Previous PageTable Of ContentsNext Page

Conflicts, agriculture and food security

"A peaceful and stable environment in every country is a fundamental condition for the attainment of sustainable food security."

World Food Summit Plan of Action, paragraph 3


Armed conflict and civil strife were major sources of food insecurity in the 1990s and will continue to be this century, although their number and the losses associated with them may have passed their peak (see Figure 13). Depending on which of the various definitions of the term is used, from 30 to 40 countries were conflict-affected at the end of the twentieth century.33 Overall, hundreds of millions of people were involved. The vast majority of these people lived in low-income countries, in which agriculture represents a major source of livelihood, foreign exchange and social stability. A disproportionate number of the countries were in sub-Saharan Africa.

Figure 13

The economic losses and the disruptions to food supply and access associated with conflicts can be disastrous, especially in low-income countries where there are no effective social safety nets. While conflicts are at the basis of food insecurity situations in many countries, it is also true that raising agricultural productivity and reducing hunger and malnutrition in poor countries is an obvious path to peace. However, this is often overlooked - as was recently emphasized by the former President of the United States, Jimmy Carter.34

Deaths resulting from reduced food security and famine, i.e. long-term effects of conflicts, can exceed the deaths caused directly from violence.


For all conflicts, as for all natural disasters, the most important impacts are the suffering, injury and death of men, women and children. The losses in output, means of production and infrastructure seem insignificant in comparison. Yet these material losses are also important, for they undermine the ability of conflict survivors to subsist and recover. This is most obvious in agriculture, where the destruction of crops and livestock results, at best, in reduced food security and, at worst, in famine and death. Indeed, in many cases, deaths resulting indirectly from conflict (through famine, for example) exceed deaths from direct violence. The estimation of material losses, narrow as it is, represents an important aspect of assessing the severity of a conflict. It also provides a guide to the design of long-term policy for dealing with conflict situations.

It has been found that the indirect costs of war are typically greater than the more straightforward direct costs; and that they continue long after the end of a conflict.35 However, estimating the total direct and indirect costs of conflict is a complex procedure.36 The outcomes of war should not be assessed against a static framework, but it is hard to accommodate changes in technology, changing institutions and shifting social relations during conflicts. It is also extremely difficult to assess the economic value and employment impact of the loss of human life.

Box 4


A study conducted for FAO on the determinants of food energy consumption levels1 included estimates of the impact of war and civil strife in a number of African countries during 1971-92. The study used regression analysis to explain the evolution of per caput food energy consumption, calculated in terms of the dietary energy supply (DES) level minus its food aid component, for a number of countries that experienced situations of civil war and/or strife (data on such situations were obtained from the United States Department of Agriculture [USDA]). Civil war and/or strife was represented in the analysis by a dummy variable.

The study concluded that the impact of civil war/strife on food energy consumption varied widely according to the frequency and severity of such situations but was in some cases considerable. Thus, among the countries for which the impact of this variable was significant (i.e. at the 10 percent statistical level), the lost food energy in Uganda during each year of civil war/strife amounted to 56 kcal, or 2.5 percent of the average DES levels. The figure of lost food energy was as high as 362 kcal in Liberia (16 percent of the total), 438 kcal (20 percent) in Somalia and 120 kcal (6 percent) in Ethiopia.

1 Prepared for FAO by Professor George P. Zanias from Athens University of Economics and Business, Athens, Greece.

In this section, which focuses on developing countries, estimates of war losses are limited to the direct impact on agricultural output, with no attempt to include capital losses or indirect effects, for example, on rural non-agricultural activities.37 Direct losses may arise from various mechanisms: the migration of population from conflict-affected areas, leaving land idle and livestock abandoned; a reduction in marketed output because transport services are disrupted; the destruction of crops in the field or of post-harvest stocks and livestock by armed groups; and declines in yields resulting from lack of access to key inputs. (These are considered in more detail in the section Characteristics of agriculture and the impact of conflict).

Box 5


The last 15 years have seen a large number of food emergencies arising from natural or human-induced factors. Their number has fluctuated from about 20 to 30 in most years, but has tended to increase during the 1990s, particularly the latter part of the decade. This increase has been associated primarily with the greater incidence of conflicts. The Figure below shows, in fact, a discernible shift in the causes of food emergencies. Whereas human-induced disasters contributed to only about 10 percent of total emergencies in 1984, by late 1999 they were a determining factor in more than 50 percent of cases.

