2010

Bringing State Institutions back to the Limelight? (*)

Simone Bertoli (**), Elisa Ticci (***)

1 State Institutions and Development

In the early days of development economics, the state undoubtedly played a pivotal role in driving and shaping the development process, as in the seminal contributions by Rosenstein-Rodan (1943) or Rostow (1959). The neo-classical revolution in economic theory progressively eroded the earlier consensus on the role of the state in the economic system, as its ability to correct market failures was questioned, and the emphasis was shifted on the inefficiencies created by state interventions. The narrower role of the state in the domestic economic system was also matched by a call for a progressive dismantling of any interference with the economic interactions between the country and the rest of the world. The state should have let the forces of globalization to play their role, which - so the reasoning went - would have contributed to promote economic growth, after the failure of state-driven approaches to development.

This long-term neglect or mistrust of the role of the state might be approaching to a turning point, given the increased emphasis on the challenges and risks posed by countries whose state institutions are deemed as fragile. The discourse around fragility is - albeit indirectly - bringing the state back to the limelight, as it opens the door for a broader reflection on the role of state institutions in helping a country to cope with the uncertainties and risk that the interdependence and interconnectedness of a globalized world creates.

2 Fragility: A New Developmental Concept

The dominant academic and policy-oriented discourse on economic development has been progressively adopting concepts that had been proposed in different political arenas or academic fields. Fragility is one of these concepts, which originally emerged in the national security arena, as the 9/11 terrorist attacks strengthened the perception that countries whose state institutions are unable to control their territory effectively could represent a threat to global security. According to the definition of fragile states provided by the OECD Principles for Good International Engagement in Fragile States and Situations, "states are fragile when state structures lack political will and/or capacity to provide the basic functions needed for poverty reduction, development and to safeguard the security and human rights of their populations" (OECD/DAC, 2007).

Robert B. Zoellick, the president of the World Bank, argued that "fragile states are the toughest development challenge of our era" (Zoellick, 2008), and the relevance of this challenge is reflected by the fact that the 2011 edition of the flagship publication of the World Bank, the World Development Report, will focus on conflicts and fragility. Fragility is currently moving up fast in the list of priorities in the development arena, but any statement about its relevance can be matched by an equal - or probably higher - number of caveats about its fuzziness and indeterminacy, which could undermine its analytical salience and operational value-added. Indeed, "despite its importance as part of the international policy discourse, the idea of state fragility remains an elusive concept" (Carment et al., 2008). This is not surprising, as "[...] there are generally no uniquely correct definitions of concepts drawn from common parlance and then used in a rather different context" (Brown and Stewart, 2009), as the lengthy process of refinement undergone by the concept of vulnerability shows (Guillaumont, 2009).

A notable dimension of this ambiguity is represented by the fact that either country or state is interchangeably matched with the label "fragile". To provide a telling example, we can observe that the World Bank, where the expression Low-Income Countries Under Stress (LICUS) had been coined, proposed to adopt, in its own work, the term fragile state in order "to facilitate a harmonized approach, and recognizing the increased international use of the term fragile states" (World Bank, 2005), as if the choice of the referring to either a country or a state was immaterial. (1)

Besides the fuzziness that surrounds the definition of fragility, Cammack et al., (2006), observe that:

"the word 'fragile' is often substituted without a precise change in meaning by 'failed', 'failing', 'crisis', 'weak', 'rogue', 'collapsed', 'poorly performing', 'ineffective', or 'shadow'; a fragile state may also be called a 'country at risk of instability' or 'under stress', or even a 'difficult partner'." (2)

It is also a common practice to refer to the countries in the Bottom Billion as fragile countries (for example, Zoellick, 2008), though this list of countries was defined - but not disclosed (3) - by Paul Collier on a set of criteria which are admittedly unrelated to any definition of fragility. (4)

The loosely-defined character of the concept of fragility is a disturbing feature from both an academic and from a policy-oriented perspective, as an unwarranted perception of coincidence among rather different approaches which use the same jargon can be highly misleading. The purpose of this paper is thus to review briefly the genesis of the concept of fragility, and the varying nuances that have characterised this word since it became a widely-referred to term in the literature on development studies. Such an effort is crucial, as different understandings of what fragility involves have relevant implications in terms of policy design.

3 How Fragility Made Its Way in the Development Discourse

Notwithstanding its recent appearance in the development discourse, the definition of fragility varies substantially across different institutions. (5) We distinguish between three separate groups of definitions, according to the basis on their focus and breadth. The first two sets of definitions relate fragility to state institutions, and differ with respect to the yardstick which is relied upon to assess fragility.

