Section I. Introduction

The problems affecting countries in conflict have been the focus of increased attention in recent years, with particular emphasis on the impact of civil wars, the incidence of which is now 10 times higher than that of international wars (Collier and Hoeffler, 2002a). The issue has attracted special attention on the African continent, where a majority of civil wars have been occurring, and directly or indirectly affect more than one in three African people. Addressing the issue of conflict has thus become an essential challenge for the leaders of the continent and external partners in the context of the New Partnership for Africa’s Development, as the restoration of stability to countries affected by conflict is increasingly perceived as a pillar of economic development, not only for the countries themselves, but also for neighboring countries and the region as a whole.

In confronting the problems characteristic of postconflict situations, an idea that has gained greater acceptance is that the traditional approach to development and aid policy is insufficient for war-affected countries. Those countries face specific internal problems, including limited administrative capacity, weak institutions, and poor infrastructure, which call for tailored solutions. Official development assistance efforts need to be designed in a manner whereby conflict is no longer treated as an exogenous factor, but as one of the key determinants of poor economic performance, whose causes and consequences should be taken into account in all efforts to enhance economic growth and reduce poverty.

This chapter attempts to summarize briefly some of the main findings from the most recent research on conflict and their implications for postconflict assistance, with a focus on the economic, rather than the social or political, dimensions of conflict.¹ The chapter is organized as follows: Section II examines the causes of conflict; Section III looks at the consequences of conflict; Section IV considers the role of aid, capacity building, and reconstruction; Section V examines the fiscal aspects of postconflict assistance; Section VI looks at the impact of military expenditure; Section VII considers the priorities for monetary policy and the financial sector; Section VIII looks at structural reforms and good governance; and Section IX offers conclusions.

Section II. Causes of Conflict

In a paper published in March 2002, Paul Collier and Anke Hoeffler (2002b) examine the causes of conflict on the basis of a sample of 78 civil conflicts that occurred between 1960 and 1999 and that involved at least 1,000 combat-related deaths per year. In particular, they construct a “grievance” model, where conflict is caused by inequality, political oppression, and ethnic and religious divisions, and a “greed” model, which emphasizes the role played by natural resources and other financing mechanisms in the emergence of conflict. Their empirical analysis gives little support to the hypothesis that conflicts are caused by grievances. Notably, they do not find that inequality or political oppression increase the risk of conflict. They find, however, that there is a systematically higher risk of conflict in countries where one ethnic group makes up 45–90 percent of the population.

The analysis lends stronger support to the greed model, showing that countries with abundant natural resources face a higher risk of conflict, and that this is also the case for countries that have large diasporas that can finance rebel movements. Exports of natural resources can increase the risk of conflict by financing rebel groups, increasing the incentive for secession, worsening corruption, and increasing exposure to shocks. In this regard, Michailof, Kostner, and Devictor (2002) observe that civil conflicts that are aimed at controlling rents and natural resources often become financially self-sustaining, making it all the harder to exert economic pressure on the warring factions. Collier and Hoeffler (2002b) point out, however, that the relationship between natural resources and conflict risk is nonlinear, because the risk of conflict diminishes when a country’s dependence on natural resource incomes is very high. They suggest that in such cases, primary commodities that are taxed at high rates become a significant source of revenue for the government and that this revenue can be used to strengthen the state.

Collier and Hoeffler also find that the risk of civil war is higher in countries with a low GDP per capita, where the economic opportunity cost of rebellion is low and unemployed youths can easily be recruited into rebel movements, as well as in countries that are mountainous or have an unequally distributed population. Conversely, the risk of conflict diminishes with a higher enrollment rate in secondary education and when societies are religiously and ethnically diverse. Luckham and others (2001) argue that conflict and poverty are in a “dynamic and mutually reinforcing relationship,” in which conflict becomes a key determinant of poverty, which, in turn, exacerbates the impact of conflict and makes civilians more vulnerable. Fearon and Laitin (2003) argue that low GDP per capita is in fact a proxy for state strength, and that weak states have more difficulty preventing insurgency and political tensions from escalating into civil war. Addison (2002) also highlights the role played by high levels of uncertainty and weak institutions, both in fueling conflict and magnifying its negative impacts once it has taken place. For example, he remarks that high levels of uncertainty lead the private sector to focus on commerce rather than long-term investment in production, because the latter is more vulnerable to predation. In Chapter 4 in this book, Yartey extends Collier and Hoeffler’s model to include institutional quality and corruption as two additional variables explaining conflict and finds that they are highly significant, underscoring the importance of good governance and institutional development in any postconflict policy framework.

