THE CONFLICT TRAP 31 the explanation,we hit a further puzzle,because although the risk of vio- lence jumps sharply if there is at least one oil well,if there are two oil wells in the district it starts to go down again.And with twenty oil wells it is lower still.That is odd because the environmental damage is presum- ably roughly proportional to the number of oil wells.To my mind this looks more like a story of a protection racket than outrage provoked by environmental damage.In the absence of an oil well there is no scope for extortion,and so no violent protest.With an oil well,the protection racket is in business.But the more oil wells there are in the district the greater the incentive for the oil company to pay up and buy peace. I do not want to overstate these results,for the disputes in the delta started out as justified environmental protests by people living in a region that was bearing the brunt of damage without seeing the benefits of oil revenues.But over time the situation has evolved.There is now a huge amount of money being directed by the Nigerian federal government to the delta region,and the oil companies are desperately spreading protec- tion money-paying ransoms to free kidnapped workers is pretty well a daily occurrence.Within the region local politicians are fighting it out for control of all this money,and violent protest has become an orchestrated part of this political rent seeking.Grievance has evolved,over the course of a decade,into greed. Let us get back to the costs of conflict.Many of the costs are borne by neighboring countries.Diseases don't respect frontiers,and the economic collapse also spreads.Since most countries are bordered by several others, the overall cost to neighbors can easily exceed the cost to the country it- self.And the costs are not limited to the immediate geographic region. Ninety-five percent of global production of hard drugs,for example,is from conflict countries.There is a straightforward explanation:conflict generates territory outside the control of a recognized government,and this comes in handy if your activity is illegal.Osama bin Laden chose to locate in Afghanistan for the same reason.So countries in civil war have what might be called a comparative advantage in international crime and terrorism.AIDS probably spread through an African civil war:the combi- nation of mass rape and mass migration produces ideal conditions for spreading sexually transmitted disease.Consequently,wars in the bottom billion are our problem as well
the explanation, we hit a further puzzle, because although the risk of violence jumps sharply if there is at least one oil well, if there are two oil wells in the district it starts to go down again. And with twenty oil wells it is lower still. That is odd because the environmental damage is presumably roughly proportional to the number of oil wells. To my mind this looks more like a story of a protection racket than outrage provoked by environmental damage. In the absence of an oil well there is no scope for extortion, and so no violent protest. With an oil well, the protection racket is in business. But the more oil wells there are in the district the greater the incentive for the oil company to pay up and buy peace. I do not want to overstate these results, for the disputes in the delta started out as justified environmental protests by people living in a region that was bearing the brunt of damage without seeing the benefits of oil revenues. But over time the situation has evolved. There is now a huge amount of money being directed by the Nigerian federal government to the delta region, and the oil companies are desperately spreading protection money—paying ransoms to free kidnapped workers is pretty well a daily occurrence. Within the region local politicians are fighting it out for control of all this money, and violent protest has become an orchestrated part of this political rent seeking. Grievance has evolved, over the course of a decade, into greed. Let us get back to the costs of conflict. Many of the costs are borne by neighboring countries. Diseases don’t respect frontiers, and the economic collapse also spreads. Since most countries are bordered by several others, the overall cost to neighbors can easily exceed the cost to the country itself. And the costs are not limited to the immediate geographic region. Ninety-five percent of global production of hard drugs, for example, is from conflict countries. There is a straightforward explanation: conflict generates territory outside the control of a recognized government, and this comes in handy if your activity is illegal. Osama bin Laden chose to locate in Afghanistan for the same reason. So countries in civil war have what might be called a comparative advantage in international crime and terrorism. AIDS probably spread through an African civil war: the combination of mass rape and mass migration produces ideal conditions for spreading sexually transmitted disease. Consequently, wars in the bottom billion are our problem as well. THE CONFLICT TRAP 31
32 THE TRAPS All in all,the cost of a typical civil war to the country and its neighbors can be put at around $64 billion.In recent decades about two new civil wars have started each year,so the global cost has been over $100 billion a year,or around double the global aid budget.This is obviously only a ballpark figure,although in building it we have erred on the side of cau- tion.Nevertheless,this sort of cost estimate can be useful.It is a critical step in valuing the benefits of interventions.As you will see in Part 4, there is a range of interventions that can cut the risk of civil war.In any one instance it is impossible to value the benefits.However,using variants of our model,it is possible to work out how much,on average,a particu- lar type of intervention will reduce the risk,and so reduce the global inci- dence of civil war.By combining this reduction in the incidence of war with our estimate of the cost of war we deduce a benefit for the interven- tion.Once this benefit is combined with the cost of the intervention,we have arrived at the cost-benefit