The first set of definitions takes a strong normative stance, identifying a set of core policy objectives, and interpreting fragility as a mis-alignment between these policy objectives and the ones that are actually pursued by state institutions. Hence, this set of definitions opts for an outcome-based concept of fragility, in which the outcomes can be either the public policies themselves, or their ensuing consequences in terms of service provision to citizens, and of the authority and legitimacy of state institutions.

The second set of definitions has a more limited normative content, and relates fragility to the social contract, whose content is endogenously determined by the interaction between the state and society. Here, the procedural dimension is predominant as far as the assessment of fragility is concerned.

Finally, the third set of definitions moves beyond the focus on state structures alone, as it emphasises that also non-state actors play a crucial role in shaping the institutional structure of a country, and in determining its fragility. A broad procedural dimension is thus crucial even for this set of definition as far as the assessment of fragility is concerned.

3.1 Fragility and the relationship with the donor community

The first set of definitions origins from a debate that is entirely developed outside the developing countries and which was triggered by the shock of the 9/11 attack, on one hand, and by the growing donors' concerns for a group of countries which increasingly appear to be left behind by the prevailing aid allocation systems in the race towards the Millennium Development Goals.

The 1990s and the early 2000s were marked both by a progressive shift away from project-based assistance towards budget support, and by an increasing recognition of the role of the policies adopted by the governments in recipient countries in mediating the impact of aid programmes (Chhotray and Hulme, 2009). The combined effect of these two major changes was an increase in aid selectivity, loosely-defined as an attempt to reward the countries which were regarded as good performers with growing aid flows. The flip-side of the coin of a performance-based allocation mechanism was that some countries became aid orphans, as countries characterised by what was perceived as poor - and not development-oriented - governance recorded sharply declining and volatile aid flows (Levin and Dollar, 2005; Fielding and Mavrotas, 2008). The growing emphasis on aid selectivity, which was reflected in the so-called Monterrey Consensus, and which led to a substantial shift in bilateral aid allocation (Dollar and Levin, 2006), confronted the donors with a hard to solve Samaritan's Dilemma. The incentive mechanism which informed aid selectivity came at the high cost of reducing aid where it was most needed, albeit where it was probably least effective.

Such a policy shift on the donors' part also endangered the progress towards the current core targets of the development efforts, namely, the Millennium Development Goals. This set of goals is to be achieved by the year 2015, a time-frame that is too short to allow the incentive mechanism agreed upon in the Monterrey Consensus to bring non-performing countries back on the track of sound, development-oriented, policies. Hence, a strict adherence to the aid selectivity principle could come at the cost of failing to achieve the targets which were set by the United Nations in September 2000.

Two of the most recent editions of the Global Monitoring Report by the World Bank (2007, 2009), evidenced that countries with ineffective or poorly-functioning institutions are not progressing towards the achievement of the MDGs, or are even moving backwards. These countries have recently begun to be labelled as fragile countries, though - as recalled in the introduction - a rich variety of differently nuanced terms can be found in the literature.

At the same time, the debate on fragile countries has come along with a renewed emphasis on the link between development and security. Countries permeable to criminal activities and networks, with latent sources of political instability, with a pervasive sense of mistrust or grievance towards state institutions, or with state institutions not responding to the Weberian definition of the monopoly on the legitimate use of physical violence have emerged as an international priority since the mid-1990s. The new interest for these countries is motivated (6), not only by the severe violations of human rights suffered by their people, but also by the possible negative spill-overs to security in other countries. After the terrorist attack of 9/11 which dramatically revealed the weaknesses of global security in the post-Cold War world order, the international community became fully aware of the lack of an adequate operational and conceptual framework to guide international interventions and assistance in these countries which were regarded as potential threat to global security (Châtaigner and Ouarzazi, 2007) (7).

Thus, the growing concern about how to deal with fragile countries "appeared as a political response to an operational issue" (Guillaumont and Guillaumont Jeanneney, 2009), namely, the troubles faced by donor agencies in inter-acting with "countries where there is a lack of political commitment and insufficient capacity to develop and implement pro-poor policies" (OECD/DAC, 2006). (8) This definition provided by the OECD Development Assistance Committee (DAC) evidences that the notion of a fragile country is intrinsically a relational one, as it refers to a mis-alignment between the political will - or its capacity - of a country and the universal priorities set out by the donor community.

The definitions adopted by other international organisations and development agencies have stressed that fragile countries represent difficult partners, and a challenging test for aid effectiveness because of the low capacity or development orientation of their state structures. The World Bank, (9) for instance, relies on the Country Policy and Institutional Assessment (CPIA), a rating system used to assess the quality of the policies in a country and its eligibility for IDA funds according to the performance-based allocation (PBA) system. The classification by USAID is also based upon the government's capacity and willingness, the only difference being that the reference to the core state functions is more explicit. The label "fragile" denotes those states which are unable or unwilling to ensure adequately the provision of security, basic and vital services to significant groups of their populations, where the legitimacy of the government is questioned, weak or non-existent, or where the central government does not exert effective control over its own territory, or where violent conflict is taking place or where there is a high risk of outbreaks of violence (USAID, 2005).

