Click here for search results
Online Media Briefing Cntr
Embargoed news for accredited journalists only.
Login / Register

Breaking the Conflict Trap: Civil War and Development Policy


Paris, France, May 14, 2003


Also available:
RealVideo of the Press Conference
in high resolution and low resolution versions

Zoltàn Agai - Good morning to you all and thank you very much for coming.  We know how difficult it was this morning to get here.  We have to apologise for Nicholas Stern's absence.  He was unable to be here due to traffic and a very important working meeting that was not scheduled until this morning.  Paul Collier and Ian Bannon will be making presentations.  Paul Collier used to be the Director of the Research Department at the World Bank until recently.  He is now a Professor at Oxford University and remains an Adviser to the Vice President for the Africa region at the World Bank.  Paul Collier will make some introductory comments on the Report we are presenting.  Ian Bannon will then present the work of the World Bank in the fields of conflict prevention and reconstruction.  Ian Bannon is Manager of the World Bank's Conflict Prevention and Reconstruction team. 

Paul Collier - Thank you very much.  I will spend a few minutes introducing the main messages from the Report.  There are two main messages.  One, as an international community we cannot afford to leave civil wars to the participants.  The damage done by civil wars spills over to the non-combatants both within the country and outside the country.  It spills over in a major way, leading to high global costs.  In other words, civil wars are very bad news and we cannot leave it to the combatants to decide whether they will occur.  That is message one: the international community has to get involved if it is to avoid the very high costs of conflict.  The second message is that the international community actually has many options in doing something about this.  The incidence of civil war globally has been rising over the past 40 years.  However, it need not have risen, and there is much that the international community could have done, and can now do, to bring that incidence down.  Those are thus the "big picture" messages.  Let me now enlarge on each of those steps.

In relation to costs, I think of the costs of conflict as having three ripples or three layers.  The first layer refers to the costs that arise within the country.  As soon as you think of civil war, you think of people killing each other.  Obviously that does happen but it is only a small fraction of the deaths, injuries and misery brought about within the country by civil war.  Most of the deaths are non-combatants because most of the deaths arise through a collapse in health.  Civil war produces flight: great movements of people flee combat.  In the process of fleeing, they pick up diseases in areas where they do not have immunity; they carry those diseases; and health services break down, with preventative health services one of the first areas to be undermined.  Many of these costs continue even when the civil war ends.  Typically, civil wars last a very long time - five to seven years - which is much longer than international wars.  However, even when they are over, many of the costs continue.  For example, typically about half the mortality costs of a war occur after it is over because health conditions stay bad for years.  That is the first ripple of costs within the country: suffering borne by non-participants.  Often participants themselves do well out of the war, and I will come to this point later.

The second ripple occurs across the region.  The regions in which these conflicts occur suffer major health and economic costs.  For example, millions of refugees move into neighbouring countries.  On average, each 1 000 refugees moving across the border into another country increases malaria cases in the host country by 1 400 cases.  Unfortunately, these refugees are spreading disease as they move.  There are also economic costs.  Most obviously, a neighbourhood gets a "bad reputation"; it is seen as a bad neighbourhood.  And these costs persist. 

The final ripple, which may well be the largest one, is the global ripple.  In the Report, we provide three examples of huge social costs that could easily be attributed to civil wars: hard drugs, AIDS, and international terrorism.  The best evidence can be found in relation to hard drugs.  95% of hard drugs production now occurs in civil war countries.  There is a very simple reason for this: to produce hard drugs on a large scale, you need territory outside the control of a recognised government, and that is what civil wars supply.  Classic examples would be Colombia or Afghanistan.  This is thus the first global phenomenon directly linked to the phenomenon of civil war. 

The second is AIDS.  You may have seen on the news yesterday a suggestion that the epidemiology of AIDS is linked to civil wars.  It is not unlikely because one of the things produced by civil wars is the conjunction of mass rape and mass flight, which spread infection very fast. 

The final and quite obvious global public bad is international terrorism.  For example, why did Al-Quaida members choose to locate themselves in Afghanistan given that they were not Afghanis?  They chose to locate in Afghanistan because it was territory outside the control of a recognised government.  The Talibans, a recently successful rebel group that had captured most of the country was not recognised. 