Luckham and others (2001) consider that conflicts in Africa originate in the legitimacy crises faced by colonial and postcolonial states. At the same time, they observe that the conflicts that took place in the immediate postcolonial era have transformed political, social, and economic realities, and that, as a consequence, “the factors which sustain present conflicts are not necessarily those which originally caused them.” Michailof, Kostner, and Devictor (2002) suggest that the end of the cold war has led to the emergence of a new type of war on the African continent, characterized by civil conflicts increasing in intensity, and gradually involving neighboring countries and regional powers. They note that the end of the cold war led to the military disengagement of non-African powers, which had transferred up to US$4 billion a year in arms to the continent, and that this disengagement was followed by a period of rising tensions fueled by deep internal tensions, poor governance and human rights records, large and underpaid standing armies, and considerable stocks of arms and ammunitions that helped sustain the fighting when conflicts erupted. Luckham and others (2001) remark that conflicts following the cold war have been associated with the delegitimization of the state, and that warfare has become a means for armed groups to accumulate power and wealth, where, increasingly, civilians have borne the costs of war.

Michailof, Kostner, and Devictor (2002) mention other causes of recent conflicts, including (1) unprecedented demographic growth, resulting in increased pressure on land and natural resources, as well as large migrations to cities, leading to social and political instability; (2) poverty, illiteracy, and large-scale unemployment; (3) exclusion from political and economic life on regional, ethnic, or social grounds; (4) mismanagement of economic rents and struggles to maintain or gain control over those rents; (5) subregional instability, with recent research pointing to a 0.55 probability that a country neighboring a conflict will suffer the same fate; and (6) easy access to small arms from regular army stocks and the international market. As pointed out by Luckham and others (2001), new financing mechanisms for civil wars, which increasingly involve neighboring countries for the purchase of weapons and the sale of mineral resources, and do not originate primarily in the major powers, have contributed to the regional instability dimension of contemporary conflicts in Africa.

Chapter 7 in this book by Ghura and Mercereau, which looks specifically at political instability and its impact on growth in the Central African Republic (C.A.R.), shows that a low domestic revenue–GDP ratio and an adverse terms-of-trade shock significantly increase the risk of a coup d’état. In particular, the authors note that weak revenue performance undermines the government’s ability to pay civil servants’ wages and provide basic social services, which, in turn, feeds discontent within the population and triggers political instability.

Section III. Consequences of Conflict

As Luckham and others (2001) underscore, it is especially difficult to estimate with precision the economic impact of conflicts, not least because they are often associated with the collapse of state institutions and, consequently, the data-collection systems that are required to assess the damage caused to economic growth and poverty reduction efforts. Typically, civil wars lead to the destruction of infrastructure, the collapse of state institutions and administrative capacity, a halt to investment activities in all sectors of the economy (with a lasting negative impact on investor confidence), high inflation levels that primarily affect the poorer segments of society, a worsening of the trade balance, an increase in foreign indebtedness, and a decline in external aid flows that is followed, once the conflict is over, by a shift from development aid to humanitarian aid (see Figure 3.1 for the estimated impacts on GDP growth). Civil wars are also accompanied, in most countries, by a sharp decline in government revenue, not only because of the narrowing of the tax base, but also because tax collection falls as the credibility of punishment for tax evasion diminishes when a government is threatened, as pointed out by Caplan (2002).

View Full Size

Real GDP Growth in Conflict Countries

(Average annual percent)

Source: Gupta and others, 2004.Note: Based on a sample of 12 countries. The real GDP per capita growth corresponding to the preconflict, conflict, and postconflict periods is –3.7, –3.5, and 1.2 percent per annum, respectively.

• Download Figure
• Download figure as PowerPoint slide

In a study published by the World Bank, Collier and others (2003) estimate that by the end of a typical civil war, incomes are on average 15 percent lower than they otherwise would have been, and that about 30 percent more people live in absolute poverty. In addition, the authors note that many of the economic costs persist for several years after the end of the conflict, notably through high military expenditures and capital flight. Collier (1995) points out that civil wars are also different from international or liberation wars, in the sense that the latter can generate some positive effects, including the strengthening of the state, technological progress, and the mobilization of productive resources, whereas civil wars lead to the impoverishment of the population and gradual depletion of an economy’s productive resources. Civil wars, from this perspective, lead only to the destruction, rather than the creation, of social capital. In that regard, research by Caplan (2002) on the impact of war on the economy shows that, although foreign wars can be associated with slightly above-average real GDP growth, domestic wars have a clear and substantial negative impact on real GDP growth.

In a draft paper on the lessons of the International Monetary Fund’s involvement in 23 low-income, postconflict countries over a period totaling 30 years, Fallon and others (2004) remark that most countries that suffered conflict in the post-1990 period entered conflict with a per capita GDP that was lower than five years earlier, and that fiscal deficits in those countries had risen in the period immediately preceding the conflict because of falling revenues and buoyant defense expenditure. They also note that, while post-1990 conflicts tended to be of shorter duration (4 years on average) than pre-1990 conflicts (12 years on average), they were also characterized by more severe economic contraction, with real GDP declining to an average of 74 percent of its preconflict level. Moreover, recovery from conflict was a lengthy process, with countries emerging from conflict with output levels far below preconflict levels and in urgent need of macroeconomic stabilization.