Stewart and Brown (2009) propose an outcome-based definition of fragility, which covers "states that are failing, or at risk of failing, with respect to authority, comprehensive service entitlements or legitimacy". These three dimensions are closely intertwined, as "failure on one of the three dimensions [...] generally predisposes to failure on other(s)", and the definition has a normative content in so far since it requires the identification of the set of services that the state should be providing to the population. (10)

Despite their differences, these definitions - either implicitly or explicitly - question the political priorities set by a country, or upon its ability to pursue them. Under this perspective, the use of the expression "fragile country" therefore becomes politically sensitive and highly contentious. Paul Collier (2007), for instance, did not initially disclose the list of countries that form its world-famous Bottom Billion as "this is not a company that countries are keen to be in". The Declaration after the EU-Africa Summit held in Lisbon on 8-9 December 2007 does not contain an explicit reference to fragile countries, although - independently of the various possible definitions and measurements - Sub-Saharan Africa always provides the majority of the countries which are classified under this label. The Declaration, which is meant to lay the foundations for "a new strategic political partnership for the future", only mentions that "Africa and the EU will also hold a dialogue on the concept of 'situations of fragility' aimed at reaching a common understanding and agreeing on steps that could be taken".

The uneasy feeling associated with being labelled as fragile is also clearly connected with the great deal of discretion that any underlying definition involves and with partner countries' fear of stereotypes and stigmatization which can jeopardize their international image, climate investment, economic and development perspectives (Faria and Ferreira, 2007). The negative reaction to the debate on fragility by these countries is also associated with the meaning that this expression has outside the development arena, where it is used - along with the stronger term failed - to designate the countries which could represent a threat to global security. Finally, recipient countries might be concerned about the possible consequences for their sovereignty of international interventions driven by a strong emphasis on the security-development nexus. Some authors, for instance, explicitly call for the adoption of a broader set of tools, including military intervention, to deal with the development challenge that these countries pose (see, for instance, Bourguignon et al., 2008, and Collier, 2009).

3.2 Fragility and the relationship between the state and the society

The reluctance by aid recipient countries to accept the label fragile, because of its intrinsically normative content and political sensitivity, could hinder the diplomatic dialogue with the donor community. This is why, just a few months after the definition that we recalled above, the OECD Development Assistance Committeee began to refer also to situations of fragility, broadening the focus from the state alone (OECD/DAC, 2007), and, in its Principles for Good International Engagement, made the statement that "a durable exit from poverty and insecurity for the world's most fragile states will need to be driven by their own leadership and people". The long-term goal of an engagement in these countries is "to help national reformers to build effective, legitimate, and resilient state institutions, capable of engaging productively with their people to promote sustained development" (OECD/DAC, 2007; emphasis added). Even if this shift might not eliminate a country's perception of being subject to an external discretionary and negative judgement, the reference to national reformers helps to convey the idea that overcoming fragility is a locally-owned process, which should respect people's right to self-determination, and which can only be marginally influenced by external actors.

One related change in the definition is a shift away from a definition of fragility that - albeit implicitly - focuses on the relationship between a country and the donor community, to definitions turning inwards to domestic dynamics. The Council of the European Union (2007) acknowledged that: "[...] fragility refers to weak or failing structures and to situations where the social contract is broken".

The reference to the social contract, which is the outcome of a never-ending bargaining process between society and the state, substantially broadens the relevance of the pervasive effects of fragility. A similar approach is pursued by OECD DAC, (2008), which proposes:

"modifying the OECD DAC definition of a fragile state, simply as one unable to meet its population's expectations or manage changes in expectations and capacity through the political process."

Such a definition does not take a stance with respect to the actual content of the expectations held by the population, and it introduces an interesting dynamic aspect in the definition, as fragility can be regarded as the inability to manage the perturbations that can affect changes in the expectations, or in the capacity of the state to meet them. Such a focus is in line with the argument by Baliamoune-Lutz and McGillivray (2008), who argue that the term "fragile state" should be used just to "delineate states only in terms of their likelihood of breaking-up or vulnerability to downside shocks", which is precisely what OECD DAC, (2008) does. The break-up of the equilibrium which occurs whenever the mis-alignment between the ability of the state to meet the expectations of the population creates an element of latent instability, which may lead to a conflict, which, in turn, represents the ultimate manifestation of fragility.

This perspective - which conceives state fragility as the risk that state institutions may lose their legitimacy and effectiveness - is close to the definition proposed by Ikpe (2007), who relates fragility to "the capacity of the state to adapt to changed circumstances, protect citizens, absorb shocks and manage conflict without resort to violence". Moving to the economic sphere, this approach is also close to the one described by Guillaumont and Guillaumont Jeanneney (2009), who portray fragility as the inability of a state to implement appropriate coping policies in the face of adverse shocks.

