Achieving Turnaround in Fragile States

by Andrew Rosser

Institute of Development Studies

at the University of Sussex

Brighton BN1 9RE

United Kingdom

Email:

November 2005

Draft Only

Not for Citation


1. Introduction

Over the past few years, the development challenges faced by fragile states have moved to the top of the international development agenda. Partly, this has reflected an apparent increase in the number of fragile states in the world over the past two decades.[1] Since the end of the Cold War, some powerful states have been much less willing and able to support and maintain weaker states, making the latter vulnerable to failure or even collapse (Doornbos 2005). At the same time, globalisation has created a range of economic, political and security-related pressures that have overwhelmed many weaker states (Clapham 2002). Combined with internal political, social and economic conditions that make countries vulnerable to the onset of civil war and economic collapse—such as natural resource wealth, ethnic diversity, and large numbers of young unemployed males (Collier and Hoeffler 1998; Collier 2000; De Soysa 2000; Ross 1999; Auty 2004)—these developments have dramatically increased the incidence of state fragility in the developing world.

In part, the increased interest in the development challenges faced by fragile states has also reflected the feelings of insecurity in the West caused by the terrorist attack in the US on 11 September 2001 and subsequent terrorist attacks elsewhere. These attacks have made it clear that extreme poverty, economic backwardness, and severe violent conflict in fragile states are a threat to people, not just in these countries, but in other countries as well. The spill-over effects on neighbouring countries of various civil wars in fragile states such as those in Rwanda and Colombia during the 1990s have reinforced this perception (Millett 2002; Gleditsch 2003). Many international development organisations have responded to these developments by making improved conditions in fragile states one of their key goals (DFID 2005; USAID 2005; World Bank 2002; AusAID 2005).

The purpose of this paper is to explore how these organisations might most effectively work to improve conditions in fragile states. Many donors have been influenced by the idea, associated with the work of Burnside, Collier and Dollar (Burnside and Dollar 1997; Collier and Dollar 1999; World Bank 1998), that aid is most effective in promoting development in countries that have sound policies and institutions. This would suggest that aid should be redirected away from fragile states, where policies and institutions are generally weak, and towards other developing countries. But donors appear unlikely to significantly reduce aid to fragile states. As the World Bank (2002: iv) has noted, to disengage from fragile states would cause severe deprivation among the populations of these countries, reduce their chances of recovering, and increase the risk of state failure, ‘with its evident adverse effects, both regional and global’. Indeed, the current global trend is towards significant increases in global aid volumes, raising the prospect of significantly more aid flowing to fragile states (Manor 2005). At the same time, however, rather than simply throwing more money at fragile states, donors are actively trying to find ways of working more effectively in them.

But before donors can work more effectively in fragile states, they need to gain a better understanding of the factors that shape development outcomes in these countries. Devising better strategies for engaging with fragile states requires some sense of the potential for change in these countries as well as the variables that shape this potential. To this end, this paper examines the factors that have shaped development outcomes—more specifically, outcomes vis-à-vis ‘turnaround’, a concept that we define in detail below—in seven current and former fragile states: Vietnam, Laos, Uganda, Mozambique, Cambodia, Burundi and Indonesia. At the same time, we also examine the role of donors in contributing to outcomes vis-à-vis turnaround in these countries and assess the implications of our findings for donor strategies for engaging in fragile states. In presenting this analysis, we draw heavily on the following papers, some of which are published here: Fforde (2005) on Vietnam, Rosser (2005a) on Laos, Robinson (2005) on Uganda, Kulipossa (2005) on Mozambique, Hughes (2005) on Cambodia, Lemarchand (2005) on Burundi, and Rosser (2005b) on Indonesia.

The structure of our paper is as follows. In Section 2, we define some of the key concepts that are used in this volume. In Section 3, we assess the performance of each of the current and former fragile states in our sample vis-à-vis turnaround. In Section 4, we identify the factors that have shaped variation in our sample countries’ respective performances vis-à-vis turnaround and in Section 5 we examine the role of donors in facilitating or hindering turnaround. Finally, in Section 6, we assess the implications of our findings for donor strategies for engaging in fragile states.