A comparison of poverty and income trends for conflict and non-conflict countries in sub-Saharan Africa over the 1972–97 period in the October 2001 edition of the IMF’s World Economic Outlook is also revealing with regard to the damage caused by conflict. Notably, real GDP per capita in purchasing power parity terms grew at an average annual rate of 5.5 percent in nonconflict countries, compared with 3 percent in conflict countries; the infant mortality rate for nonconflict countries fell by 36.5 percent, compared with a decline of 25.5 percent for conflict countries; life expectancy increased by 17.5 percent for nonconflict countries, compared with 9.5 percent for conflict countries; and gross primary school enrollment increased from 61 to 89 percent between 1972 and 1992 in nonconflict countries, compared with an increase from 46 to 66 percent in conflict countries.

Looking at the consequences of conflict from an African perspective, Addison (2002) observes that war “destroys the human and physical capital of the poor and undermines the family ties and relationships that are central to the livelihoods of Africa’s communities.” Combined with the destruction of essential services and infrastructure, he adds, these effects may weaken the poorer segments of the population to the extent that they are not able to share the benefits of recovery once the conflict is over. Referring, in particular, to the work of Ahmed (1994) on Somaliland, and Cliffe (1994) on Eritrea, Luckham and others (2001) emphasize that rural areas have been particularly affected by civil wars in the case of sub-Saharan Africa, not least because rebel movements often originate from parts of the territory over which the incumbent regime has little or no control. The impact of conflicts on rural areas includes the destruction of crops, the killing of livestock, the destruction of irrigation networks and, eventually, the dislocation of markets as farmers are compelled to leave their farms because of excessive insecurity. The displacement of rural populations to urban areas that are illequipped to face a surge in population adds to the political instability and the decline in living standards faced by the war-affected country. The consequences of the decline in agricultural output can also have dramatic consequences in terms of food security, as was most strikingly demonstrated by the famine suffered by the Horn of Africa region in 1983–84, when more than 2 million people are estimated to have died.

Referring to the works of Addison (1998), Bruck (1997), Collier (1995), Fitzgerald (1997), Harris (1999), and Mubarak (1997), Luckham and others (2001) note that civil wars in sub-Saharan African have led to the development of informal war economies that have created new forms of inequality and poverty. Although those parallel economies can help ensure the survival of the population during the conflict, they generally become an obstacle to the reconstruction of the economy once the conflict is over. In particular, informal war economies tend to favor activities with low productivity and short-term returns, and households tend to favor consumption rather than savings and productive investment. In addition, informal war economies lead to a significant increase in transaction costs because of the uncertainty linked to the weak legal and regulatory environment, discouraging investments in the private sector. As noted by Collier (2000), “one of the casualties of civil war is trust.” Another common feature of informal war economies is the segmentation of markets, because parts of the territory are controlled by rebel groups and the main transport routes often become the target of attacks. The combination of segmented markets, rising costs, a weak regulatory environment, and reduced administrative capacity feeds the opportunistic, rent-seeking behavior of armed groups. This is especially the case for resource-rich areas that come under the control of armed groups that have a vested interest in the perpetuation of conflict, thus complicating the task of reconciliation and recovery.

The economic costs of conflict depend on the way conflict is financed. When conflicts are financed by accumulating debt, the repayment commitments mean that less financial resources will be available in the future to invest in social capital. When conflicts are financed through taxation, segments of society will suffer, and economic activity is likely to decline, especially in the formal sector. And when no additional financing is available, a country may accumulate expenditure arrears, or allocate a share of existing resources to defense expenditure. In this event, Mohammed (1999) considers that the opportunity cost of a conflict becomes the social rate of return to expenditure in any area where expenditure is reduced. This problem is exacerbated in countries where the state’s ability to collect revenue has already been diminished by the conflict.

Michailof, Kostner, and Devictor (2002) underscore that the destruction of social capital and institutions poses a significant challenge to postconflict economic revival, with a 0.5 probability that countries return to war within five years of a peace agreement. They express concern about the regional economic impact of conflicts, noting that even when neighboring countries succeed in maintaining peace, the presence of a large number of refugees, the accumulation of small arms, disruptions in trade, and a heightened sense of insecurity by potential investors inevitably take a toll on their economies. The regional contagion effects of civil wars have been highlighted by Sambanis (2003), who cites the case of the African Great Lakes region, where recurrent wars and population displacements in Burundi and Rwanda helped fuel instability and conflict in the Democratic Republic of the Congo (DRC), and led to the involvement of Uganda and Zimbabwe. Sambanis’s (2001) analysis of ethnic civil wars during the 1945–99 period also indicates that a country was three times as likely to face a civil war if its neighbors were undemocratic or confronted with an ethnic war of their own. In that regard, Lake and Rothschild (1998) point out that contagion can be driven “by alliances between transnational kin groups … or by predatory states that seek to take advantage of the internal weaknesses of others.” Yartey’s accompanying chapter also examines the case of the African Great Lakes region and finds that the average risk of war for the region during the 1960–99 period stood at 27 percent, far in excess of the 9 percent average for sub-Saharan Africa as a whole. Looking at the experience of the C.A.R., Ghura and Mercereau’s accompanying chapter tends to confirm the regional contagion hypothesis, indicating that recent military upheavals in the country have been fueled in part by conflicts in neighboring countries, including the DRC and the Republic of Congo, which had adverse economic effects on the economy and provided politicians with military allies.