Following this line of argument, what is fragile, namely, what "can be easily damaged or broken", as the Oxford English Dictionary reads, is the peaceful, though dialectic, inter-action between society, population groups and the state. When this dynamic system is fragile, it is exposed to a high risk of breakdown even in the face of small shocks. As specified in the definition proposed by the Crisis State Research Centre (2006), the opposite of a "fragile state" is a state "where dominant or statutory institutional arrangements appear able to withstand internal and external shocks and contestation remains within the boundaries of reigning institutional arrangements". State fragility still pivots on state functioning, but is not evaluated in relation to external or international development goals, but in relation to state ability to mediate, to settle, to meet - either converging or conflicting - expectations and claims expressed at different levels of the social structures (families, clans, political parties, multinational and domestic enterprises) regardless of the content of these expectations. These are regarded as being endogenously determined, even though the donor and development community has tried to identify when states are more at risk of failing to fulfil the expectations of its citizens. Thus, the Council of the European Union (2007) suggests that the social contract is most likely to be broken when the state is incapable or unwilling:

"to deal with its basic functions, meet its obligations and responsibilities regarding the rule of law, protection of human rights and fundamental freedoms, security and safety of its population, poverty reduction, service delivery, the transparent and equitable management of resources and access to power." (Council of the European Union, 2007)

The Crisis State Research Centre (2006) attempts to detect some typologies of state institutional arrangements that can embody and reproduce conditions of crisis in the economic, social and political domains, such as property rights which re-inforce stagnation or low rates of economic growth, or imply extreme inequalities. Kaplan (2008) pushes this argument further by defining fragile countries upon the basis of some structural features in the design of the state institutions. The author observes that the arbitrariness of the borders drawn in Africa at the time of the colonial rule forced "multiple identity groups with little common history to work together" with the result that the current legal and political systems in Sub-Saharan African countries do not reflect the values, beliefs, and ways of social organisation recognised by the local populations.

3.3 Fragility and institutions beyond state structures

According to the definitions which refer to the social contract and to the expectations held by the population, both society groups and individuals are meant to express their needs, wishes, and also to bring their contestations and to exert checks and balances towards state institutions, which, in turn, are called to respond to their requests. Therefore, this perspective implicitly entails a passive view of citizens, even though they are acknowledged to be last holders of the right to self-governance. Some authors have moved to definitions which attempt to account for the active role of non-state institutions and social groups in shaping a country's self-governance system. An interesting definition, for instance, is offered by Engberg-Pedersen et al., (2008), who move beyond the exclusive focus on the state, which characterises the definitions that we have reviewed so far, and define a situation of fragility as "institutional instability undermining the predictability, transparency and accountability of public decision processes and the provision of security and social services to the population". Though OECD DAC (2007) also refers to situations of fragility, it states that donors should "focus on state-building as the central objective". Conversely, Engberg-Pedersen et al. (2008) argue that, "in fragile situations where the state is absent or very weak, non-state authorities often perform state-like functions with respect to the provision of security and social services". (11)

This interpretation is based upon the recognition that multiple authorities perceived as legitimate and multiple sources of service provision can exist at a level above the state in a given territory. While this is not a cause of instability per se, when the inter-actions, interests, and governance forms of these non-state groups and institutions are irreconcilable each other, or beyond the control of the state and in opposition to it, the existing formal and informal institutional arrangements are in a situation of fragility, namely, they run the risk of failing to meet the different sets of expectations and of bringing them into equilibrium.

4 Fragile - or Fragility?

The differences among the alternative sets of definitions could - at first sight - be regarded as terminological disputes whose relevance is solely academic, but they do actually have a salient content also as far as policy design is concerned. The various definitions of fragility hide a further - and possibly more profound - divergence: Is it possible draw a line to separate fragile from non-fragile countries upon the basis of a relevant discontinuity in the factors that are thought to identify a situation of fragility? The answer to such a question is crucial for policy design: the external engagement towards these countries can be reflective of specific priorities or modality of interventions only if fragile countries can be credibly singled out. Conversely, even if such a distinction was not grounded solidly, fragility would still be relevant for policy design, as it would emphasise the central role of local institutions in mediating the effects of external engagement towards any country.

The definitions pertaining to the first set are more easily susceptible to a quantitative measurement of fragility, and provide an affirmative answer to the question above, by setting a - either relative or absolute - threshold that identifies fragile countries. But such an option is not uncontested in the literature, with arguments that are unrelated to the - admittedly controversial in itself - challenge of operationalising a definition of fragility into some form of measurement (see DIE and UNDP, 2009).