There are thus three layers of costs.  Is there anything we can do about them?  In fact, there is much that we can do.  First of all, the risk of civil war is concentrated in a minority of countries.  It is associated with a few major economic drivers.  We think of civil wars as acutely political or idiosyncratic.  However, in fact, there are patterns, and the patterns are predominantly economic.  There are three major characteristics that make a country at risk of conflict.  First, if the country has low per capita income.  Second, if the country is in economic decline.  Third, if the country is dependent on natural resources.  If it has all three of those, a country is playing Russian roulette.  There are about 1.1 billion people living in countries that we see as either in conflict or in severe risk of conflict.  That represents about one-sixth of the world's population.  However, it is responsible for about five-sixths of the world's civil wars. 

In that sense, it is a manageable problem because if we focus our efforts on the people in this bottom billion group, we can have a much bigger impact.  What impact do we need?  There are two major directions.  First, to promote economic development much more successfully than we have in the past in this bottom billion.  During the 1990s, development missed this group.  The 1990s was quite a good decade for the world as a whole in economic terms.  However, it was not a good decade for this bottom billion: on average, per capita incomes were falling by 1% per year.  Therefore, even in the best decade we have had for a while, economic circumstances were dreadful for this group, which had low incomes and falling incomes.  We therefore need a major focused development effort and not just business as usual.  We need more aid that is better targeted, and better reform efforts to try and get growth in these countries. 

However, that is just part of the story.  The other part focuses on the natural resources risk.  This is the part we are particularly pushing at the moment in time for the G8 meeting.  The reason that the Report is being issued now is that it is specifically timed to influence G8 discussions.

There is a strong link between dependence on natural resources and the risk of civil war and, more generally, of failed development.  This link is due to a number of reasons.  First, natural resources help finance rebel groups.  Spectacular examples include UNITA with diamonds, or the FARC with drugs.  Sometimes this occurs through direct looting of the natural resources themselves; sometimes this occurs through extortion rackets or kidnapping and ransoms of the workers in the industry.  This is the most common way in which rebel groups gain access to oil income, for example.  Another route is that, in the past, the significant revenues from natural resources have tended to be very badly used by governments.  This produces resentment.  In particular, it produces especially intense resentment in the regions of the country that have the natural resource.  It thus encourages secessionist pressures.  There are small, romantic groups all over the world that are urging secession.  Usually it does not come to anything.  Every now and then, a small romantic groups finds that it is sitting on top of a very valuable natural resource.  The language can then change.  Instead of just saying, "we are different" they say "if we were independent, we would be rich".  For example, the Gam in Aceh (Indonesia) had been around for over 100 years.  Once oil was discovered, they began to say that they could be as rich as Brunei.  In fact, this was not true; it was a massive exaggeration.  However, this is the sort of thing secessionist groups will do. 

Let us now come to some specific policies that can address this link between natural resources, conflict risk, and development in reverse.  We are suggesting that the G8 should think of a package of policies.  The first part of the package consists of transparency and scrutiny: transparency of revenues and scrutiny of expenditures.  You will recognise that the transparency agenda has already been heavily promoted by one member of the G8, that is, the UK.  This is Tony Blair's transparency agenda.  We believe that, by itself, this is too narrow, and I think the UK would agree with this and they have themselves broadened to scrutiny.  Scrutiny is an agenda of particular concern to the United States: how can we empower people domestically to be able to scrutinise how those revenues are used?

The original push for transparency came from the NGO, Global Witness, and George Soros.  Their original idea was to force the international oil companies to make payments transparent.  I think we have moved on from there.  Our thinking now is that the best approach is to have governments of resource-rich countries volunteer to sign up to a template of good governance.  Some will; some will not.  However, it gives governments that have bad reputations and want to start again, the chance to say that the future will not be like the past.  The sort of template they would be signing would be a mixture of requiring companies to report revenues in their country, and setting up domestic institutions for the scrutiny of how those revenues are used.  This is building block one in our package: transparency and scrutiny.

Building block two is to improve the tracking of the commodities themselves, and the revenues from them in order to try and keep those revenues out of the hands of rebel groups.  Major progress has already been made in this area with the Kimberly process on diamonds, which now has a certification procedure for tracking diamonds.  It will not literally keep rebel groups out of a diamonds market.  However, it will force a deep discount on the price that rebels can receive.  We carried out a small study in the Report and we found that where commodity prices are lower, conflicts are shorter.  Thus, if we can depress the price that rebels are getting for their natural resource looting, we believe that this would shorten on-going wars.  That agenda of going for the financial jugular of rebel groups is popular with the governments of resource-rich countries.  For example, the Angolan government does not want a UNITA II.  After 30 years of UNITA 1 they have finally attained peace and do not want this to occur again. This is thus the second building block in our package. 