Source: FAO.

Figure A

It must be emphasized that cost estimates are tentative. The accuracy of estimates is undermined by the frequent lack of reliable statistics. Many of the conflict-affected countries are low-income countries where agricultural statistics are of questionable accuracy even during peacetime.38 Conflict frequently disrupts the normal data collection and estimation process. Furthermore, in wartime the unmeasured proportion of output is likely to increase because it is channelled through unofficial marketing networks.

Table 7






Total losses



Losses (Million $ -1995 prices)


Losses (Million $ -1995 prices)


Losses (Million $ -1995 prices)


Latin America

and the


Colombia, Nicaragua, Honduras (3)

1 328 (7%)

Nicaragua, Peru,

El Salvador, Guatemala, Colombia, Honduras, Panama (7)

8 686 (12%)

Nicaragua, Peru, El Salvador, Guatemala, Colombia (5) [Haiti, no data]

5 011 (8%)

15 025 (10%)



Angola, Burundi, Chad, Dem. Rep. Congo, Guinea-Bissau, Ethiopia, Mozambique, Namibia, Rwanda, Sudan, Zimbabwe (11)

9 427 (15%)

Angola, Cent. Afr. Rep., Chad, Dem. Rep. Congo, Ethiopia, Liberia, Mozambique, Namibia, Nigeria, Sudan, Zimbabwe (11)

21 951 (26%)

Angola, Burundi, Cent. Afr. Rep., Congo, Dem. Rep. Congo, Ethiopia, Kenya, Liberia, Mozambique, Somalia, Sudan, Zimbabwe (12)

21 005 (40%)

2 383 (30%)

South and




Lao People's

Dem. Rep.,

Sri Lanka,

Viet Nam (4)

4 346 (21%)

Cambodia, Philippines,

Sri Lanka (3)

707 (13%)

Cambodia, Philippines,

Sri Lanka (3)

6 360 (10%)

17 413 (14%)

West Asia

and North


Iran, Iraq (2)

(no data for other CACs)

206 (6%)

Afghanistan, Iran, Iraq (3) (no data for other CACs)

13 211 (11%)

Afghanistan, Algeria, Iraq (3) (no data for other CACs)

2 382 (58%)

5 800 (40%)


Countries: 20 Per year: 10.3

15 307 (16%)

Countries: 24 Per year: 16.9

50 556 (18%)

ountries: 23 Per year: 15.3

54 758 (41%)

120 620 (28%)

Notes: The countries listed are those for which the model (see below) produced statistically significant results for the conflict variable (except for exceptions noted below). Omitted are conflict-affected countries (CAC) for which the method showed statistically insignificant results, or countries without data. Percentages in parentheses are the share of losses in agricultural value added for the conflict-affected years, weighted across countries by total value added for each country. Losses are estimated by the following regression model by country:
agric = ao + a1(ag/man) + a2(Trd/ntrd) + a3(T) + a4(conf) + e
Where (with variables in logarithmic form except T and conf):
agric = agricultural value added in constant 1995 dollars
ag/man = the ratio of the agricultural to the manufacturing relative price among tradables
Trd/ntrd = ratio of tradable prices (agriculture, mining and manufacturing) to non-tradables (services), the real exchange rate
T = time trend
conf = conflict variable, which takes the value of 1 in conflict years, and
e = the error term
If the conflict variable was significant at the .1 level, in the war-affected years this was multiplied by the actual output to obtain estimated losses. For several countries, this method could not be applied and another method was used.

Figure 14

The estimated output losses for all developing countries are illustrated in Table 7, p. 73, and they are extensive. Over the 28 years from 1970 to 1997, the estimation yields an aggregate of almost $121 billion at 1995 prices, or an average of $4.3 billion per year. This can be compared with the estimated costs of raising nutrition to minimum required levels. Notional estimates calculate that it would cost approximately $13 per caput/year to provide enough food to feed the undernourished at a minimally adequate level.39 Dividing the estimated loss of $4.3 billion per year by $13 produces a figure of 330 million undernourished people who could have raised their food intake to minimum required levels for the year. Such a decrease in malnutrition would rival the combined impact of all food aid. In both the 1980s and the 1990s, conflict-induced losses in developing countries exceeded the total food aid of all types to those countries (see Table in Box 6). For the full decade of the 1980s, the former were about $37 billion and the latter $29 billion (both in current prices).

