2 1688 DEVELOPMENTBRiEF Number 47 The World Bank January 1995 Economic consequne of the transition from civil wa rdto pea eooThe P Evidence from Africa-especially Ethiopia and Uganda- sheds light on why, when a civil war ends, it takes time to claim a peace dividend ivil wars differ from inter- national wars: they are in- formal, often having no clear beginning and end, they weaken rather than strengthen the authority of the state, and they leave two opposing armies to be de- mobilized within one territory. Per- haps most important, they erode the institutions of civil society, such as respect for the law. But what happens when they end? What are the economic consequences of the transition to peace? Is there a peace dividend? Civil wars vary greatly in the amount of economic damage they do. Nigeria's relatively short civil war appears to have caused only a small loss of output, and much of this loss was recovered after three years of peace. By contrast, Uganda's long and intermittent civil war caused a drastic decline in that country's output: by the end of the war GDP was half what it would have been had there been no war. Angola and Mozambique suf- fered similarly dramatic declines.* 'For more details, see Jean-Paul Azam, David Bevan, Paul Collier, Stefan Dercon, Jan Gunning, and Sanjay Pradhan, 'Some Economic Consequences of the Transition from Civil Wrto Peace,"PolicyResearchWorking Paperl1392, World Bank, Washington. DC, 1994. These economies emerging from long, destructive civil wars might be expected to face a large potential peace dividend, but in fact, postwar recoveries have been only partial. Even eight years after peace Uganda's economy remains far There I IS rarely a substantial peace dividend in the short run below its prewar peak. The private sector remained wary after the war: it neither invested nor consumed as might have been expected had it an- ticipated recovery with confidence. The fiscal implications of peace also are discouraging. Because civil wars often do not end decisively, military expenditure cannot easily be reduced. In Uganda military spending actually rose 40% in the early years of peace. Further, rev- enue may not be buoyant. In war- time Ethiopia the state used coercive revenue-raising measures that had to be abandoned during peace, causing revenue to decline, In Uganda revenue collapsed dur- ing the war and was very slow to recover. Moreover, during the last years of a civil war the fiscal posi tion may deteriorate to a level that is unsustainable. post-civil war government therefore inherits a pressing need for fiscal retrenchment. The only revenue relief lies in the enhanced flow of aid that is typically triggered by the ending of hostilities, which can help moderate the amount of fiscal adjustment required. In the short run, then, there is rarely a substantial peace dividend for the economy or for the government. Reestablishing security The end of a civil war does not automatically lead to the end of insecurity. Indeed, the fear of personal violence and theft may increase in the aftermath of a war. Demobilization is sometimes accompanied by a rise in criminality. But a well- planned demobilization program that provides compensation to the demobilized soldiers, as in Uganda, can help prevent increased insecurity. Uganda actually experienced a net reduction in crime because most of the demobilized soldiers had access to land and the demobilization of men with military training appears to have discouraged existing criminals. But in such countries as Angola, Chad, and Mozambique the key phases of a successful transition- assembly, disarmament, discharge, demobilization, and reintegration-were poorly planned. The release of unprepared and often armed soldiers led to a significant increase in crime, rural banditry, and circulation of weap ons in these countries. The fear that the state may be overthrown also persists during the transition. There often is a continu- ing risk that the war will be re- sumed. Governments can reduce military expenditures in the after- math of civil war, but to prevent an increase in personal or state-level insecurity they may need to make offsetting increases in expenditure in the form of compensation to the losers. Accelerating the return to the market Civil war puts assets and transac- tions in jeopardy. How much a par- ticular sector of the economy is affected depends on how transac- tion-intensive it is and on how vul- nerable and visible its assets are. During Uganda's civil war, be- tween 1971 and 1986, the manufac- turing sector, both transaction- intensive and asset-vulnerable, shrank by half relative to GDP (which itself declined by half). A1- though manufacturing recovered after 1986, by 1994 it was still far below its 1971 share of GDP. The sector that contracted most severely during the war was the asset- producing sector, construction. But construction was also the most suc- cessful sector in the postwar pe- riod. Commerce and transport, a transaction-providing sector, also suffered severely during the war. Although this sector grew more rapidly than GDP after the war, it remained a much smaller share of GDP than before the war. The sec- tor that did relatively well during the war was subsistence agricul- ture. Relative to GDP, subsistence grew during the war and con- tracted after peace was restored. The postwar recovery involves a return to the market and a restora- tion of confidence in assets. The government can encourage private agents to shift from subsistence ac- tivities back into the market by set- ting both explicit taxation and implicit (inflation) taxation of trans- actions low in the early postwar years. This means that the govern- ment should keep inflation below the revenue-maximizing rate. During the transition back to the market the return on the rehabilita- tion of transport infrastructure is likely to be high, as shown by the very high rates of return on road re- habilitation projects in Uganda dur- ing its transition. Aid can play a critical role in accelerating the The peace dividend comes from a adual recove graua ry of confidence return to the market by permitting a much more rapid rehabilitation of roads and other market infrastruc- ture than would be feasible if the state relied only on taxation. Reassuring private investors During civil war private profits can be high. As wartime conditions dis- rupt markets, competition declines and marketing margins widen. As long as trade is not eliminated, these wider margins can offset re- ductions in the volume of trade enough to sustain or even increase profits. And once the state has aban- doned its role of impartial defender of property rights, there are oppor- tunities to acquire assets by force. Thus, during civil war a significant group of private agents is likely to accumulate substantial assets. At the same time the uncertainty, vulnerability of physical assets, and high and variable inflation tax that prevail during civil wars lead to substantial capital flight. During the transition from war to peace a key imperative for the government is to take actions to reassure wary private agents and encourage them to repatriate this capital and to make irreversible investments. The emphasis therefore needs to be on the early sequencing of investment-sensitive reforms, such as temporary undervaluation of the exchange rate, the preservation of low inflation using direct targeting of the price level, sale of expropriated housing stock, and restraint in revenue collection. Restoring social capital During a prolonged civil war the institutions of civil society and the trust that are essential to the functioning of the economy are eroded. The economy loses "social capital." Like other types of capital, social capital takes time to restore: in the aftermath of a civil war private agents fear both each other and the government. This fear is perhaps an even greater obstacle to a private sector-led recovery than physical damage to infrastructure, because it delays irreversible investment even when financing is available. Thus, the peace dividend comes not from a swift resumption of activities disrupted by the war, but from a gradual recovery of confidence that induces repatriation of financial and human capital. During the transition, then, the most important needs are to rebuild the institutions of civil society and to boost confidence by early sequencing of investment-sensitive reforms. DevelopmentBriefsare issued by the World Bank to inform the media, business, academic, and government policy communities about development policy analyses and results from the Bank's research activities. Theyaredrawn from the work ofindividual Bank researchers and do not necessarily represent the views of the World Bank and its member countries-and should not therefore be attributed to the World Bank or its affiliates. Briefs are issued periodicllyby the Research Advisory Staff, Development Eonomics Vice Presidency, The World Bank, 1818 H Street, NW, Washington, DC 20433. Tel: (202)473-3984, Fax: (202)477-0955. Briefsare not copyrighted and may be reproduced with the appropriate attribution.