Collier and others (2003) add that civil wars impose costs at the global level, because territories are created that are outside the control of recognized government, facilitating the production and transport of hard drugs. The authors report that 95 percent of the global production of hard drugs occurs in countries with civil wars.

Section IV. Aid, Capacity Building, and Reconstruction

In a paper on the economics of postconflict aid, Demekas, McHugh, and Kosma (2002) analyze the differences between postconflict aid and conventional development aid and, in particular, their different effects on the recipient economy, including capital accumulation, growth, welfare, and resource allocation. Whereas development aid’s traditional objectives are to encourage investment and growth and reduce poverty, postconflict aid has a humanitarian objective and a reconstruction objective. Reconstruction aid is aimed not only at repairing physical and capital infrastructure, but also at institution-building efforts. The two forms of aid also differ in the magnitude and timing of disbursements, postconflict aid being massive and concentrated in a short period, while conventional development aid is a relatively steady flow over time.

Demekas, McHugh, and Kosma suggest that while humanitarian aid, which is intended primarily to support basic consumption needs, can lead to the same unintended effects as conventional development aid in terms of reducing the competitiveness of export industries, through the so-called “Dutch disease” effect, this is not necessarily the case for reconstruction aid. Reconstruction aid is different in the sense that it can help ease supply bottlenecks and thus contribute to an increase in productivity in the tradable goods sector.

The analytical framework developed by the authors shows that humanitarian aid achieves its main objective, which is to enhance welfare by addressing urgent basic human needs. However, by raising permanent income, enabling the representative agent to achieve his/her desired level of consumption with a lower capital stock, it reduces capital accumulation in the long run and steady state growth. The authors do not find an unambiguous relationship between reconstruction aid and capital accumulation. While the marginal productivity of reconstruction aid tends to reduce the long-run capital stock, the impact of reconstruction aid on labor supply has an ambiguous effect, whereas the impact of reconstruction aid on savings and on the productivity of capital both tend to increase the long-run capital stock. The authors’ presumption on the basis of their findings is that postconflict reconstruction aid tends to raise the equilibrium capital stock. In particular, they note that by boosting productivity in both the tradable and the nontradable goods sectors and influencing consumption decisions, reconstruction aid may result in a sustained expansion of the tradable goods sector.

Demekas, McHugh, and Kosma offer the following recommendations from their study: First, donors should sequence their aid flows. In the immediate postconflict phase, aid resources are best devoted to humanitarian aid and will have the greatest impact in terms of enhancing welfare. Second, to preserve long-term economic development, humanitarian aid should be disbursed rapidly, with modest consumption targets. The authors note that overly generous or protracted disbursement of humanitarian aid risks creating aid dependency and reducing labor supply, ultimately proving counterproductive. And third, reconstruction aid should be designed so as to stimulate the development of the tradable goods sector, with an emphasis on public services that facilitate the development of the private sector.

Collier and others (2003) also note that in postconflict cases, aid flows tend to be massive immediately following the end of the conflict, when media coverage is at its peak, but rapidly taper out. They underscore the need for more important overall aid flows in the first postconflict decade and note that those resources should be disbursed more gradually. Collier and others also find that aid flows tend to be most effective in terms of stimulating economic growth after a period of about five years, when institutions have strengthened, the infrastructure has improved, and absorptive capacity has increased. For this reason, they consider that external assistance in postconflict situations should be designed to cover a period of at least 10 years. Fallon and others (2004) also emphasize the need for improved coordination among donors regarding the magnitude and the timing of their financial support, pointing to the negative impact shortfalls in foreign financing have had on development expenditures that are essential for reconstruction and economic recovery. Based on their analysis of the regional economic implications of civil wars, Murdoch and Sandler (2003) also underscore that assessments of the effectiveness of aid flows should take into account the reduced growth rates of some recipient countries owing to their proximity to conflict areas, and that peacekeeping efforts funded by the international community have regionwide public benefits.

