichen Asyl und Einwanderung. Baden-Baden, Nomos.

Shackleton, M. (1990). Financing the European Community. London, Chatham House / Pinter.

Suhrke, A. (1998). “Burden-sharing during Refugee Emergencies: The Logic of Collective Action versus National Action.” Journal of Refugee Studies 11(4): 396-415.

Thielemann, E. (2001). Beyond the rational-/sociological institutional divide? Explaining stability and change in European asylum policy. Paper presented at the 2001 ECSA Conference, Madison, Wisconsin.

Tsoukalis, L. (1993). The New European Economy: The Politics and Economics of Integration. Oxford, Oxford University Press.

UNHCR (1995). 'Collection of international instruments and other legal texts concerning refugees and displaced persons', Volume II, Regional Instruments, UNHCR, Geneva 1995. Geneva, UNHCR.

UNHCR (1998), Asylum in Europe: Arrivals, Stay and Gender From a Data Perspective, Geneva: UNHCR.

Wallace, H. (1980). Budgetary Politics. London, Allen and Unwin.

Wallace, W. (1983). Less than a Federation, More than a Regime: The Community as a Political System. Policy-Making in the European Community. H. Wallace, W. Wallace and C. Webb. Chichester, Wiley and Sons.

---

[1] Previous versions of this paper were presented at the 42^nd Annual ISA Convention, Chicago, 21-24 February 2001, two workshops of the 'UACES Study Group on EU Burden-Sharing', the ECPR/IR Conference in Canterbury on 8-10 September 2001 and the 2001 APSA Annual Meeting in San Francisco. The author gratefully acknowledges comments and suggestions received on these occasions.

[2] Despite its prejudicial connotations (in particular in the context of forced migration), the term ‘burden-sharing’ is used throughout the text to address questions of how to share responsibility between different territorial units for the provision of collective goods. Other terms, such as the Commission’s call for an ‘equal balance of efforts’ between the Member States have so far failed to become common usage.

[3] Institutions here are defined in a broad sense as rules, norms and conventions that shape and inform human interaction. This definition has similarities with Krasner's definition of regimes which he defines as 'implicit or explicit principles norms, rules and decision-making procedures around which actors' expectations converge in a given area of international relations (1983: 2). It is useful to distinguish between institutions/regimes and organisations. The EU or NATO are understood here as organisational embodiments of particular economic and security institutions/regimes.

[4] To reiterate the general point. The key difference between the two accounts is not the dichotomy between material and non-material motivations. Even if goals are non-materialist, like in the area of international relations when actors adhere to certain norms for reasons of international standing, the underlying logic of action is often still consequentialist—means-ends—in nature (Checkel 1999).

[5] For an overview see Chalmers (2000: 1-20).

[6] The principal reason for such under-provision is linked to the problem of free-riding. To explain this problem Olsen (1965) points to the underlying tension between individual and collective rationality. If members of a group share an interest in having some policy adopted, this does not mean that it is in each individual's interest to provide a certain part of that good. Instead, according to Olson's logic of collective action, it is in the interest of the individual member to 'free-ride' on the efforts the others. Such arguments have for example been applied to the NATO context to explain not only the motives for co-operation on national defence but also to demonstrate that members of an alliance have an incentive to cheat by 'free-riding' on the common security provided by the other members. The difficulty to prevent free-riding therefore points to one of the principal obstacles to the provision of collective goods. The public goods literature suggests that the principal way to persuade an actor to contribute disproportionately to a burden-sharing deal is through the strategic use of side-payments, linking one's contributions in one bargain to the receipt of benefits in another bargain that can be linked to the original one. The existence of side-payment opportunities may thus help to reopen blocked negotiations.

[7] See also the contribution of Noll in this issue.

[8] For problems and inconsistencies with this general argument see Boyer (1989: 715).

[9] Durkheim distinguishes two types of solidarity. In his view mechanical solidarity exists in traditional societies where it emerges from the unmediated identity of its members. He distinguished this from a more modern form of solidarity—organic solidarity—which emerges from the functional interdependence of individuals' roles in modern societies. He argues that solidarity is therefore not just an old-fashioned concept that has all but lost its relevance. Increased interdependence that characterises modern societies, while undermining traditional notions of mechanical solidarity, also provides a new source of solidarity (Durkheim 1983).

[10] Noll, in this issue, makes the useful distinction between the sharing of norms, money and people as three interrelated aspects of burden-sharing in the area of forced migration. I will concentrate here on the latter two which have an explicit redistributive dimension. For a detailed account on legal harmonisation attempts in this area (the sharing of norms), see Ucarer and Hurwitz in this issue.

[11] Not published in the Official Journal but can be found in (UNHCR 1995).[also printed in: BMI (ed.) Texsammlung zur Europaeischen Asylpraxis).