2. Concepts and Definitions

The notions of ‘fragile state’ and ‘turnaround’ are central to the analysis here. In much of the development literature—and in particular that produced or commissioned by donors—these terms are defined in terms of the perceived quality of countries’ governance, policies, and institutions, as measured by the World Bank’s Country Policy and Institutional Assessment (CPIA).[2] The problem with this approach is twofold. First, it suggests that the principal development challenge faced by fragile states is weak governance, policies and institutions. Second, it suggests that there is only one way to improve governance, policies and institutions—that defined by the CPIA. The criteria that make up the CPIA emphasise the importance of deregulated markets, conservative macroeconomic and fiscal policies, and public administrative and other institutional structures that provide transparency and accountability (World Bank 2002; 2003). While it could be argued that improving governance, policies and institutions is an end in itself, our view is that these states’ principal challenges are to overcome economic stagnation, extreme poverty, and severe violent conflict. At the same time, as numerous studies of the East Asian experience have demonstrated, there are other pathways towards these goals besides that defined by the CPIA (Chang 2005; Weiss 1999; World Bank 1993; Wade 1990; Amsden 1989; Jenkins 1991; Haggard 1990, Robison 1989; Rodan 1989). In this way, the use of the CPIA as the primary indicator of ‘turnaround’ confuses causes with outcomes. Better definitions of fragile states and turnaround need to distinguish between desired outcomes and the models of development that are thought to lead to them.

For these reasons, we have defined ‘fragile states’ and ‘turnaround’ in this volume to refer to particular development outcomes rather than the supposed quality of countries’ governance, policies and institutions. More specifically, we have defined fragile states as low-income countries that (i) are performing poorly in terms of achieving real economic growth; (ii) are failing to significantly reduce poverty; or (iii) are experiencing severe violent conflict or have done so in the recent past; and turnaround as involving (i) sustained high levels of real economic growth; (ii) sustained significant reductions in poverty; and (iii) a durable cessation to severe violent conflict.[3]

This focus on development outcomes rather than the models for achieving these outcomes is arguably a more ideologically neutral means of defining fragile states and assessing their development performance. At the same time, however, we acknowledge that it has certain weaknesses. One is that it focuses on particular dimensions of development—real economic growth, poverty reduction, and peace—and ignores other important dimensions such as gender equality, environmental sustainability, and inclusion and democracy. Another is that it relies on what are to some extent arbitrary judgements about how much real economic growth or improvement in human development, or how few battle-related deaths are necessary to constitute turnaround.[4]

3. The Cases

The seven countries on which we focus have varied in terms of their respective performances vis-à-vis turnaround in recent decades. Some—specifically, Vietnam, Laos, Uganda, and Mozambique—have performed well vis-à-vis turnaround and are no longer fragile states. Cambodia and Burundi, on the other hand, have performed less well vis-à-vis turnaround and are still considered fragile states. Indonesia represents a third category of country in our sample—during the 1960s and 1970s, it achieved turnaround but during the 1990s, it then experienced a renewed period of fragility.

Vietnam and Laos were the site of a destructive Cold War-related conflict during the 1960s and early 1970s. The end of this conflict in 1975 ushered in a period of relative peace and stability in both countries but limited progress in terms of economic growth. Between 1975 and the late 1990s, these countries experienced erratic growth, periods of recession, and extreme poverty. In the late 1980s and the first half of the 1990s, however, the two countries experienced major economic booms characterised by increased foreign investment and growing exports, particularly in manufacturing. The Asian economic crisis in 1997-1998 brought an end to these booms but the countries have still achieved respectable rates of economic growth since the crisis. At the same time, both countries experienced marked improvements in their HDIs during the late 1980s and 1990s.

Uganda was in the midst of a civil war in the mid-1980s. Since then, peace and stability have been restored to virtually all of the country except the North, the economy has grown strongly (after a short period of stagnation in the mid- to late 1980s), and the country’s HDI has improved rapidly.