Economic losses from conflict in developing countries exceeded total food aid to those countries in the 1980s and 1990s.

Figure 15

From 1976 until 1990, the estimate of constant price losses grew at an alarming 12 percent per year (see Figure 14, p. 76), levelling off in the 1990s to about $6.5 to $7 billion per year. The pattern of output losses supports the conclusion that a peak was reached in 1990, although there has been no reversal of the previous trend. While this is cause for a degree of optimism, in the 1990s conflict remained a major cause of output loss and food insecurity in developing countries. The relative size of the losses can be indicated by comparing them with the agricultural trade of all developing countries (average imports and exports). While the comparison is not rigorous (trade involves gross output and losses refer to value added), the numbers are startling (see Figure 15, p. 76). After oscillating between 1 and 3 percent from 1970 until 1980, losses began an upward trend, peaking at slightly more than 7 percent of total agricultural trade in 1990.

Latin America is the developing region to have been least affected by conflicts since the 1980s.

Box 6


Food aid can, and does, play an important role in reducing the impact of conflict on food-insecure groups. Food aid assigned in response to conflicts would provide an indication of agricultural conflict losses, were it possible to make the necessary disaggregation. Available data make it possible to disaggregate only "emergency" food aid, which includes disbursements for the full range of emergencies, both natural and human-induced. The Table below reports shipments of cereal food aid to the conflict-affected countries listed in Table 7, with the exception of the Lao People's Democratic Republic and Viet Nam (their internal conflicts were over by 1989/90 when the Table begins). The overwhelming majority of emergency food aid to the conflict-affected countries went to the sub-Saharan Africa region during the 1990s - 86 percent over all years.



Thousand tonnes





Sub- Saharan Africa

South and Southeast Asia

North Africa and West Asia


Latin America

Sub- Saharan Africa

South and Southeast Asia

North Africa and West Asia





































































































Note: Figures cover the same countries as Table 7, except for the Lao People's Democratic Republic and Viet Nam in South and Southeast Asia. The first year of the 12-month period is given, i.e. 1989 = 1989/90.
Sources: FAO and WFP.

Figure 16

The situation by region

Among the regions, Latin America and the Caribbean, and South and Southeast Asia show substantially lower levels of conflict-induced losses (see Figure 16, p. 77 and Table 8, p. 78). The former was unique in that, at the end of the 1990s, only one Latin American country, Colombia, remained at serious risk of conflict. In this region, losses were concentrated in the 1980s and early 1990s. In Central America, the conflicts in El Salvador, Guatemala and Nicaragua were caused by a variety of political problems and socio-economic instability, but they also arose from long-term disputes over the distribution of land (especially in El Salvador and Guatemala). The losses resulting from conflict in Honduras were primarily "collateral" effects of armed struggles in the other states.40 While the right of

indigenous peoples to resources has been a major issue in Latin America, only in Guatemala was ethnicity an important contributing factor to a major conflict. Losses by country were generally lower in relative terms than in other regions, below 10 percent of total output in affected years. Nicaragua is the exception: in the late 1970s, estimated losses from conflict represented more than three times this percentage. In Nicaragua land disputes also continued to fuel conflict in the 1990s to a greater degree than in the other Central American countries. Despite a long history of rural insurgency in Latin America, typically over the distribution of land, at the end of the twentieth century it was the least conflict-affected region.

In South, Southeast and East Asia, none of the countries with large populations (i.e. Bangladesh, China, India, Indonesia and Pakistan) showed statistically significant losses as a result of conflict. In Southeast Asia, conflict was closely associated with cold war rivalries, evidenced by two major conflict-affected countries, the Lao People's Democratic Republic and Viet Nam, which achieved a lasting internal peace in the mid-1970s. In tragic contrast, war in Cambodia continued into the 1990s, with peace problematic even at the end of the century. The case of Cambodia demonstrates how easily losses can accumulate. If, from the early 1970s onwards, agricultural output in the country had been constant (even accounting for the great loss of lives in rural and urban communities), per caput production in 1997 would have been more than double its actual value. In spite of the Cambodia case, across all countries during their conflict years output losses in the region were a relatively low 14 percent of the sector's production (see Table 7, p. 73).