While noting that international actors can help in the postconflict recovery process, notably through peacekeeping, humanitarian aid, budgetary support, debt relief, and foreign direct investment, Addison (2002) emphasizes that for international assistance to work well, the capacities of national actors, encompassing communities, the private sector, and the state, must be strengthened. In addition, rather than being based on a wish list for projects, the provision of external financial assistance should focus on a set of core priorities for broad-based recovery on the basis of the needs of communities as identified from household consumption surveys or participatory poverty assessments. The author mentions that the immediate priorities in the context of a reconstruction and reform agenda include the demobilization and reintegration of former combatants, caring for refugees and displaced populations, and a timetable for multiparty elections. While acknowledging the commonly held view that economic reforms should be delayed until reconstruction is well under way, because of the need to secure peace, social stability, and political commitment in the immediate postconflict period, Addison considers such an approach to be unrealistic, because economic policymaking cannot be put on hold during the initial stages of reconstruction. In his opinion, rapid economic reforms, if they are well designed, can in fact help ensure that economic recovery will be broad based, rather than narrow, in its benefits.

Section V. Fiscal Aspects of Postconflict Assistance

Fallon and others (2004) observe that countries tended to emerge from recent conflicts with severely reduced domestic revenues and damaged tax administration. To restore macroeconomic stability it is important to increase domestic revenues by restoring the tax base and administrative capacity to collect taxes.

As in other countries, the IMF has advised governments in postconflict countries to implement a set of broad-based, equitable, and easily monitored revenue-generating instruments that can be expanded as the capacity of the country grows. In postconflict countries, limited capacity restricts the choice of instruments so that meeting revenue objectives sometimes has involved advocating a second-best option in the near term that will not impede the development of a wider and deeper revenue system in the longer term. In countries with sufficient preexisting capacity, the IMF’s advice on tax-revenue instruments has focused on identifying which instruments are viable given the administrative capacity available with some modification, and on revitalizing existing administrative capacity (see also IMF, 2004, forthcoming). Where necessary, the IMF has advised that tax instruments be pared down to a realistic set that can be implemented effectively and monitored. In countries, especially those born out of conflict, with almost nonexistent capacity, the introduction of indirect taxes, especially border taxes, is typically the first step toward establishing a revenue-generating system because these taxes are paid at the source of the taxable transaction and therefore are relatively easy to implement and monitor. In that regard, the paper by Fallon and others indicates that although good tax administration is essential for successful revenue-generating systems, it is also one of the most challenging tasks in countries born out of conflict, as it involves the development of institutions from scratch, including the establishment of their legal basis, training of personnel, and the introduction of tax instruments. In addition, political factors such as lack of political will or the inability of parliaments to adopt new legislation can act as major obstacles to the reconstitution of an effective tax administration. Fallon and others note that the countries with IMF postconflict programs under review have been broadly successful in meeting their domestic revenue program targets.

Looking specifically at government revenues linked to natural resources, Collier and others (2003) underscore that such revenues are too often used inefficiently or corruptly, and that there is a need to better harness the potential growth of natural resources for economic growth. To that effect, they propose the creation of an international template for the good governance of natural resource revenues to which governments and companies could choose to adhere. The proposal would work along the following lines: National and international companies in extractive industries would report their payments to governments, governments would report their receipts from the national and international companies, the World Bank would collect the information and reconcile payments and receipts, and the IMF would integrate the net revenue figure into budgetary data and analysis. Apart from ensuring a more efficient use of such resources, the template could also be used to attract more reputable investors, according to the authors. Collier and others add that another priority is to put an end to illicit payments by extraction companies to bribe country officials, and that some progress has been made on this front, notably through new legislation in the member states of the Organization for Economic Cooperation and Development banning companies from bribing government officials.

In the area of government expenditures, Fallon and others (2004) note that postconflict countries are faced with a complex set of challenges, including the need to demobilize and reintegrate ex-combatants, resettle displaced persons, rebuild infrastructure and institutions, and improve public service delivery to create political stability. These challenges are made all the more daunting by the limited amount of resources available. The IMF’s recommendations on expenditure composition have included a shift in public spending from recurrent to capital activities, higher priority to social sector spending, and a reduction or at least a freeze in the size of the civil service. Fallon and others note that although government expenditures typically increased in per capita terms following the end of the conflict, they were nevertheless well below program targets in most cases. This was due mainly to weakened implementation capacity of governments and, in some instances, to shortfalls in external financing. Fallon and others also note that the record concerning the recovery of social spending has been mixed. Even though public expenditure management systems improved during the postconflict period, fiscal institutions generally remained weak. Underscoring the limited institutional capacity of governments, the paper indicates that, in many cases, weakened civil services hindered the preparation and execution of budgets according to postconflict priorities, and that the link between public expenditures and outcome was weak, partly because of lack of fiscal accountability and transparency.

Addison (2002) emphasizes that in the immediate postconflict period, a thorough review of spending priorities and public expenditure management is required. Otherwise, there is a strong likelihood that the additional resources generated by reduced military spending and reconstruction aid will not be used efficiently. Public expenditure reform should therefore be undertaken as early as possible following the end of the conflict. In addition, Addison suggests that a fairer allocation of public spending across regions and ethnic groups is necessary to address some of the social inequalities that often characterize the prewar pattern of public infrastructure and services that may have contributed to the conflict. Figure 3.2 summarizes the evolution of the fiscal aggregates in a sample of 14 countries as they moved from the preconflict to the conflict stage, and from the conflict to the postconflict stage.