The final building block in our package is to try and cushion the price shocks that have badly affected the countries dependent on natural resources.  Periodically, world commodity prices crash, producing an implosion of the economy in the countries that are heavily dependent on exports.  In turn, those implosions are periods of high risk for conflict.  When an economy is declining rapidly, this is the peak moment of risk.  In the past, the international community has been very poor in responding to these circumstances of crisis.  Its response depended on the type of crisis: in telegenic crises such as earthquakes or draughts the donor community would move in a big way.  When we had pictures on the television screen, development ministers would open their cheque books.  You could even turn a profit on it.  However, if the nature of the crisis was a price crash, which in economic terms is much more serious, there was no donor response.  We are therefore recommending that the donor community should create the aid instruments that provide a  short-term cushion to soften these crashes. 

These are thus the three elements of our package: transparency and scrutiny; tracking of commodities to keep the money out of rebel groups; and cushioning the shocks.

I should say that there are many other things we can do to reduce the risk of conflict, and the Report discusses them.  The one area I will discuss very briefly is post-conflict. The reason the Report is called the Conflict Trap is because once a country has had a conflict, there is a very high risk of conflict-repeat.  Far from civil wars producing a legacy of democracy, happiness and prosperity, they produce a vicious cycle of more conflict.  We thus have many specific recommendations on policies in post-conflict countries.  I will, however, close here.  Our main focus now is that in the next two weeks, the build up to the G8 actually internalises this package of proposals for breaking the link between natural resources and development in reverse. 

Ian Bannon - Paul Collier represents the part of the World Bank that carries out research.  I represent the part of the Bank that tries to do something with this research.  I therefore try to determine what this means for our own policies and activities. 

Up to about three or four years ago, the World Bank was very much about reconstruction; cleaning up the mess after a civil war.  We have now turned that around and believe that we also need to do a better job in preventing conflicts from occurring in the first case.  This has been part of that on-going effort.  On the one hand, we need to better understand the causes of conflict.  Otherwise, we mis-diagnose the conflict.  That is the research that has been carried out by Paul Collier and his team.  Our challenge now is not only to present an agenda for global action but also to be consistent in what we do internally in support of that agenda.  Let me provide you with a few examples of what we are doing internally in the Bank. 

One of the things we are doing is to look at our own policies and the kinds of interventions and projects that we finance in extractive industries.  This is an independent review being carried out in consultation with civil society and our independent evaluation department.  We hope that this will enable us to develop better policies and better ways of supporting extractive industries.  We are very much aware of the connection between poverty and conflict.  We are working to make poverty strategies take account of the things that can either prevent or contribute to conflict.  That is, we are working to better integrate conflict aspects into the development agenda.  We are also encouraging our teams to carry out conflict analysis before they embark on country strategies, reform programmes or projects.  We believe that, in some cases, good policies lead to growth and a reduction in poverty.  However, there may be instances where policy reform or policy recommendations could have unanticipated negative consequences on the drivers of conflict.  For example, if they support one ethnic group over another, or if they favour policies that some groups object to.  We are therefore passing projects and strategies through a conflict filter to see whether we are taking account of some of these conflict drivers.  Hopefully, if we notice such a matter, we can change it. 

Other matters we are emphasising include our belief in the need for a stronger effort on the governance side.  The connections between natural resources and governance are a very strong and powerful risk factor.  As you all realise, oil is a resource that has very rarely, if at all, become a source of wealth.  In fact, it has become more of a curse than an asset for developing countries.  In the spirit of this template, we believe that especially for new oil producers such as Timor, parts of West Africa and particularly Sudan, we need to work with these countries to get them to buy into this template in order that oil being a wealth and not a curse.

Finally, by way of another example, we have taken a close look at how our aid patterns function.  Some of the Report findings suggest that aid needs to be increased in post-conflict countries.  We have therefore made special allocations for countries coming out of aid.  We are also concerned about the pattern of aid.  Research has shown that the critical years for post-conflict countries are not the initial years which attract all the headlines.  At that time, the aid community rushes in with amounts of money and much frustration and criticism arises because all the funds cannot be spend.  For example, why can the highway in Afghanistan not be built?  It cannot be built because the country does not have the absorbing capacity.  In most cases, it will have that capacity in years four and five and not in years one and two.  We are therefore looking at the way in which we disperse and commit aid to better match the way capacity in these countries evolves.  We do much more work at the sub-sectoral level, working with communities to try and reinforce the community's ability to manage conflict, whether it be through the education system, community development and those types of projects. 

