Context and caution

Development Policy26 Jan 2011Lars Engberg-Pedersen

What do countries like Zimbabwe, Liberia and Afghanistan have in common? The answer is that they are all considered ‘fragile states’. Whether they like it or not, they are said to suffer from poor state capacity or the lack of political will to carry out effective policies. What purpose does such an ahistorical and generalizing categorization serve? The answers given in Aidland are instrumental, normative and pragmatic.

The term ‘fragile states’ has become very widely used in recent years by development agencies such as the UK Department for International Development (DFID) and the German Federal Ministry for Economic Cooperation and Development (BMZ), and in the European Commission’s (EC) development policy. Though donors emphasize the need to take context as the starting point, in practice, they treat fragile states as a clearly identifiable, separate category of countries requiring a particular development approach unlike the one used in non-fragile states. Donor agencies end up creating a crude distinction between countries with and without particular problems – the picture is black and white, with no shades of grey.

The security and development sectors’ views on how to engage in fragile states have converged since the end of the Cold War. A radically changed world made both sectors realize that they have overlapping agendas and in fact depend on each other. The focus has swung decidedly to the state itself and all that comes with it: institutional stability, effective service provision, and social, political and economic equality. This has come to be referred to as the ‘security-development nexus’.

For the last twenty years, the state has been seen increasingly as the central force for achieving progress among those working in the fields of development and poverty reduction. Indeed, donors have devoted more attention to the state, which may be linked to the decline of the structural adjustment approach, with its focus on liberalization, privatization and deregulation.

Actors providing humanitarian relief in the contexts of natural disasters and civil wars also see the state as a major concern. The end of the Cold War meant that superpowers no longer saw a political need to uphold regimes across the globe. On the other hand, the international community became increasingly concerned about the effects of conflict on human security in countries such as Rwanda and Somalia.

The global security sector has also increasingly directed its attention to the state, arguing that societies with weak states are more likely to harbour groups that constitute a threat to these societies as well as to neighbouring and more distant countries. Since the War on Terror was declared, several security strategies have identified fragile states as constituting significant threats and possible homes for international criminal networks.

Caution and context

This raises the question of how governments and development agencies should engage in fragile states. One of the most authoritative statements reflecting the evolving discourse on aid and fragile states is the 2007 publication by the OECD DAC, Principles for good international engagement in fragile states and situations. This four-page document strongly advocates taking ‘context as the starting point’ for international engagement in fragile states and avoiding ‘blueprint approaches’. Its ten principles call for political analyses to go ‘beyond quantitative indicators of conflict, governance and institutional strength’.

The document also warns of the unintended consequences of aid interventions, which may deepen social cleavages and fuel latent conflicts. One reading of the DAC principles therefore suggests that context and caution are key elements in the development discourse on how the international community should engage fragile states.

The only really tangible principle of the ten – and therefore probably the one that is likely to have the biggest impact on donor practices – is the proposal to consider ‘state-building as the central objective’. Moreover, this principle tentatively defines a fragile state as one that lacks the ‘political will [or] capacity to provide the basic functions for poverty reduction’ and ‘development’. Nor can a fragile state ‘safeguard the security and human rights of their populations’.

Everything converges in these two phrases: What fragility is, why it is a problem and how it should be addressed. Such a condensed understanding of fragile states and fragile situations puts into perspective the emphasis on context as a starting point.

We are all fragile in some respects

These points become clearer if we take a brief look at how different donors define and address fragile states. In its 2007 strategy document, Development-oriented transformation in conditions of fragile statehood and poor government performance, the German Federal Ministry for Economic Cooperation and Development (BMZ) distinguishes between three kinds of situations in partner countries:

  • ‘Situations where government performance is development-oriented and shows a positive trend’
  • ‘Situations where government development orientation is low’
  • ‘Situations of continuously deteriorating government performance, with no development orientation’

This categorization is not based on a dichotomy between fragile and non-fragile states. Indeed the document does not refer much to the term fragile state. Yet, it seems to assume that development is linear and homogeneous.

The European Commission (EC) does attempt to define fragile states. In a 2007 issues paper entitled EU response to situations of fragility in developing countries, the EC argues that fragility refers to ‘weak or failing structures and institutions but also to situations where the social contract is broken due to [a] state’s incapacity [or] unwillingness to deal with its basic functions and meet its obligations in terms of public service delivery and its responsibilities regarding the rule of law’.

The issues paper rejects the notion of fragile states, however, because of the stigma attached to it. They also fear that it could potentially reflect ‘a new form of conditionality, which may involve more or less formal “sanctions [and the] suspension of cooperation” or have consequences in terms of private sector readiness to invest’.