The objection is theoretical, and is based upon the argument that "[...] a rigorous distinction between fragile and non-fragile states seems unsustainable" (Engberg-Pedersen et al., 2008), as there is a spectrum of fragility, which "is found in all but the most developed and institutionalised states" (OECD/DAC, 2008). Such a criticism is not limited to the definitions which focus on the procedural dimensions: Baliamoune-Lutz and McGillivray (2008) argue that "if we associate fragility to aid effectiveness, all countries are fragile to the extent that their ability to use aid differs". In this latter respect, the identification of the existing best practices in aid policies to fragile states by Manor (2007) revealed that:

"we expected that the approaches widely-used by donors in countries that are not fragile states would be inappropriate in the extremely difficult environments examined here. That turned out to be true, but only to a limited extent. Many well-accepted principles and strategies have proved their worth in these countries, although they often need some adjustment." (Manor, 2007)

Given the current low-rating of the one size fits all approaches to the design of development policies, there is little surprise at the fact that interventions in fragile states "need some adjustment", while what is remarkable here is that well-established patterns of interventions have been found to work well even in countries that are portrayed as the toughest development challenges. Though this argument cannot go all the way to claiming that fragile countries do not pose specific challenges to the engagement of donors, what needs to be emphasised is that this evidence gives rise to a further question: namely, if well-functioning policies need to be context-specific, what do we gain from including some countries under the broadly-defined group of fragile countries?

Several authors argued that the ensuing gain is, indeed, limited. Briscoe (2008) wonders "whether an effective, targeted policy can be derived from an all-encompassing label", as "there is little to hold state fragility together other than its symptoms: poverty, insecurity, proneness to conflict, corruption". Chesterman et al., (2004), quoted in Faria and Magalhães Ferreira (2007), extended to the case in question the initial lines of Anna Karenina by Leo Tolstoy: "all happy families are happy alike; while every unhappy family is unhappy in its own way". Fragility is a heterogeneous and multi-faceted phenomenon, no differently from familial unhappiness. The empirical analysis that Stewart and Brown (2009) adopt reveals that most of the countries fail along just one of the three dimensions of state fragility, and none of them fails in all the dimensions, so that every country is fragile in its own way.

5 Theoretical and Policy Implications of the Different Definitions of Fragility

The debate around fragility has, to date, provided a valuable contribution in highlighting the side-effects of performance-based aid allocation systems and of the poverty-efficient allocation paradigm which are adopted by the donor community. These can also lead to the paradox that development assistance channels less aid to countries which need it most, though they are possibly the least able to use it effectively. The discourse on fragile countries, therefore, has had an important advocacy role. Econometric estimates by Feeny and McGillivray (2009), in particular, find that many fragile countries, identified on the basis of the CPIA score, were under-aided also relative to their capacities to convert aid efficiently into economic growth. The widespread bad neighbour effects due to fragility (Chauvet and Collier, 2004) reinforce the case for increasing aid flows to these countries, in as much as it contributes to increase the chance for a sustained institutional turnaround (Chauvet and Collier, 2008).

Moreover, the literature on fragility has brought back the political dimension to the attention of debate on development and aid policy design. While it is widely recognised that institutions and governance matter to economic development and aid effectiveness, the debate on fragility has stressed that the functioning of the state and, more generally, the functioning of institutions need to be considered in the definition of aid and development assistance modalities and tools.

However, the contribution of the discourse on fragility to the design of development policies could be substantially expanded if the fuzziness around the definition of fragility itself were - at least partly - clarified. This is so because, at present, alternative definitions of fragility have considerably different - and, at times, contrasting - implications when it comes to selecting both the priorities and adequate instruments for intervention.

The first set of definitions emphasises the distinctive traits of a specific group of countries, and it thus supports the case for tailor-made delivery systems, which reflect the specific challenges that fragile countries pose. Supporting their progress towards the MDGs necessarily requires taking the institutional context of the recipient countries as given, because of the achievement of the goals is set for 2015. Pursuing this objective might be in conflict with the long-term goal of strengthening state institutions (OECD/DAC, 2007), as it could call for a channelling of funds outside state institutions, given their current limited capacity or development-orientation. Hence, aid effectiveness could be improved through tailor-made modalities which need not be consistent with state-building. Aid modalities, such as budget-support, which could contribute to the strengthening of state institutions, face the risk of a limited effectiveness - as far as major development outcomes are concerned - in the short run.