analysis. Cost-benefit analysis is the basis of how governments make decisions on public spending.If we can get interventions to reduce the risk of civil war into this conventional framework of public decision making,we can escape the world of political make-believe-the posturing fantasies to which politicians resort when unrestrained by evidence.That is ultimately the agenda of Part 4. The Conflict Trap Now we reach the aspect of civil war that is crucial for the thesis of this book:it is a trap.Suppose a country starts its independence with the three economic characteristics that globally make a country prone to civil war: low income,slow growth,and dependence upon primary commodity ex- ports.It is playing Russian roulette.That is not just an idle metaphor:the risk that a country in the bottom billion falls into civil war in any five-year period is nearly one in six,the same risk facing a player of Russian roulette.The country may be lucky and grow its way out of the danger zone before it gets caught.Growth directly helps to reduce risk;cumula- tively it raises the level of income,which also reduces risk,and that in turn helps to diversify the country's exports away from primary com- modities,which further reduces risk.But it may not be so lucky.Suppose
All in all, the cost of a typical civil war to the country and its neighbors can be put at around $64 billion. In recent decades about two new civil wars have started each year, so the global cost has been over $100 billion a year, or around double the global aid budget. This is obviously only a ballpark figure, although in building it we have erred on the side of caution. Nevertheless, this sort of cost estimate can be useful. It is a critical step in valuing the benefits of interventions. As you will see in Part 4, there is a range of interventions that can cut the risk of civil war. In any one instance it is impossible to value the benefits. However, using variants of our model, it is possible to work out how much, on average, a particular type of intervention will reduce the risk, and so reduce the global incidence of civil war. By combining this reduction in the incidence of war with our estimate of the cost of war we deduce a benefit for the intervention. Once this benefit is combined with the cost of the intervention, we have arrived at the cost-benefit analysis. Cost-benefit analysis is the basis of how governments make decisions on public spending. If we can get interventions to reduce the risk of civil war into this conventional framework of public decision making, we can escape the world of political make-believe—the posturing fantasies to which politicians resort when unrestrained by evidence. That is ultimately the agenda of Part 4. The Conflict Trap Now we reach the aspect of civil war that is crucial for the thesis of this book: it is a trap. Suppose a country starts its independence with the three economic characteristics that globally make a country prone to civil war: low income, slow growth, and dependence upon primary commodity exports. It is playing Russian roulette. That is not just an idle metaphor: the risk that a country in the bottom billion falls into civil war in any five-year period is nearly one in six, the same risk facing a player of Russian roulette. The country may be lucky and grow its way out of the danger zone before it gets caught. Growth directly helps to reduce risk; cumulatively it raises the level of income, which also reduces risk, and that in turn helps to diversify the country’s exports away from primary commodities, which further reduces risk. But it may not be so lucky. Suppose 32 THE TRAPS
THE CONFLICT TRAP 33 that for one reason or another growth stays slow.(I will be looking at why the countries of the bottom billion have failed to grow in the following chapters.)Then the peace might not last long enough to bring risks down before the chamber with the bullet in it comes around and the country slides into civil war.That is basically what has happened:the unlucky countries got war shortly after independence,as in the case of Nigeria, while others maintained peace for many years and then succumbed,as did Cote d'lvoire,which was destabilized by a coup d'etat,and Nepal,where Maoists were confronted by a fratricidal monarchy.All have been living dan- gerously.Sooner or later some combination of personalities and mistakes that in a more economically successful country would be brushed aside es- calates into rebellion.Call the personalities and mistakes the"causes"if you must.I think that in such fragile societies it is generally even harder to avoid these triggers than it is to develop the economy.Persuading everyone to be- have decently to each other because the society is so fragile is a worthy goal, but it may be more straightforward just to make the societies less fragile, which means developing their economies.How we can help these countries to do that is the agenda for Part 4. Once a war has begun,the economic damage undoes the growth achieved during peace.Worse,even aside from this economic damage the risk of further war explodes upward.Civil war leaves a legacy of organized killing that is hard to live down.Violence and extortion have proved prof- itable for the perpetrators.Killing is the only way they know to earn a liv- ing.And what else to do with all those guns?Currently one of my gradu- ate students,Phil Killicoat,is trying to collect data on the price of a Kalashnikov around the world year by year,the Kalashnikov being the weapon of choice for any self-respecting rebel.That is the sort of innova- tion in data that would make a real contribution to work on conflict.It is not an easy task,which is why until now nobody has done it,but he is re- sourceful.The emerging pattern seems to be that guns become cheap dur- ing conflict because so many get imported through official and semioffi- cial channels that a proportion of them leak onto the informal market. The legacy of conflict is cheap Kalashnikovs. Anke Hoeffler and I looked to see what happens to the crime rate in postconflict societies.Crime is one of the phenomena that are very badly measured:countries differ massively in their definitions and in the degree