The paper argues that the instability that comes with fragility is ‘rooted in a mix of political, institutional, economic, social, cultural [and] environmental factors’. The paper is less clear about what is needed to solve situations of fragility. However, in another 2007 paper, Towards an EU response to situations of fragility, the EC does clearly state that every ‘case requires a differentiated, articulated and holistic response, articulating diplomatic action, humanitarian aid, development cooperation and security’.

If every case is different, can the idea of a fragile state even be defined? DFID questions the very term in a 2005 paper on fragile states entitled Why we need to work more effectively in fragile states. The organization argues that all ‘states are fragile in some respects and states move in and out of fragility. People also disagree about what constitutes fragility and no state likes to be labelled as fragile by the international community’.

Yet, the paper subsequently states that ‘DFID’s working definition of fragile states covers those where the government cannot or will not deliver core functions to the majority of its people, including the poor’. This illustrates the dilemma of defining or even identifying fragile states. On the one hand, the notion is slippery and covers very diverse situations, but on the other hand there are some indicators, such as dysfunctional governance, that point to state fragility. As a result, DFID – along with DAC and the EU – uses state capacity and political willingness as the criteria for creating a separate category of fragile states for certain countries.

The DFID paper is more cautious, however, about the approach that should be adopted to engage fragile states. The paper suggests a ‘drivers of change’ approach, which entails understanding ‘the history of a country and its people, who holds power and how it is brokered and used, the informal “rules of the game”‘. Nevertheless, certain governance reforms, including reforms in the security sector, improving public financial management and strengthening service delivery, are considered particularly important.

Indices of fragility

Aid agencies therefore appear to be in a bind. They use the term ‘fragile states’ but are uncomfortably aware of the need to acknowledge a marked difference in states’ situational contexts and histories. In practice, however, fragile states are treated by donors as a clearly identifiable, separate category of countries requiring a particular development approach unlike the one used in non-fragile states.

A qualitative difference between fragile and non-fragile states emerges when the former is described as troubled by conflict and having weak state capacity, whereas the latter, by implication, are assumed not to suffer from these difficulties. This suggests that rapid, preventive and mitigating measures, together with the building of basic state structures, will go a long way towards helping fragile states. The implication, though, is that non-fragile states do not need this kind of support.

It is perhaps not surprising that donor agencies end up creating a rather crude distinction between countries with and without these problems. It is, however, a far from logical distinction. Many donors rely on the World Bank’s Country Policy and Institutional Assessment (CPIA) index to identify fragile states. Formerly labelled Low-Income Countries Under Stress, countries with ratings of 3.2 or below on the CPIA index have been described b y the World Bank as fragile states since 2005.

The World Bank identified 34 fragile states in 2007 using this method. DFID labels the countries in the two lowest quintiles on the CPIA index as fragile. In 2005, 46 fragile states were identified. The German BMZ strategy paper estimates the number of countries with fragile statehood at more than 50, based on ratings from three different indexes (the CPIA index, the Failed States Index and the World Bank’s Governance Matters indicators).

The inconsistent ways of identifying fragile states should make donor agencies much more attentive to the relative nature of fragility. The arbitrariness of the cut-off points on the CPIA index also seriously calls into question the qualitative difference between fragile and non-fragile states.

Different degrees of failure

Moreover, it should be noted that states may be strong in some respects and weak in others. In a recent study entitled Fragile states, Frances Stewart and Graham Brown, from the Centre for Research on Inequality, Human Security and Ethnicity in Oxford, distinguish between three types of failure at country level.

  • Authority failure, where states lack the authority to protect their citizens from various kinds of violence
  • Service failure, where states fail to ensure access to basic services to all citizens
  • Legitimacy failure, where there are no accountability mechanisms between the state and the population

The study concludes, despite the many methodological challenges, that no country fails on all three dimensions, while eight fail on two and 43 on one. Moreover, the correlations between the three types of failure are either weak or absent. All this indicates that fragile states are rarely fragile in all respects and typically possess some important levels of capacity.

Furthermore, fragility may be quite significant in some parts of a country and not in others. Fragility may exist where rebel movements dominate or where the state is absent, without the state as such being fragile. In many countries, the presence of the state is much more felt in the capital and other major cities, while remote areas experience substantial service failures.

The creation of a separate group of fragile states is a misreading of development problems. As a result, fragility in supposedly non-fragile countries is overlooked. Similarly, so-called fragile states risk being the object of similar approaches focusing on state building, although this may not be the main priority from a poverty reduction perspective.

The development discourse on fragile states takes great pains to emphasize the need for cautious approaches that do not create unintended negative consequences and that adapt activities to the context and to the level of capacity existing in fragile states. But it seems that donors want the whole panoply of internationally sanctioned development objectives to be taken at least as seriously in fragile states as elsewhere.

Indeed, the Council of the European Union, in its 2007 Conclusions on a EU response to situations of fragility, says that its approach to situations of fragility ‘highlights the importance of democratic governance, rule of law, respect of human rights and fundamental freedoms, and the eradication of poverty, as well as of climate change and migration issues.’ Is this what the Council calls adapting to local contexts and limited state capacity?