The label of fragile country would thus signal the high priority for the donor community to improve aid effectiveness towards the countries which are defined as such, but it could still be of limited use in pursuing such an improvement. This is so because of the high heterogeneity of fragile countries which was recalled above, which creates the need to adapt modalities of intervention to context-specific priorities, institutional settings and political settlements. This, in turn, indirectly cast doubts on the operational relevance of the concept of the fragility. (12)

The second set of definitions identifies state-building as the central objective of external engagement towards fragile countries. The emphasis on the procedural dimensions suggests that the efforts to promote the strengthening of state institutions should not be limited only to what concerns the capacity of state institutions, but also to their perceived legitimacy. Restoring or creating capacity is per se insufficient to overcome the fragility which emerges from the relationship between the state capacity and the expectations held by the various social groups. Externally-driven improvements in development outcomes which are not perceived as being the result of state interventions, would produce little to no effect on the legitimacy of state institutions, which should be pursued even at the cost of a lower effectiveness.

The third set of definitions does not actually identify a distinct set of priorities, but instead signals which are the relevant local factors that should be carefully considered by external actors. Engberg-Pedersen et al. (2008) argue that a narrow focus on the strengthening of the capacity and legitimacy of the state to abide by the social contract could be ill-founded, due to the crucial role played by non-state actors in situations of fragility.

Along the same lines, González Aimé (2008) criticises the view of societies as "passive victims rather than political actors", in what she considers the prevailing conceptual and analytical framework applied in the design of international development assistance towards fragile countries. Specifically, "the international response to these states - which are perceived as anomalous spaces - has neglected the possibility that they may be political spaces, in the process of re-organising themselves" (González Aimé, 2008). Indeed, anecdotic evidence suggests that there may be well-functioning institutions which external actors should not ignore once they engage in the country, even in what is a clear case of state collapse. (13) The role of local political actors, which could act either as spoilers or as drivers of change, need to be carefully analysed and understood, in order to understand what the scope of external engagement is, and what its chances of success are.

The definitions of the third group - and some of the second one as well (14) - also imply that the notion of fragility should be used as a dimension, rather than as a categorisation or as a criterion for classification, since it does not allow for the identification of a separate group of countries which are labelled as fragile. Under this perspective, "fragility" could be assessed with reference to all political and socio-economic systems, and this concept could provide an analytical framework, instead of identifying a group of countries requiring specific tools of interventions, and with specific priorities. This approach could be also consistent with the need for development assistance to be based upon conceptualisation and theoretical constructs accepted by all counterparts, and which do not create a sense of stigmatisation in partner stakeholders.

6 Concluding Remarks

This brief overview of the concept of fragility has allowed us to shed some light on the theoretical and policy implications of this notion, on its potential and actual contributions, as well as on its existing weaknesses and ambiguities. In the light of this critical review, we argue that the debate on fragility has led to two main conclusions which deserve further research. On the one hand, the priorities, modalities and tools for aid and development assistance should be informed by the understanding of state functioning and of the sense of trust and legitimacy attributed to institutions. On the other hand, external interventions should be rooted in baseline assessments of - either state or non-state - basic service providers, power-selection mechanisms, public decision-making processes, and existing authorities, acknowledged as legitimate by the populations. This approach is, first and foremost, meant to complement and mutually reinforce humanitarian principles. Though citizenships' expectations are culturally and country's specific, the need to ensure the provision of the basic needs and human rights of the population remains the guiding principle of development assistance. At the same time, the capacity of existing institutions to meet the basic needs of the population and respect its rights constitutes the condition and the primary source of their legitimacy.

The need for a deeply-rooted knowledge of the political and institutional dimensions of any context in which external actors engage themselves is a relevant methodological contribution of the third set of definitions of fragility. This requires accurately respecting and operationalising the first of the OECD DAC principles, which calls for "taking context as the starting point" (OECD/DAC, 2007). If external actors fail to take account of the inherently political character of interventions in situations of fragility, then their engagement may not only be ineffective, but could also even be counter-productive. The road to be travelled is a long and difficult one, as the operationalisation of this principle requires analytical instruments which have not yet been refined, and that often do not pertain to the toolbox of the actors engaged on the ground. Nevertheless, a move in this direction is necessary to bring the tough challenge posed by fragility within reach.