[12] Council Document 7773/94 ASIM 124.

[13] The form of the suggested redistributive mechanism followed the example of German domestic legislation, which stipulates a similar key for the distribution of asylum seekers among the German Länder. See section 45 of the German Asylum Procedure Act (Asylverfahrensgesetz).

[14] A look at the table of indicative percentage shares contained in the German draft shows that the proposal would above all have improved the situation of Germany. The percentage shares read as follows: Germany (21.58), France (19.40), Italy (15.83), United Kingdom (14.28), Spain (13.63), Netherlands (3.55), Greece (3.20), Portugal (2.65), Belgium (2.42), Denmark (1.78), Ireland (1.54) and Luxembourg (0.12) (quoted in Noll 2000: 293, fn. 884).

[15] BMI, Pressemitteilung vom 1.12.1994, FAZ 27.1.1995, p.2; BT-Drs. 13/1070, 55; Integrationsbericht, p.92.

[16] OJ No C 262/1, 07.10.1995.

[17] In order to specify procedural steps necessary for the implementation of the September 1995 resolution, the March decision established a procedure to note whether a situation exists which requires concerted action by the European Union for the admission and residence of displaced persons on a temporary basis. It holds that the presidency, a member State or the Commission can call for an urgent meeting of the K4 Co-ordinating Committee. Council Decision on an alert and emergency procedure for burden-sharing (4 March 1996) (OJ No L63/10, 13.3.1996) (96/198/JHA).

[18] For an overview of the Dublin Convention see e.g. Hurwitz (1999).

[19] Council Document (SN 200/99).

[20] Joint Action of 22 July 1997 concerning the financing of projects in favour of asylum seekers and refugees (OJ (1997) L 205, pp. 5-6); Joint action of 22 July 1997 concerning the financing of specific projects in favour of displaced persons who have found temporary protection in the Member States and asylum seekers (OJ (1997) L 205, pp. 3-4). Corresponding initiatives a year later were the two Joint Actions of 27 March (OJ (1998) L 138, pp. 6-9)(COM (97) 93 Final, OJ 1997 No. C106/13). These initiatives were aimed at improving admissions and at facilitating repatriation respectively. The Joint Action of 26 April 1999 (OJ (1999) L 114, pp. 2-5), which was in part a reaction to the Kosovo crisis, develops these earlier joint actions by combining reception and voluntary repatriation in one instrument.

[21] OJ (2000) L 252/12 of 06/10/2000.

[22] To support these objectives, the Fund is to support Member States’ measures relating the reception, integration and the repatriation of refugees and displaced persons (Art. 4(1). Such action may apply in particular to accommodation, material aid, health care, social and legal assistance (Article 4(2)

[23] Five percent of the Fund are reserved for Community Initiatives (Article 5(1)).

[24] UK and Ireland have opted in (para 22); Denmark is not participating in the adoption of the decision (para 23).

[25] OJ (2000) L 252/12 of 06/10/2000 (para 11).

[26] The fact that there indeed appear to be strict limits to cross-border solidarity in the Union should not surprise, if one keeps in mind how vulnerable this notion is in the national context.

[27] The Court derives it from Article 5 EC Treaty which states that 'Member States shall take all appropriate measures, whether general or particular, to ensure fulfilment of the obligations arising out of this Treaty or resulting from action taken by the institutions of the Community. They shall facilitate the achievement of the Community's tasks. They shall abstain from any measure which could jeopardise the attainment of the objectives of this Treaty.' The principle of solidarity was first explicitly used and accepted as a general principle of European law arising from the particular nature of the Communities in the case 'Commission vs. Italy' (ECJ 1973, 102).See also Commission vs. Great Britain 128/78 (ECJ 1978, 419). For more details see Schieffer (1998: 204). Marias (1994), Legal Issues of European Integration, p. 85, 88. [see Schieffer 1998: 204, footnote 12] [LSE Library: MC JN18]

[28] Council Directive 2001/55/EC of 20 July 2001 on minimum standards for giving temporary protection in the event of a mass influx of displaced persons and on measures promoting a balance of efforts between Member States in receiving such persons and bearing the consequences thereof, OJ L 212/12, 7.8.2001

[29] This of course is only rough proxy for the notion of closeness which with better data should be extended by reference to other factors such as historic cultural, linguistic, religious, economic and other ties.

[30] Correlation analysis suggests a weak negative relationship but not at statistically significant levels.

[31] See Ucerer in this volume.

[32] See also Lumsdaine's (1993) analysis of the post-war foreign aid regime.

[33] As for example highlighted by Jack Straw, the British Home Office Minister (The Guardian, Home secretary wants Geneva convention rewritten, Tuesday February 6, 2001).

[34] For a discussion of a more extensive list of variables see Böcker and Havinga (1997).