And what say you?

The obvious question, then, is why the term fragile states is so prominent in development discourse. In an attempt to better understand Aidland, I interviewed three officials from two donor agencies – one bilateral and the other multilateral, all of whom work or have worked in the field of fragile states and are posted at the headquarters of their respective organizations – to get their opinion on the matter.

The interviewees appeared to be concerned primarily about some of the latest ideas circulating, which may bolster their attempts to elaborate and implement development policies. In other words, they were less interested in whether the discourse on fragile states is logically consistent, and more concerned about whether it provides them with tools they can use in their work.

Their responses are characterized by three concerns. The first concern is instrumental. One respondent said the advantage of having a category like fragile states is that it ‘spells things out for us. It would be nice if we could use specific tools, like state building, in specific countries’. Another respondent argued that there is ‘a need for a category like fragile states to reject the comprehensive development framework [approaching diverse countries in a uniform manner with an emphasis on good governance, democracy and sound macro-economic policies, for example] and the emphasis on good performers’.

In other words, the main concern is not whether the notion of fragile states provides insight into the development of specific countries, but whether it helps to identify and legitimize particular actions, including building a case against a particular development approach. So the notion of fragile states is not expected to shed light on particular countries, rather it should help explain particular interventions.

Moreover, the term helps to establish a focus on the state and state capacity. ‘State-building is manageable,’ one respondent said. ‘It does not have to take so long, and the good effects of a stronger state will probably trickle down into society. Moreover, there is the issue of “value export”. You know – democratization and so forth.’ Another said, ‘It is difficult to work with non-state actors, and we don’t have a tradition doing it.’

The second concern is a normative one. One respondent said, ‘I cannot understand this argument that you typically hear from academics that development is not normative. A donor agency evidently has to pursue particular objectives.’

Another interviewee said that it is not right to ‘adopt a completely relativistic approach’. It seems that to these respondents political willingness is a way of legitimizing support to ‘good’ governments and the sanctioning of ‘bad’ governments. However, to one respondent the issue of willingness was less important. He argued that governments in non-fragile states are just as unwilling to implement the ‘right’ policies, so the lack of political willingness is more a condition for development cooperation than a criterion for selecting partners.

Be pragmatic

The final concern is the need for pragmatism. One respondent said, ‘We don’t spend much time classifying states. We adopt a pragmatic approach. The conflict issue is important in identifying fragile states, as well as the issues of governance and human rights. Countries potentially on the agenda of the Security Council and countries that have already been there are fragile states to us. Our focus is on integrated approaches in specific countries. We don’t think that you can get much further with the conceptual questions regarding fragile states.’

The need to be pragmatic in terms of setting priorities was raised by another respondent. ‘Particularly in relation to fragile states, there may be contradictions at the policy level as there are so many different needs here. Thus, fixing a set of priorities at the level of implementation is crucial.’ In such cases, pragmatism is needed to solve existing policy problems.

Instrumental concerns, normative ideas and pragmatism do not fully explain the discourse on fragile states, but they give some indication of how donor officials make sense of the discourse.

It appears that foreign policy and security concerns have been very instrumental in stimulating the emergence of the discourse on fragile states, while development concerns have had some influence on its elaboration. The discourse has suffered a fate during this process that is all too common in Aidland: the adoption of a crude and misleading categorization.

Despite years of recognition that aid should be adapted to the particular conditions of individual countries, donor objectives are generally uniform across recipient countries. The case of fragile states is no exception.


Unfortunately, due to the age of this contribution and several migrations to online content management systems, the footnotes in the text may have been lost. The footnotes below are listed in its original order of appearance in text.
  1. OECD DAC (2007) Principles for Good International Engagement in Fragile States and Situations. OECD DAC High Level Meeting, April 2007.
  2. It has become increasingly popular to refer to fragile situations in international discussions of fragile states partly to capture fragility in particular geographical areas within a country or of a temporary nature in otherwise well-performing countries, partly to avoid stigmatising whole countries as fragile (see Engberg-Pedersen, L., Andersen, L. and Stepputat, F. (2008) Fragile Situations: Current Debates and Central Dilemmas. DIIS Report 2008:9. Danish Institute for International Studies, pp. 24-26).
  3. Commission of the European Communities (2007) EU Response to Situations of Fragility in Developing Countries – Engaging in Difficult Environments for Long-term Development. Issues Paper, July.
  4. Commission of the European Communities (2007) Towards an EU Response to Situations of Fragility. Issues Paper 643, 25 October.
  5. Department for International Development (2005) Why We Need to Work More Effectively in Fragile States. January.
  6. The Council of the European Union (2007) Conclusions on a EU Response to Situations of Fragility. 2831st External Relations Council Meeting, Brussels, 19-20 November.