Zoltàn Agai - Thank you very much.  We are now going to open the session to questions.  Please introduce yourselves before asking your questions.

Genc Myftiu, Sustainable Development Agency, Albania - The presentation was very interesting in terms of theory and practice.  Are there any commonalities between African cases and Asian or Eastern European ones?  The stage of development and conflict in these areas could be similar but there could also be particularities.  The most important factor is the diagnosis.  What do you think of the particularities of the regions?  Does this influence the way in which the World Bank or other organisations would handle the situation?  I am particularly concerned by the situation in my region.  We have a source of conflict with Macedonia, a border country containing some Albanian ethnics.  My perception is that this case represents a major difference from Sudan or Afghanistan.  How can the World Bank address these two cases?

Paul Collier - Pages 112 to 115 of the Report contains pictures that show the trends region by region over the past forty years.  These regional patterns have been quite different.  First of all, we should say that every civil war is different, and each contains its own idiosyncrasies.  Therefore, we cannot just develop one grand general theory, and we always have to look at the particularities.  However, the patterns we have suggested do seem to be global ones.  That is, there are three major drivers of conflict across all regions: poverty (low per capita income), economic decline, and natural resource dependence.  They are not the end of the story but they are certainly the beginning of the story.  This plays in to the regional differences.  For example, over the last 20 to 30 years, the incidence of civil war has tended to rise in Africa and has tended to fall in other regions, on average.  This is related to economic developments.  Over the last 30 years, Africa has missed the boat on economic development.  If anything, it is poorer now than it was 30 years ago, and large parts of Africa have negative growth rates, and it is at least as independent now on primary commodities as it was 30 years ago.  In contrast, large tracts of the rest of the developing world have broken out of primary commodity dependence, and have broken into manufactured exports.  Large parts of the non-African developing world have raised per capita incomes, albeit in a quite bouncy way in some cases.  There are thus major differences in regional patterns. 

However, the economic story is not the end of it, and there is also a political story to be told.  That is, when countries first obtain independence, they have a period of political fragility: their political institutions are new and challengeable.  They thus tend to go through a fairly high risk phase.  Those phases of independence occur at different stages in different regions.  For example, the break up of the Soviet empire occurred much later than the break up of the French and British empires.  We thus have different timetables for these political instability factors in different regions.

I should add that there are two people in the in the room who are co-authors of the Report: Håvard Hegre, who worked on much of the political analysis; and Marta Reynal-Querol, who worked on much of social costs aspects of conflicts.  They are very welcome to comment on any of these matters.  Neither of them are Bank staff.

Ewa Latoszek, Warsaw School of Economics - I would like to raise the question of absorption.  You mentioned that you take into account absorption in the financing of post-conflict countries.  However, I think it is a very difficult question to resolve.  For example, it is much easier to resolve this question in the existing situation in Poland, which has quite well prepared people in the education and administration sectors.  However, it took many years to achieve this, and it will take several years before we are ready to take advantage of structural funds from the European Union. Therefore, I believe that it would not be easy to financially assist a country such as Iraq, even with specialists from abroad.  A solution may be found for two or three years, but what about afterwards?  You cannot educate people in such a short time frame given the existing situation from an educational view at least. 

Ian Bannon - This is one of the most difficult problems.  However we have learnt a certain number of things that we could be doing differently.  For example, in countries where you do not even have the basic institutions, perhaps some of the initial aid could be used to pay salaries.  This is a simple point but if officials such as police officers are not getting paid, then this is not good for institution building or absorption capacity.  Some donors have difficulty in doing this: donors, including parts of the Bank, are still caught in the mentality that you must carry out a specific project, with an official being photographed in front of something that has been funded by taxpayers' money.  However, very often in these situations, taxpayer funding of the salary of a police officer might be the best contribution you could make.  Traditional institution building is a slow, painful and careful process.  You could insist, for example, that when experts are used to provide outside technical assistance, a training component should always be included.  As you design contracts for technical assistance projects, make sure there is always a training component.  Very often, nationals who have migrated can be a tremendous help in building that institutional capacity.  Again, each country will be different.  A country such as Sri Lanka, which has a functioning state, may be very different from a country such as Afghanistan, where you are starting from scratch.  In terms of absorptive capacity, years four and five post-conflict are the ones when countries can begin to make better use of aid.  Unfortunately, this is when most donors begin to disappear and move on to the next crisis.