Losses from conflicts were especially severe in sub-Saharan Africa. Over the entire period, this region suffered the greatest losses in absolute terms, $52 billion (in 1995 prices). Across all affected countries during their conflict years, this represented almost 30 percent of agricultural output (Table 7). Unlike the two regions previously discussed, the estimates for sub-Saharan Africa show no marked tendency for either the incidence of conflict (number of countries) or the losses caused by conflict to decline in the 1990s. The number of countries affected was virtually the same in each decade. Relative to the total agricultural output of the affected countries in conflict years, losses grew at almost 5 percent per year from 1976 until 1996, with a sharp drop in 1997. While peace has come to several countries, including Mozambique, after long and devastating wars the region remained seriously plagued by conflict at the end of the century.

Economic losses from conflict in sub-Saharan Africa amount to almost 30 percent of agricultural output in affected countries since 1970.

For most countries in sub-Saharan Africa, conflicts were relatively short, but often recurred. In a number of cases, conflict dominated the entire 28-year period. Angola was an extreme case, with virtually continuous internal warfare since before 1970 and no peace imminent at the turn of the century (despite a formal peace accord earlier in the 1990s). Estimates suggest that, at the end of the 1990s, agricultural output in Angola was well below half of what it would have been in the absence of war. The Sudan has suffered for almost as long, although the relative conflict losses were less.

The severity of the impact of conflict on sub-Saharan African countries is shown in Table 9 in dollar terms (1995 prices). For the conflict-affected countries, estimated agricultural losses were 75 percent of ODA for the 28 years, a percentage that increased with each decade. Conflict losses in sub-Saharan Africa were considerably greater than foreign direct investment (FDI) in the affected countries. Since more than 80 percent of FDI went to two conflict-affected countries, Angola and Nigeria, agricultural losses far exceeded these private capital inflows for the other countries.

Work by Luckham, Ahmed and Muggah41 supports these estimated losses and suggests that, of the 11 most war-affected countries of sub-Saharan Africa, measured per caput agricultural production declined most sharply in the four where the state collapsed during the 1990s (Sierra Leone, Liberia, Somalia and Rwanda). One of the 11 was Chad which, although conflict-affected, showed no significant losses according to the estimation. This may be because warfare was relatively small-scale and took place outside the main agricultural production areas.

North Africa and West Asia were, according to some indicators, even more conflict-prone than sub-Saharan Africa, with internal wars or invasion affecting a long list of countries. However, agricultural losses were statistically significant in only four cases: Afghanistan, Algeria, the Islamic Republic of Iran and Iraq. According to the statistical estimates, conflict losses did not begin until the late 1970s, after which they increased dramatically (see Figure 16 and Table 8), rising by 26 percent per year in constant prices between 1979 and 1988. This was overwhelmingly the result of conflicts in Afghanistan and the Iran-Iraq war. So severe was the estimated effect of these conflicts that, in the 1990s, the losses of three countries alone - Afghanistan, Iraq and Algeria - exceeded the total for sub-Saharan Africa. These countries were on average the most intensely affected of all regions. Across affected countries in conflict years, losses amounted to 58 percent of total output for the 1990s.

Table 8







Latin America and the Caribbean





Sub-Saharan Africa





South and Southeast Asia





North Africa and West Asia






The impact of conflict on agriculture can be placed in the context of the nature of contemporary conflicts, which has changed since the Second World War. Until then, wars typically involved conflict between governments. For the most part, these wars were fought with regular armies, a well-defined division was made between combatants and civilians, there were recognized battle lines and the termination of conflict was relatively clear.

Table 9









(Million $ - current value)



11 924

31 160

21 916

64 999


7 999

50 811

65 715

124 525


2 740

5 984

14 030

22 753















Since the Second World War there have been few and usually only brief intercountry conflicts (although the wars in Korea and Viet Nam and between the Islamic Republic of Iran and Iraq were notably prolonged). Most contemporary conflicts are not between countries, but between a government and political or military groups seeking to overthrow that government or create a separate state. Such conflicts are typically not fought by two regular armies, but involve a complex struggle between the government and various contestants, which may be hostile to one another.42 In some cases it is difficult to determine who is officially or de facto in charge of the government. Because conflict is not formalized,43 its beginning and its end are ambiguous. As the conflict in Angola shows, not even a formal peace settlement signed by the warring parties need signal an end to war.44

Recent conflicts have tended to be between governments and opposition groups within the same country.