7 References

Baliamoune-Lutz, M., McGillivray, M., 2008, State Fragility - Concept and Measurement, Research Paper No. 2008/44, Helsinki, World Institute for Development Economics Research.
Bourguignon, F., Bénassy-Quéré, A., Dercon, S., Estache, A., Gunning, J. W., Kanbur, R., Klasen, S., Maxwell, S., Platteau, J.-P., Spadaro, A., 2008, Millennium Development Goals at Midpoint: Where Do We Stand and Where Do We Need to Go?, background paper for the European Report on Development 2009.
Briscoe, I., 2008, The EU response to fragile states, European Security Review 42.
Cammack, D., McLeod, D., Rocha Menocal, A., Christiansen, K., 2006, Donors and the 'Fragile Sates' Agenda: A Survey of Current Thinking and Practice, Report submitted to the Japan International Cooperation Agency, London, Overseas Development Institute.
Carment, D., Yiagadeesen, S., Prest, S., 2008, State Fragility and Implications for Aid Allocation: An Empirical Analysis, Conflict Management and Peace Science 25, 349-373.
Chandler, D., 2006, Empire in Denial: The Politics of State-Building, London: Pluto Press.
Châtaigner, J.M, Ouarzazi, L., 2007, Fragile states and the new international disorder, FRIDE Comment, Madrid: Fundación para las Relaciones Internacionales y el Diálogo Exterior.
Chauvet, L., Collier, P., 2008, What are the Preconditions for Turnarounds in Failing States?, Conflict Management and Peace Science 25 (4), 332-348.
Chauvet, L., Collier, P., 2004, Development Effectiveness in Fragile States: Spillovers and Turnarounds, Center for the Study of African Economies, Department of Economics, Oxford University.
Chesterman, S., Ignatieff, M., Thakur, R., 2004, Making States Work: From state failure to state-building, New York, International Peace Academy/ United Nations University,
Chhotray, V., Hulme, D., 2009, Contrasting Visions for Aid and Governance in the 21st Century: The White House Millennium Challenge Account and DFID's Drivers of Change, World Development 37 (1), 36-49.
Collier, P., 2009, Wars, Guns, and Votes, New York: Harpers Collins Publishers.
Collier, P., 2007, The Bottom Billion, Oxford: Oxford University Press.
Council of the European Union, 2007, Council Conclusions on a EU response to situations of fragility, 2831st External Relations Council meeting, Brussels, 19-20 November 2007.
DIE, UNDP, 2009, Users' Guide on Measuring Fragility, Bonn: German Development Institute, and Oslo: United Nations Development Programme, Oslo Governance Centre.
Dollar, D., Levin, V., 2006, The Increasing Selectivity of Foreign Aid, 1984-2003, World Development 34 (12), 2034-2046.
Engberg-Pedersen, L., Andersen, L., Stepputat, F., 2008, Fragile Situations - Current Debates and Central Dilemmas, Report No 2008:9, Copenhagen: Danish Institute for International Studies
Faria, F., Magalhães Ferreira, P., 2007, Situation of Fragility: Challenges for an European Response Strategy, Maastricht: European Centre for Development Policy Management.
Feeny, S., McGillivray, M., 2009, Aid allocation to fragile states: Absorptive capacity constraints, Journal of International Development 21 (5), 618-632.
Fielding, D., Mavrotas, G., 2008, Aid Volatility and Donor-Recipient Characteristics in 'Difficult Partnership Countries'", Economica 75, 481-494.
González Aimé, E., 2008, Fragile states and neoliberalism in Sub-Saharan Africa, mimeo, Madrid: Fundación para las Relaciones Internacionales y el Diálogo Exterior.
Guillaumont, P., 2009, Caught in the Trap. Identifying the Least Developed Countries, Paris: Economica ,
Guillaumont, P., Guillaumont Jeanneney, S., 2009, State fragility and economic vulnerability: what is measured and why?, background paper for the European Report on Development 2009.
Hehir, A., 2007, The Myth of the Failed State and the War on Terror: A Challenge to the Conventional Wisdom, Journal of Intervention and State Building 1 (3), 307-332.
Ikpe, E., 2007, Challenging the discourse on fragile states, Conflict Security & Development 7 (1), 85-124.
Kaplan, S., 2008, Fixing Fragile States: A New Paradigm for Development, Praeger.
Leeson, P.T., 2007, Better off stateless: Somalia before and after government collapse, Journal of Comparative Economics 35, 689-710.
Levin, V., Dollar, D., 2005, The Forgotten States: Aid Volumes and Volatility in Difficult Partnership Countries (1992-2002), Summary paper prepared for the DAC Learning and Advvisory Process in Difficult Partnership, mimeo.
Manor, J., 2007, Aid That Works, Washington, the World Bank.
McGillivray, M., 2006, Aid Allocation and Fragile States, Research Paper No. 2006/1, Helsinki, World Institute for Development Economics Research.
Newman, E., 2007, Weak States, State Failure, and Terrorism, Terrorism and Political Violence 19 (4), 463-88.
OECD/DAC, 2008, Concepts and Dilemmas of State Building in Fragile Situations - From Fragility to Resilience, Paris.
OECD/DAC, 2007, Principles for Good International Engagement in Fragile States and Situations, Paris.
OECD/DAC, 2006, Fragile States: Policy Commitment and Principles for Good International Engagement in Fragile States and Situations, DAC Senior Level Meeting, December 5-6.
Patrick, S., 2007, "Failed" States and Global Security: Empirical Questions and Policy Dilemmas, International Studies Review, 9: 644-62.
Rosenstein-Rodan, P.N., 1943, Problems of Industrialisation of Eastern and South-Eastern Europe, The Economic Journal, 53: 202-211.
Rostow, W.W., 1959, The Stages of Economic Growth, The Economic History Review, 12: 1-16.
Scheye, E., 2009, State-Provided Service, Contracting Out, and Non-State Networks: Justice and Security as Public and Private Goods and Services, Discussion Paper, INCAF/PDG, Paris, OECD.
Stewart, F., Brown, G., 2009, Fragile States, Working Paper No. 51, Oxford, Centre for Research on Inequality, Human Security and Ethnicity.
UNDP, 2001, Human Development Report 2001 - Somalia, New York. USAID, 2005, Fragile States Strategy, U.S. Agency for International Development, Washington D.C.
World Bank, 2009, Global Monitoring Report 2009 - A Development Emergency, Washington D.C.
World Bank, 2007, Global Monitoring Report 2007 - Confronting the Challenges of Gender Equality and Fragile States, Washington D.C.
World Bank, 2005, Low-Income Countries Under Stress: Update, Operations Policy and Country Services, Washington D.C.
Zoellick, R.B., 2008, Fragile States: Securing Development, Survival 50 (6), 67-84.