Subidey Togan, Ankara University, Turkey - Our region demonstrates an example of conflict.  In your presentation, you mentioned that the importance of economic factors low-income economic decline and dependence on natural resources.  When I consider the conflicts in our region, such as Iraq, these factors are important.  However, a mechanism for dispute settlement must exist.  If the parties enjoy access to a fair dispute settlement mechanism, the solution would become easier to solve.  Regarding the institutional aspect, as long as certain rules are enforced by institutions and people believe in those rules, conflicts could be solved more easily.  Without such institutions to enforce the rules, they take their own measures.  Therefore, besides economic factors, one could also emphasise the institutional aspect and judicial reform.

Paul Collier - There are many good reasons to agree with your statement.  However, it would represent a romantic delusion to think that democratic institutions alone can guarantee peace in low-income countries.  In low-income countries, although we wish it did, we find that democracy does not necessarily reduce the risk of civil war.  Basically, people have to believe in democratic institutions.  Typically, within low-income democracies, institutions remain fragile.  If a party is unhappy with the result of an election, they challenge the entire arrangement.  Although, this statement does not devalue democracy, we should not hold naive beliefs that once established, democratic institutions can survive unaided.  Especially in the early stages, these institutions are fragile.  Consequently, in conflict situations, we need economic intervention and perhaps external military intervention.  Frequently, post-conflict governments make the mistake of trying to secure their position by high military expenditure.  In our view, high domestic military expenditure increases the risk of conflict and offers a signal that the government is not committed to the peace settlement.  If any military presence is required, it should be accompanied by an external guarantor.

Ian Bannon - Part of problem is related to the fact that we focus on the civil wars that occur.  However, in similar countries, it is equally important to understand why conflict is avoided.  In a way, we need to conduct more research on why people do trust their institutions and find alternative ways of avoiding conflict.  Perhaps our next report could focus on this issue.

Agni Vlavianos Arvanitis, Biopolitics International Organisation, Greece - I find it encouraging to hear a clear statement for the prevention of civil war stressing the motivating role for economic conditions.  Thanks to technology and other possibilities of international co-operation, we hope to have found the opportunities to create a new vision where peace is more profitable than war.  Perhaps new mechanisms can create buffer zones within areas of conflict or problems to soften differences.  In this regard, he environmental and sustainability dimensions could provide unlimited opportunities.  Do you believe that it is too idealistic to believe in our ability to shift the paradigm from civil wars for profit to peace for profit?

Paul Collier - I believe that there is a real possibility of achieving this ideal.  In the report, we provide the example of Sierra Leone and Botswana.  30 years ago both countries enjoyed similar per-capita income, large diamond resources and potential.  Subsequently, Botswana used these resources to become one of the fastest growing economies in the world.  Eventually, it became a middle income country.  In Sierra Leone, the same diamond resources cause collapse into civil war environmental degradation and utter poverty.  At present, it props up the table of national human development index.  In terms of per-capita income, the gap between both countries is now 20 to 1.  This example dramatically demonstrates how much is at stake.  Sadly, we have recorded more examples like Sierra Leone than Botswana.  However, Botswana succeeded and proved that improvement was possible.  For the international community, the challenge consists of shifting the balance to create more examples like Botswana than Sierra Leone. 

The report examines several proposals that should make this more likely.  In this respect, following the example of the millennium development goals, we have identified monitorable goals.  By 2015, the world has promised to half global poverty.  IN our view, it should also promise to reduce the occurrence of civil wars by half.  Over the past 40 years, the number of civil wars has increased dramatically.  However, we believe that it is possible to half the number.  Indeed, the objective of halving the number of civil wars provides a more monitorable goal than that of halving poverty.  Civil wars are easier to count and more observable than poverty.  We need action to reduce the consequences on the ground as well as across regions and the globe.  If we succeed in raising consciousness I believe that it is feasible.  Despite recent attentions, civil wars remain the most common form of war.  Today, about 95% of all wars are civil wars.  While a typical international war lasts for a few weeks, a typical civil war lasts for a few years.  There are some very good reasons for this fact.  In practice, many measures are possible to reduce the occurrence of civil wars: let us implement them.

Zoltàn Agai - I would like to remind you of a workshop on this topic tomorrow at the ABCDE Europe conference, which will take place at the French Ministry for Finance.





Permanent URL for this page: http://go.worldbank.org/NA01QIDVS1