Since the distinction between combatant and civilian cannot be made with any precision, intracountry conflicts tend to be indiscriminate in their effect on the population. In fact, the bulk of the casualties tend to be among civilians. The instruments of war are often of a low technological level, involving small arms and land mines. Land mines wreak particular havoc with agriculture, excluding vast areas of arable land from cultivation until they are cleared, a problem discussed below. Thus, it should not be surprising that the estimates of agricultural output losses are large; it is in the nature of intracountry wars. However, even more than their effects on the productive capacity of farmers, these conflicts are likely to create food insecurity situations because of the disruptions they cause to the transportation and distribution of food.

Intracountry conflicts harm rural populations more than intercountry conflicts.

Another characteristic of intracountry conflicts is that they are usually fought in the countryside rather than in cities. For this reason, they tend to have devastating effects on the rural population and agriculture (see Box 8, which discusses survival strategies). This is in contrast to the intercountry wars of the past (and the recent conflicts in Iraq and the Balkan states), in which aerial bombing brought destruction chiefly to cities.

There is evidence that competition over scarce resources, such as land and water, enhances the probability of conflict. The control of fertile lands, water and food stocks is likely to be a strategic aim of all parties in a conflict. Conversely, the destruction of food stocks and the means of agricultural production may be a military objective of both the government and the insurgents. A "scorched earth" tactic may be pursued by either side, but particularly the government, as a means of undermining support for insurgents. This tactic has a long history.45 In such an approach to warfare, food insecurity becomes a powerful weapon, with disastrous effects on the rural population.

Box 7


The two central features of Rwandan agriculture are high population density in rural areas and the concentration on export crops - tea and coffee. Coffee production increased by an average 4.4 percent per year between 1969 and 1981, while tea output grew by 17 percent per year. However, the quality of coffee, the main export crop, deteriorated after the late 1970s. This left Rwandan coffee farmers and the economy less able to capture windfall gains from price increases, and more exposed to price declines. During 1985-1992 the world price of coffee fell by 72 percent in real terms and that of tea by 66 percent. While Rwanda produced
40 percent more coffee in volume during 1989/90 compared with the early 1980s, earnings were down by 20 percent. After 1992, coffee price stabilization transfers to farmers were terminated, as part of the country's structural adjustment programme. In addition to this crisis, rapid population growth over a longer period generated a drop in the size of landholdings, and there was little effective intensification of production. Droughts in the 1980s worsened conditions. According to one estimate, by 1989 one in six Rwandans was affected by famine. According to official estimates, food energy output fell from 2 055 kcal per farmer/day in 1984 to 1 509 kcal in 1991. Thus, during the 1980s a variety of severe pressures built up within Rwandan agriculture.

The onset of war in 1990 led to the displacement of 15 percent of the population, according to some estimates. The effect of the 1994 war on agriculture was dramatic. When the Rwandan Patriotic Front came to power, effectively ending the genocide, 2 million people had fled the country and hundreds of thousands were classified as internally displaced people.

In the aftermath of the genocide, many sources predicted unprecedented famine. However, these fears proved to be exaggerated. Later estimates of the conflict losses suggested that they varied substantially between and within provinces. Furthermore, they indicated that the problems of Rwandan agricultural production and rehabilitation were more complex than seed and harvest problems and extended to the entire organization of rural production. Far from being self-sufficient, most farmers relied on cash for crucial inputs such as seeds. The two most important income sources, apart from selling marketed farm output, were interprefectural seasonal migration by males and agricultural wage labour on commercial farms by females. Both were disrupted in the 1990s, the former by direct conflict and the latter by the flight or death of large-scale farmers. In 1995 the Ministry of Agriculture estimated that, since April 1994, three out of four farmers had been displaced and that the December-January harvest had fallen to 50 percent below normal output.

The recovery of agriculture after 1995 was uneven, as farmers returned to their lands, legal conflicts were partially resolved and labour migration revived. However, the problem of post-conflict rehabilitation, and agricultural policy more generally, was not merely one of restoring production levels and institutions to their pre-war status quo. Long-term solutions require a reorganization of the entire agricultural production system.