that for one reason or another growth stays slow. (I will be looking at why the countries of the bottom billion have failed to grow in the following chapters.) Then the peace might not last long enough to bring risks down before the chamber with the bullet in it comes around and the country slides into civil war. That is basically what has happened: the unlucky countries got war shortly after independence, as in the case of Nigeria, while others maintained peace for many years and then succumbed, as did Côte d’Ivoire, which was destabilized by a coup d’état, and Nepal, where Maoists were confronted by a fratricidal monarchy. All have been living dangerously. Sooner or later some combination of personalities and mistakes that in a more economically successful country would be brushed aside escalates into rebellion. Call the personalities and mistakes the “causes” if you must. I think that in such fragile societies it is generally even harder to avoid these triggers than it is to develop the economy. Persuading everyone to behave decently to each other because the society is so fragile is a worthy goal, but it may be more straightforward just to make the societies less fragile, which means developing their economies. How we can help these countries to do that is the agenda for Part 4. Once a war has begun, the economic damage undoes the growth achieved during peace. Worse, even aside from this economic damage the risk of further war explodes upward. Civil war leaves a legacy of organized killing that is hard to live down. Violence and extortion have proved profitable for the perpetrators. Killing is the only way they know to earn a living. And what else to do with all those guns? Currently one of my graduate students, Phil Killicoat, is trying to collect data on the price of a Kalashnikov around the world year by year, the Kalashnikov being the weapon of choice for any self-respecting rebel. That is the sort of innovation in data that would make a real contribution to work on conflict. It is not an easy task, which is why until now nobody has done it, but he is resourceful. The emerging pattern seems to be that guns become cheap during conflict because so many get imported through official and semioffi- cial channels that a proportion of them leak onto the informal market. The legacy of conflict is cheap Kalashnikovs. Anke Hoeffler and I looked to see what happens to the crime rate in postconflict societies. Crime is one of the phenomena that are very badly measured: countries differ massively in their definitions and in the degree THE CONFLICT TRAP 33
34 THE TRAPS of underreporting.For that reason we settled on the homicide rate as the proxy for violent crime in general.Homicide is the best-defined violent crime and it is also likely to be the best-reported.We found that political peace does not usher in social peace.The end of the political fighting ush- ers in a boom in homicides.Presumably,this is part of a wider surge in violent crime.Add in mutual distrust and recriminations over atrocities, and it is not surprising that the typical postconflict country has little bet- ter than a fifty-fifty chance of making it through the first decade in peace. Indeed,about half of all civil wars are postconflict relapses. A country such as the Democratic Republic of the Congo (formerly Zaire)will need around half a century of peace at its present rate of growth simply to get back to the income level it had in 1960.Its chances of getting fifty continuous years of peace with its low income,slow growth,depend- ence upon primary commodities,and history of conflict are,unfortunately, not high.This country is likely to be stuck in a conflict trap no matter how many times it rebrands itself unless we do something about it. Poverty,economic stagnation,dependence on primary commodities- do these characteristics sound familiar?Yes,they are endemic to the bot- tom billion.This does not mean that all such countries are in the conflict trap,but they are all prone to it.We have,in fact,the building blocks for a system.The risk of conflict differs according to economic characteristics, and the economic characteristics are affected by conflict.It is possible to set up this interaction as a model that predicts in a stylized fashion how the incidence of conflict is likely to evolve.I joined forces with Harvard Hegre,a young Norwegian political scientist,and we built one.The world, as modeled,starts in 1960 with three different groups of countries:rich, bottom billion,and developing.We then see how many countries fall into conflict.The predictions rest upon the risks generated by the analysis I had already done with Anke,as well as assumptions about growth perfor- mance that extrapolate from the past forty years of experience.We project the incidence of conflict through until 2020 and even,somewhat fanci- fully,to 2050.Rich countries have such a low risk of civil war that even over such a long period none gets into trouble.A few of the developing countries stumble into civil war,and those that do get derailed for a while-examples of these are countries such as Colombia and Lebanon
of underreporting. For that reason we settled on the homicide rate as the proxy for violent crime in general. Homicide is the best-defined violent crime and it is also likely to be the best-reported. We found that political peace does not usher in social peace. The end of the political fighting ushers in a boom in homicides. Presumably, this is part of a wider surge in violent crime. Add in mutual distrust and recriminations over atrocities, and it is not surprising that the typical postconflict country has little better than a fifty-fifty chance of making it through the first decade in peace. Indeed, about half of all civil wars are postconflict relapses. A country such as the Democratic Republic of the Congo (formerly Zaire) will need around half a century of peace at its present rate of growth simply to get back to the income level it had in 1960. Its chances of getting fifty continuous years of peace with its low