View Full Size

Fiscal Aggregates in Conflict Countries

(Percent of GDP)

Source: Gupta and others, 2004.Note: Based on a sample of 14 countries.

• Download Figure
• Download figure as PowerPoint slide

Section VI. Impact of Military Expenditures

In a paper originally published in 1996, Knight, Loayza, and Villanueva show that high levels of military spending have a negative impact on growth by reducing productive capital formation and distorting resource allocation within the economy, and that substantial cuts in military spending achieved through international peace-building efforts could significantly enhance economic growth, notably through higher expenditures on infrastructure, education, and health. According to another study by Shieh, Ching-Chong, and Wen-Ya (2002), defense expenditures can have three different impacts on the economy. First, they lead to a crowding-out effect, by reducing the resources in the economy available for private investment and public spending on sectors that have a positive impact on growth. Second, defense expenditures may lead to positive supply-side spillover effects on other sectors of the economy; however, this impact is likely to be limited in low- and middle-income countries that tend to import their armaments. And third, defense expenditures may lead to an increase in private savings and investment by enhancing security conditions. However, this third impact is less relevant for the analysis of countries that have already suffered from conflict.

A paper by Gupta and others (2004) uses two approaches to provide a cross-country examination of the fiscal consequences of conflict. First, the study analyzes the evolution of various macroeconomic and fiscal variables during 22 episodes of conflict and in the years immediately preceding and following the conflicts. This analysis suggests that the higher level of government expenditure on defense that takes place during conflict tends to generate greater macroeconomic instability, as reflected for example in higher budget deficits and a pickup in inflation, but does not necessarily lead to lower levels of spending on education and health as a percentage of GDP (see Figure 3.3). However, because conflict is associated with lower real GDP growth, real per capita government spending on education and health is lower during conflict periods. This is consistent with the findings of their second approach, which uses panel data for a large number of low-income countries to estimate econometrically the fiscal consequences of conflict. Their econometric results suggest that conflict has a direct and significant negative impact on growth, rather than an indirect effect through its impact on the composition of government spending. Moreover, armed conflict does not appear to have a significant impact on the government revenue–GDP ratio independent of its impact on real economic activity.

View Full Size

Composition of Government Spending in Conflict Countries

(Percent of GDP)

Source: Gupta and others, 2004.Note: Based on a sample of 12 countries for defense expenditure and on 6 countries each for education and health spending.

• Download Figure
• Download figure as PowerPoint slide

Recent research by Collier and Hoeffler (2002c) emphasizes the regional dimension of military spending, with a country’s expenditure heavily influenced by expenditure in neighboring countries. In particular, they estimate an arms race multiplier effect and find that an initial exogenous increase in military spending in one country eventually leads to a doubling of expenditure in both the originating country and its neighbor. They find that, through the arms race multiplier effect, civil war leads to an increase in military expenditure by 2.9 percent of GDP in the country where conflict originates and by 1.7 percent of GDP in neighboring countries. Also, Collier and Hoeffler find that a 10 percentage point increase in the risk of civil war within a subregion, linked to the greater availability of armaments for instance, leads to an increase in military spending by 1 percent of GDP in each country.² In addition, the authors do not find in their regressions any offsetting positive effect of military expenditure in terms of acting as a deterrent to rebellion. Military spending by the government as a response to security concerns is ineffective, because it is matched by spending by the rebel groups, and also because it gives a signal to the population that the government’s survival strategy is based on coercion rather than inclusion. In that sense, military spending might in fact become counterproductive, notably through its adverse effect on economic growth. For this reason, Collier and Hoeffler consider that military expenditure is a regional public bad, which inflicts negative externalities across borders. An important implication the authors draw from their analysis is that in the absence of negotiated reductions in postconflict military spending, much of the cost of a war might accrue after it is over. Therefore, the solution they advocate is for governments and regional political organizations to negotiate reciprocal reductions in military spending, and for international financial institutions that have more reliable information on military expenditure to act as an honest broker and, if necessary, to respond to increases in military budgets that breach regional agreements by reducing aid flows. In that regard, an interesting finding by Davoodi and others (2001) is that the easing of international and regional tensions accounted for up to 66 percent and 26 percent of the decline in global military spending during 1972–89 and 1990–94, respectively.

With regard to the existence of a peace dividend once the conflict is over and defense expenditures are reduced, Luckham and others (2001) underscore that this depends essentially on the capacity of states to restore their revenue base and to invest in the social sectors and infrastructure. They also emphasize that for the peace dividend to last, efforts to demobilize and reintegrate former combatants are crucial.

Section VII. Priorities for Monetary Policy and the Financial Sector

In a study published in 2000, Addison, Murshed, and Le Billon emphasize the importance of rebuilding the financial system following the end of conflict, because it plays a key role in contributing to macroeconomic stability and facilitating the resumption of investment by the private sector.