Sources: P. Gourevitch. 1999. We wish to inform you that tomorrow we will be killed with our families. London, Picador; J. Pottier. 1996. Agricultural rehabilitation and food insecurity in postwar Rwanda: assessing needs, designing solutions. IDS Bulletin, 27(3); A. Storey. 1999. Economics and ethnic conflict: structural adjustment in Rwanda. Development Policy Review,17: 43-63; and P. Uvin. 1996. Development, aid and conflict: reflections from the case of Rwanda. Research for Action No. 24. Helsinki, UN University/World Institute for Development Economics Research.

Box 8


When households are in danger of becoming food-insecure, they employ various types of coping strategies to maintain their access to food and to protect their livelihoods during the crisis. They may collect wild foods, look for credit, sell their labour or reduce consumption. People's reactions depend mainly on their perception of the severity of the crisis and their economic and social position.

In deciding which options to employ, a household will carefully weigh the economic and social costs of each action, although during conflicts people may be forced to take greater risks more suddenly, since their first goal is to save lives. In addition, freedom of movement is often restricted because of insecurity, and this limits access to foods. Collection of unfamiliar wild foods can be risky; many contain natural toxins and, when procedures for their preparation and utilization are unknown, their consumption may lead to toxicity and poisoning. Very often homes and farms are destroyed when people flee, which makes it extremely difficult for them to re-establish normal lives when they return to their home communities. In addition, many of the economic and social networks that households normally employ during times of crisis may be shattered in conflict situations; the community, or even the family, may no longer be available as a safety net. In those cases the range of successful coping strategies may be severely limited.

A good knowledge of the various survival strategies available to households is fundamental for any efficient system of prevention and protection against food insecurity in conflict-prone areas.

Source: Adapted from A. Hussain and M. Herens. 1997. Child nutrition and food security during armed conflict. Food, Nutrition and Agriculture No. 19.


By their nature, intracountry wars affect agriculture in a variety of ways. The impact also varies with the characteristics of each country's agriculture. In some countries there may be surplus labour in rural areas. In such cases, war casualties, regardless
of their obvious human cost, would not necessarily weaken agricultural productivity and output trends. In other circumstances, agriculture may be constrained by labour shortages, especially, at seasonal peaks (harvesting, weeding, etc.). When this is so, war casualties or the diversion of men from agricultural production to the armed forces will undermine the viability of agriculture.

The direct cost of war damage tends to be higher when agriculture is more capitalized, i.e. when mechanization and irrigation are more advanced, purchased inputs are used and a large portion of the output is marketed. Disruptions in marketing channels may have costs for commercialized output, but might not necessarily force households into food insecurity, should they still produce for own consumption.

The costs incurred by war are greater for mechanized agriculture than for subsistence agriculture.

Conflict can affect agriculture and food security in many other ways. In areas that specialize in labour-intensive export crops, war will have an impact on foreign exchange earnings, which may have severe implications for development and food security. If the rural population consists of net food purchasers, who perhaps provide wage labour for commercial farms, the disruption of incomes and marketing networks will expose people to greater food insecurity than if there is a preponderance of subsistence food production. In cases where commercial farming areas depend, at least partly, on labour migration, war can interrupt established flows of seasonal labour migration and, therefore, damage productivity on commercial farms. This disruption also reduces income in areas that are not directly affected by war but that send migrant labourers to conflict zones. For example, workers from southern Kordofan in the Sudan could no longer migrate to jobs in other regions during phases of civil war. Guest workers who were repatriated to Bangladesh and the Philippines from Iraq, together with households dependent on their remittances, became victims of the Persian Gulf conflict through loss of earnings.46

Livestock operations are highly vulnerable to war, as animals are left untended and disease spreads easily.