Mozambique endured civil war for sixteen years until the Rome peace agreement was signed in 1992 and a UN mission was established in the country. Since then, it has experienced relative peace and stability, rapid economic growth, and marked improvement in its human development index.

Cambodia was severely affected by civil war and poor economic performance during the 1970s and 1980s. Since 1990, it has experienced continued violent conflict but has managed to avoid a return to full-blown war. At the same time, its economy has grown strongly. It has not yet managed to achieve turnaround, however, because it has failed to make significant progress in improving human development levels.

Burundi experienced relative peace and stability and modest economic growth during the 1980s, although it performed poorly in terms of improving human development levels. In the 1990s, the country descended into civil war and its modest record vis-à-vis growth and human development grew worse. Recent years have seen efforts to achieve a durable war to peace transition but the country remains fragile.

Indonesia was one of the poorest countries in the world and experienced a severe economic crisis in the mid-1960s. During the late 1960s and 1970s, the country achieved sustained high levels of growth and a dramatic improvement in human development levels while avoiding severe violent conflict (at least outside East Timor, which it invaded in 1975, and West Papua, which was forcefully incorporated into Indonesia in the 1960s). In the late 1990s, however, the country experienced a partial reversal vis-à-vis turnaround characterised by a sharp drop in economic growth.

The fact that these countries vary in terms of our dependent variable—performance vis-à-vis turnaround—makes our sample a good one for exploring the factors that determine why some fragile states achieve turnaround and others do not. At the same time, however, our use of a case study approach means that our findings have a higher degree of validity in relation to the particular cases examined here than generalisability. Examination of the dynamics of turnaround in other fragile states is required to establish whether the findings here can be generalised beyond our sample countries.

4. Achieving Turnaround

What factors shaped whether or not the countries in our sample achieved turnaround?

In addressing this question, it is useful to distinguish between the ‘durable cessation of conflict’ and ‘sustained high rates of real economic growth and significant improvement in human development levels’ dimensions of turnaround. All our turnaround countries—or at least those that experienced severe violent conflict prior to achieving turnaround—experienced a cessation of severe violent conflict prior to achieving sustained high rates of real economic growth and significant improvements in human development. In other words, achieving a durable cessation to conflict was the first stage in the process of turnaround. At the same time, however, achieving a durable cessation of conflict was not sufficient in itself to generate turnaround. In Vietnam and Laos, for instance, an end to conflict did not automatically lead to sustained high rates of economic growth and improvement in human development levels—it was more than a decade before these things happened. This suggests that different factors were at work in relation to each of these dimensions of turnaround. Our discussion below is consequently organised according to these two dimensions.

Achieving a Durable Cessation of Conflict

All of the countries in our sample experienced severe violent conflict, as that term has been defined here, at some point during the periods under examination, except for Indonesia.[5] In all these cases, severe violent conflict initially came to an end as a result of one of two main factors: a change in geo-political conditions or a military victory by one party involved in the violent conflict.

In most cases the former was the important factor. In Mozambique, for instance, the fall of the apartheid regime in South Africa and parallel developments in Rhodesia removed the rationale for governments in these countries to continue supporting the resistance movement in Mozambique. Similarly, the end of the Cold War in the late 1980s provided the trigger for peace in Cambodia by reducing the capacity and willingness of various foreign governments to continue supporting the different sides in the conflict. In both cases, these changes led to international efforts, led by the UN, to ensure a transition to peace and the creation of a more democratic political system. In Laos and Vietnam, it was growing public opposition to the war in Indochina in the US and other Western countries that ultimately led to the initial cessation of conflict because it forced Western governments to withdraw their troops. In the Ugandan case, however, such geo-political factors were not important. In this case, it was the ability of the National Resistance Movement (NRM) to defeat its opponents on the battlefield that ultimately brought an end to the conflict. Importantly, the war to peace transition in Burundi has been less decisive, partly because the strategic location of the country has meant that geo-political changes have had little influence over the willingness or ability of warring parties in that country to continue fighting, and partly because both parties to the conflict have been able to maintain viable military forces.