To begin with, they observe that the financial sector often plays a role in “narrow development,” a development process that fails to reduce poverty and exacerbates inequalities, thus contributing to the causes of conflict. In particular, this can happen when the state banking system is used to finance private accumulation, or when weak financial regulation facilitates the accumulation of wealth by fraud. The capacity to mobilize financial resources is what enables belligerents to engage in conflicts and to prolong them, the authors note, whether through domestic financing or the mobilization of external capital inflows via commercial borrowing, official borrowing, or remittances from diasporas. Therefore, they suggest that action to cut the external financing of wars would be one way to hasten their end, while cautioning that such action, if not carefully planned, might have asymmetric effects on belligerents, leading to outcomes that might not be welfare improving. In particular, they note, as do Collier and others (2003), the need to stop the financing of wars through trade in natural resources, such as conflict diamonds, and the money-laundering activities of organized crime.³

Following the end of conflict, Addison, Murshed, and Le Billon (2000) note the importance of a well-functioning financial system to facilitate reconstruction and, in particular, to stimulate the investment activities of the private sector that will help recreate the market networks that were destroyed during conflict. They note that the track record of direct state control of the financial system in conflict countries has been poor, and that greater financial liberalization is generally advocated, including by allowing market mechanisms to determine the level of interest rates. Also, the fact that state banks generally need to be recapitalized often leads to their privatization, as private sector money is used to “capitalize and reinvigorate the management of former state banks.” In addition, financial liberalization needs to be accompanied by measures to enhance prudential regulation and supervision by the monetary authorities. In that regard, the authors underscore the risk of “regulatory capture” in conflict and postconflict societies, when the absence of democratic institutions to impose and supervise an adequate regulatory framework leads to the abuse of the financial system for personal gain by powerful political actors. Finally, they recommend measures to stimulate the growth of domestic capital markets, so that public spending on reconstruction can be financed without resorting to monetization of the fiscal deficit.

Fallon and others (2004) note that the ability of central banks to conduct independent monetary policies and exercise their supervisory role over banks is considerably weakened by conflicts. Countries have tended to emerge from recent conflicts with increased inflation rates linked to the monetization of fiscal deficits aggravated by reduced money demand (see Figure 3.4). With the exception of countries in a monetary union, the loss of confidence in the currency and the banking system and excessive monetary expansion also led to the nominal depreciation of the exchange rate. The IMF’s advice has generally focused on reducing inflation and rebuilding foreign reserves, through a combination of fiscal consolidation and monetary constraint, appropriate exchange rate policies, and improvement of monetary instruments. Fallon and others note that countries with IMF postconflict programs under review have tended to be very successful in meeting their inflation targets. In addition, emphasis has been placed on financial restructuring, including, in particular, measures to recover overdue loans, accelerate provisioning for risky assets, inject new capital, and transfer nonperforming loans to special agencies, as well as promoting the establishment of a sound banking supervision system by enhancing the legal and regulatory framework and providing training to supervision staff at the central bank (see also IMF, 2004, forthcoming).

View Full Size

Consumer Price Inflation in Conflict Countries

(Average annual percent)

Source: Gupta and others, 2004.Note: Based on a sample of nine countries.

• Download Figure
• Download figure as PowerPoint slide

Section VIII. Structural Reforms and Good Governance

In the area of structural reforms, Fallon and others (2004) note that the countries reviewed were in urgent need of structural reforms to help the recovery process and strengthen longer-term growth. Structural measures have been a key part of IMF postconflict programs, but have varied widely in scope and included arrears such as public sector reform, establishment or rehabilitation of fiscal and monetary institutions, and privatization. The main priorities typically have included the strengthening of good governance, privatization efforts, the rehabilitation of customs and tax administration, the implementation of an expenditure control and monitoring system, the improvement of public procurement, and the simplification of the external tariff system. There were two sets of countries with particular needs. The structural reform agenda was particularly important in those new countries born out of conflict that entirely lacked an institutional infrastructure. In these countries, the priority was to put in place new government entities to handle fiscal, monetary, and exchange rate policies. Structural reform was also important in the transition economies that needed to reorient the economy from a centralized economy to market-based economic structures. The authors conclude that countries have tended to have difficulty in implementing their structural reform agenda. The new countries were typically the most successful, followed by the transition economies. Countries whose implementation of structural reforms was the weakest were those that suffered from fragile political institutions, including dissent from entrenched political and institutional interests that favored the status quo. In addition, Fallon and others emphasize that the ambition and size of the structural reform agenda were not always realistic, with compliance on the more difficult issues waning over time, and that the lack of sufficient human resources has contributed to the difficulty of the reform process.

A similar point is made by Collier and others (2003), who underscore that in the difficult environments characteristic of postconflict countries, there is a need to resist the temptation to address too many issues simultaneously, and that structural reform efforts should be based on a sequential and selective approach, with sufficient attention given to political considerations and the need to build a constituency for reform among the population.