The long-term effects of war on agriculture will vary with, among other factors, the composition of cropping patterns. For example, in areas with tree crops such as coffee or cashew nuts, refugee flight and conflict are likely to cause neglect or abandonment of trees and, subsequently, greater exposure to pests and diseases.47 This was the case with the production of bananas in Rwanda and Burundi, where the fruit is a staple. As a result, there will be a delay in the postwar recovery of agricultural productivity, as trees and bushes lose productivity through age, and the costs of reviving competitive export production will rise. Similarly, one of the most vulnerable agricultural activities during wartime appears to be livestock raising. Estimates suggest that more than half the total livestock was lost through direct and indirect conflict effects in Somaliland during the mid-1990s.48 The estimate for Mozambique is roughly 80 percent.49

Large farms, whether state-run, private, or joint ventures between the two, may be relatively protected from direct war impacts, but still be vulnerable as targets for insurgents. For example, a state farm will often be a target for rebels whose strategy includes the selective destruction of government property. Governments may use troops to protect such enterprises, but it is likely that the linkages between large and small farms will be weakened, if not altogether severed, by wartime insecurity.

Agriculture's other linkages may suffer disruption during war. Rural people in low- and middle-income countries draw income from a portfolio of sources, including subsistence production, processing of crops for local and distant markets and small enterprises such as transport and crafts. There are linkages between the agricultural and the manufacturing sectors: in many instances the core of early industrialization is agroprocessing, such as cotton ginning, sugar refining, soft drinks production, industrial brewing of beer, furniture making and pulp and paper mills. There are also resource flows between sectors. All of these linkages are liable to be disrupted by war.


Land mine victims: Farmers are sometimes forced to
cultivate fields dotted with land mines

- FAO/21328

If links between rural and urban areas are weakened by conflict, or if tourism in rural areas is reduced by war, many of the diverse income sources of the rural population are likely to be curtailed by lack of demand or inputs. Agro-industry may be badly affected by conflict if the supply of raw materials is disrupted, the marketing of raw material from producers to industrial enterprises is severed, or the demand for raw materials is reduced. Even if intersectoral resource flows remain in place, the mechanisms may be undermined by war. Marketing boards may have less ability to purchase crops, or their organization and personnel may be weakened by the diversion of resources to the war effort. The positive effects of resource transfers from agriculture to other sectors and of reinvestment in agriculture will suffer. A further mechanism of violence disrupts agriculture: the targeting of transport and communications infrastructure during war.

Land mines expose agricultural workers to danger and reduce their ability to produce.

Land mines are commonly scattered over rural areas - indiscriminately in some cases (see Box 9). They may be laid in crop fields, along footpaths linking villages to fields and rivers and on rural feeder roads, major highways and bridges. They have a direct and significant impact on agricultural production simply by making production too dangerous. Neutralizing the effect of land mines requires mapping their location. Often, however, neither the ruling regime nor its adversaries have records of where they have laid mines. In Afghanistan, Angola and Cambodia, countries with a particularly high density of mines, the overall impact on agricultural production is immense and likely to last for years after the end of conflict.

During war, marketing and distribution systems are disrupted and food becomes more expensive.

Production and communications infrastructure is critical to the vitality of agricultural production.50 Poor transport adds to production and marketing costs, and damage to infrastructure clearly worsens this situation by, for example, raising traders' margins. Added to this, it is common in wars for access to petroleum to be disrupted, and for transport equipment, especially trucks, to be commandeered by the government or armed groups.51

A corollary of the increased risk of marketing agricultural produce and inputs is that traders' profits tend to rise. For those willing to take the risks, there are substantial war-monopoly rents; indeed, such profits may be necessary for marketing to continue and are sometimes referred to as "war taxes". This is a specific example of how, in a civil war, the established checks and balances of social and economic rules are eroded, are poorly monitored or collapse. In these circumstances, agricultural production and related activities may continue at a reduced level under conditions that have been characterized as a "vicious market fundamentalism".52 Some people do very well out of such conditions.53 While the activities of "profiteers" may offer a lifeline for people whose income-earning opportunities might otherwise be non-existent in war, the conditions that allow people access to income sources are likely to be worse than they are in peacetime.

An extreme example is when producers suffer a direct levy on their products in order to feed military forces for little or no payment. In Sierra Leone, for example, "... [Armed] young men ... discovered that control over business operations gave them access to additional benefits, such as the opportunity to extort food from farmers, set up roadblocks to levy tolls on internal commerce and organize protection and looting rackets.... This promise of survival, exemption from widespread exploitation of local populations, and even material gain amid the destruction of war, is a major incentive drawing young men and boys to be fighters."54

Alternatively, war may include, either directly or as a by-product, land and other asset accumulation by some at the expense of others; i.e. an increase in inequality, of which the long-term economic implications are not straightforward.55

Macroeconomic policies can be designed to prevent the worsening of tensions and conflicts that can lead to war.