income, slow growth, dependence upon primary commodities, and history of conflict are, unfortunately, not high. This country is likely to be stuck in a conflict trap no matter how many times it rebrands itself unless we do something about it. Poverty, economic stagnation, dependence on primary commodities— do these characteristics sound familiar? Yes, they are endemic to the bottom billion. This does not mean that all such countries are in the conflict trap, but they are all prone to it. We have, in fact, the building blocks for a system. The risk of conflict differs according to economic characteristics, and the economic characteristics are affected by conflict. It is possible to set up this interaction as a model that predicts in a stylized fashion how the incidence of conflict is likely to evolve. I joined forces with Harvard Hegre, a young Norwegian political scientist, and we built one. The world, as modeled, starts in 1960 with three different groups of countries: rich, bottom billion, and developing. We then see how many countries fall into conflict. The predictions rest upon the risks generated by the analysis I had already done with Anke, as well as assumptions about growth performance that extrapolate from the past forty years of experience. We project the incidence of conflict through until 2020 and even, somewhat fancifully, to 2050. Rich countries have such a low risk of civil war that even over such a long period none gets into trouble. A few of the developing countries stumble into civil war, and those that do get derailed for a while—examples of these are countries such as Colombia and Lebanon, 34 THE TRAPS
THE CONFLICT TRAP 35 which are not part of the bottom billion but for one reason or another have been unlucky.The bulk of the countries that fall into civil war are from the bottom billion.Periodically they get back to peace,but often they fall back into conflict.The model is useless for telling us which coun- tries will be in conflict,but its prediction as to how the global incidence of conflict evolves is depressing.By 2020 the world is much richer than today, and by 2050 it is fabulously richer:most countries are developed.But the incidence of civil war declines only modestly because most civil war is gen- erated by the minority of countries in the bottom billion,and their growth is slow.Our model quantifies the grim implications of the failure of the growth process in the bottom billion,given the link between poverty,stag- nation,and conflict. Coups Rebellion is not the only form of violent,illegitimate challenge to govern- ments in the countries of the bottom billion.Many governments are more at threat from coups than from rebellions.You might have thought that coups had died out;your image of a coup is likely to involve a Latin American general from the 1960s.There is some justice to that image,as coups have largely gone out of fashion and outside the bottom billion they are now very rare.But among the bottom billion they are still depressingly common.As of December 2006 the latest successful one had occurred just two weeks earlier,in Fiji.Coups are not as disastrous as civil wars;to adapt the famous newspaper headline about an earthquake,this event might have been reported as"small coup in Fiji,not many dead."But they are not a very good way of changing a government.The political instabil- ity that they manifest is known to be detrimental to economic develop- ment.So what causes coups? We drew upon the data of an American political scientist who had as- siduously trawled through thousands of pages of newspaper reports to produce a comprehensive list of all the reported coup plots,failed coup attempts,and successful coups in Africa,and we also found data on all the successful coups in other parts of the world.We followed broadly the same approach that we had taken in understanding civil wars.I should
which are not part of the bottom billion but for one reason or another have been unlucky. The bulk of the countries that fall into civil war are from the bottom billion. Periodically they get back to peace, but often they fall back into conflict. The model is useless for telling us which countries will be in conflict, but its prediction as to how the global incidence of conflict evolves is depressing. By 2020 the world is much richer than today, and by 2050 it is fabulously richer: most countries are developed. But the incidence of civil war declines only modestly because most civil war is generated by the minority of countries in the bottom billion, and their growth is slow. Our model quantifies the grim implications of the failure of the growth process in the bottom billion, given the link between poverty, stagnation, and conflict. Coups Rebellion is not the only form of violent, illegitimate challenge to governments in the countries of the bottom billion. Many governments are more at threat from coups than from rebellions. You might have thought that coups had died out; your image of a coup is likely to involve a Latin American general from the 1960s. There is some justice to that image, as coups have largely gone out of fashion and outside the bottom billion they are now very rare. But among the bottom billion they are still depressingly common. As of December 2006 the latest successful one had occurred just two weeks earlier, in Fiji. Coups are not as disastrous as civil wars; to adapt the famous newspaper headline about an earthquake, this event might have been reported as “small coup in Fiji, not many dead.” But they are not a very good way of changing a government. The political instability that they manifest is known to be detrimental to economic development. So what causes coups? We drew upon the data of an American political scientist who had assiduously trawled through thousands of pages of newspaper reports to produce a comprehensive list of all the reported coup plots, failed coup attempts, and successful coups in Africa, and we also found data on all the successful coups in other parts of the world. We followed broadly the same approach that we had taken in understanding civil wars. I should THE CONFLICT TRAP 35