Addison (2002) notes that in the postconflict period, strong and sustained investment from the private sector, both domestic and foreign, is required to support the recovery process. To that effect, the state must create the right conditions for investment by providing macroeconomic stability; reforming the legal and regulatory framework so that property rights are respected; and enhancing public utilities, notably through investments in the areas of telecommunications, water, and road infrastructure. At the same time, Addison adds, the state should regulate the private sector to protect the public interest, which can be a challenging task when democratic institutions to oversee and protect the public interest are only emerging. In addition, the process of privatization needs to be carried out in a transparent and competitive manner to avoid the transfer of valuable assets to local vested interests.

Collier (2000) remarks that countries for which natural resources represent more than a quarter of GDP face a higher risk of conflict. A priority for those countries should be to generate growth and diversify their economy away from dependence on primary commodities. Key factors to achieve diversification include sound macroeconomic policy; trade liberalization; investments in human capital; improvements in the country’s infrastructure, notably in the areas of water, electricity, and telecommunications; and enhanced investor confidence through a reliable judiciary system and a transparent regulatory framework. Collier also notes that, rather than playing a destructive role through their financial support to rebel groups, diasporas can play a constructive role in the development process through their skills and business connections. This points to the importance of a national reconciliation process being in place in the immediate postconflict period.

Section IX. Conclusions

From the above analysis, a few recommendations can be offered regarding the main economic areas in which to concentrate efforts to assist countries.

First, with regard to the causes of conflict, existing research by Collier and Hoeffler (2002b) shows that, in addition to poverty, a dominant economic cause of conflict has been the poor management of revenues linked to the extraction of natural resources. This situation calls not only for domestic efforts to reform extractive industries and ensure greater transparency in the management of the revenues linked to those industries, as noted by Collier and others (2003), but also international efforts to curb the trade in natural resources from conflict-affected areas.

Second, there is broad agreement that capacity building has a key role to play in a postconflict environment. Institutions and administrative capacity must be strengthened as quickly as possible, notably through the provision of technical assistance. As indicated by Fallon and others (2004), this step is essential to achieve revenue targets, conduct an independent monetary policy, ensure effective banking supervision, and facilitate the implementation of structural reforms. In this regard, it is worth emphasizing that, in light of the administrative and institutional constraints of postconflict countries, the technical assistance efforts of external partners should be coherent and consistent in their objectives.

Third, as indicated by Demekas, McHugh, and Kosma (2002), aid flows should be well sequenced and adequately coordinated among donors, with an initial focus on humanitarian aid, followed by a gradual increase in reconstruction aid as the recipient country’s capacity to use that aid effectively is enhanced. External assistance requirements should be assessed more accurately, notably through closer donor involvement in program countries, as indicated by Fallon and others (2004) and Addison (2002), and they should be designed from a long-term perspective, as noted by Collier and others (2003).

Fourth, in the area of revenue generation, tax instruments should be chosen on the basis of simplicity of implementation and operation. As indicated by Fallon and others (2004), tax instruments should be reasonably nondistortionary and modified as the tax system grows. Also, tax and customs administration need to be strengthened to improve revenue collection. If revenue can be increased without augmenting the fiscal pressure, and the additional revenue is invested in propoor policies, social cohesion is likely to be enhanced as a consequence.

Fifth, with regard to government expenditures, a thorough review of public expenditure management and spending priorities is called for, as pointed out by Addison (2002), with a particular focus on the need to demobilize and reintegrate ex-combatants, rebuild infrastructure and institutions, and improve public service delivery, with a fairer allocation of public spending across ethnic groups and regions. In that context, Collier and Hoeffler (2002c) note that efforts to reduce military spending and to reorient expenditures to the social sectors should be encouraged, with increased emphasis on reciprocal reductions negotiated on a regional basis.

Sixth, in the area of monetary policy, the experience of postconflict countries shows that, with the exception of countries in a monetary union, those economies tend to suffer from high inflation levels and a depreciated exchange rate. As noted by Fallon and others (2004), recommended strategies to reduce inflation and rebuild reserves include a combination of fiscal consolidation and monetary constraint, appropriate exchange rate policies, and improvement of monetary instruments.

Seventh, measures to restore a stable banking system are an important element to stimulate the investment activities of the private sector. Enhancing prudential regulation and restoring adequate supervision of the financial sector is required not only to reassure potential investors, but also to avoid the accumulation of wealth by fraud and the financing of war by illegal means, as indicated by Addison, Murshed, and Le Billon (2000).

Finally, with regard to structural reforms, there is broad agreement that simplicity is desirable, that reform efforts should not be overly ambitious, and that the often limited administrative capacity of postconflict countries must be taken into account. Initial reforms should focus on measures to strengthen good governance; improve expenditure management, notably in the area of public procurement; enhance the regulatory framework to attract investment; and improve public services that facilitate development of the private sector.