Notes

*. The authors are grateful to Luca Alinovi, Stefano Bartolini, Nicolò Bellanca, François Bourguignon, Sara Turchetti and Franco Volpi for their comments on earlier drafts of this paper, which was prepared for the European Report on Development 2009, Overcoming Fragility in Africa - Forging a New European Approach; the usual disclaimers apply.

**. Institute for Employment Research, Nuremberg, sb3151@gmail.com.

***. Robert Schuman Centre for Advanced Studies, EUI, elisa.ticci@eui.eu.

1. In this paper, the use of either of the two expressions does not reflect its endorsement; instead, it reflects the stances taken by the various definitions to which we refer.

2. Zoellick (2008) recently added, to this - rather long - list, the expression "broken states", but "in most cases, these labels do not have a meaning that is clearly understood far beyond the author who has used them", as Cammack et al., (2006), argue.

3. The list of 58 countries was disclosed by the author in his recent book Wars, Guns and Votes - Democracy in Dangerous Places.

4. For instance, landlockedness is one of the four criteria that Collier (2007) refers to when identifying and characterising countries which are classified in the Bottom Billion.

5. The overview of definitions of fragility that we provide in this paper is deliberately partial, as this is meant to convey some basic features of the debate around this concept; see DIE and UNDP (2009) for a detailed and exhaustive list of definitions.

6. The centrality of security concerns in the debate on fragile countries is mirrored by distribution of aid flows: in 2007, the two main conflict-settings in the scenario of world politics - Afghanistan and Iraq - and two big players in highly unstable regions - Pakistan and Ethiopia - absorbed half of the aid directed to the countries classified as fragile by OECD (OECD, 2009).

7. It worth underlining that, however, the link between state fragility and trans-national threats does not reflect a shared consensus in the literature; on this topic, see Chandler (2006), Hehir (2007), Newman (2007) and Patrick (2007).

8. "When it appeared clear that countries facing particularly difficult situations could not receive much from the PBA, although they could need more, the answer was then to give them a specific treatment for effectiveness or security reasons. Fragile states were first identified, by one way or another, as countries where the PBA should not apply." (Guillaumont and Guillaumont Jeanneney, 2009).

9. The World Bank definition is available online.

10. "No matter what the claimed purpose is, the practice of measurement [of fragility] will always contain a normative dimension, and this foundation of values often stems from the producers' interest." (DIE and UNDP, 2009).

11. Many examples of non-state systems of provision of security, justice and basic social services in Afghanistan, Burundi, Rwanda, Sierra Leone, Somalia, Sudan, Timor Leste fragile are illustrated by Scheye (2009).

12. A telling example of the limited operational relevance is reflected in the fact that many contributions on fragility introduce further taxonomies of fragile countries or situations of fragility once it comes to choosing from among the alternative strategies of interventions; see, for instance, the discussion about budget support in OECD/DAC (2008).

13. For instance, the provision of law and order in Somalia, the text-book case of state fragility, is ensured by the so-called Islamic courts, and "shari'a courts perform an instrumental function in creating legal order [...] under anarchy, dispute resolution is free and speedy by international standards", as Leeson (2007) observes. The author argues that, though the current condition is far from being ideal, it has, nevertheless, improved, compared to the pre-1991 situation, and this is not limited to the judicial system. The Human Development Report 2001 - Somalia observes that, there were more primary schools in the country in the early 2000s than in the late 1980s, and the private sector has been effective also in the provision of water and electricity (UNDP, 2001).

14. See, for instance, Ikpe (2007), and Baliamoune-Lutz and McGillivray (2008).