Box 9


Angola has experienced warfare almost continuously since 1961. Land mines were first used in mid-1961 in the war between Portuguese colonial authorities and nationalist groups. With the growth of foreign support for nationalist movements, which started in 1968, land mines became an increasing feature of the conflict. One of the most intense phases of military activity occurred in the late 1980s, during major battles in the southern Cuito Cuanavale area. Both sides laid large numbers of mines. Mine laying intensified after the failure of the peace process and elections between 1990 and 1992. The People's Movement for the Liberation of Angola (MPLA) government and UNITA both laid thousands of new mines to obstruct roads and bridges around besieged towns. There were mine "belts" up to 3 km wide.

One estimate puts the total number of land mines laid in Angola at some 20 million, with 4 million still in the ground in 1999, 6 million removed or detonated, and 8 million unaccounted for. Sixty different types of anti-personnel mines from 19 different countries have been identified. However, there is no comprehensive survey of their dispersion and no one knows precisely how many mines have been laid. The country has one of the highest rates of land mine injuries per caput in the world. Out of a population of some 9 million, there are tens of thousands of amputees, most caused by mines. With a conservative estimate of 20 000 serious land mine injuries, this implies a rate of injury that would be equivalent to 5 million people in the United States.

A great number of mines were laid on footpaths to rivers and small agricultural plots. Others were placed on roads and bridges. The effects on agriculture, although impossible to quantify, are clear: mines restrict access to farmland and prevent cultivation; when people are injured or killed by mines, the agricultural labour force is affected, with potentially severe effects on household labour allocation and food security; and mines on roads and bridges undermine trading and hinder effective relief efforts. Together with active fighting, refugee flight and recruitment into armed forces, land mines are responsible for destroying much of the agricultural supply base in the country and for reducing many Angolans to periodic famine.

"The rebuilding of rural communities after conflict can be made impossible by the presence of mines and ordnance. New buildings cannot be constructed and land use cannot be extended to support more people. Refugees and people displaced by the fighting, desperate to restart their lives, must either remain separated from their homelands or return despite the mines. An inability to grow crops or safely collect fruit, wood or other natural resources is a common product of land mine contamination. In some cases the local economy and the pattern of land mine contamination mean that land is not only denied altogether, but the land that can be farmed cannot be used to best effect. By denying land for agriculture, and making land use less efficient, land mines and unexploded ordnance (UXO) can create food insecurity. This in turn can promote dangerous activities, such as salvaging scrap metal from UXO, as people look for ways of supporting themselves or supplementing their income. Spurred on by the need to feed their families, people are forced to farm land that is known to be mined - often trying to clear mines themselves. This desperation causes accidents and a spiral further into poverty."

The Mines Advisory Group

Sources: Africa Watch.1993. Landmines in Angola. London; Human Rights Watch. 1999. Landmine monitor - toward a mine-free world. New York; S.H. McCormick. 1994. The Angolan economy: prospects for growth in a postwar environment. Washington, DC, Center for Strategic and International Studies; Mines Advisory Group. 1999. The effects of landmines on the community.


That conflicts tend to be closely associated with low levels of development combined with inequity and competition over scarce resources points to an obvious conclusion: countries should seek to promote growth that is both rapid and inclusive. If the bases for growth are to be created through measures aimed at macroeconomic stabilization and reform, the challenge is to manage such measures in a manner that contains social stress and reduces, or at least does not accentuate, inequities. This view is now widely supported, despite difficulties experienced in translating it into fact. In a 1997 World Bank report,56 it was observed that:

"The Bank needs to integrate concern for conflict into development operations. It must ensure that its interventions do not aggravate existing inequities ... and that they ameliorate potential conflict situations, through judicious social analysis and adequate attention to distributive policies...."

Promoting rural development can reduce the risk of conflict in agriculture-dependent countries.

An excessive emphasis on rapid stabilization involving stringent fiscal austerity might conflict with the investments required to rehabilitate the agricultural sector, as well as with the so-called peace dividend (see Box 10, p. 92). While stabilization and adjustment programmes usually call for reduced state intervention in markets, post-conflict situations frequently require an increased role of the state in aiding local communities during reconstruction, for example through the direct provision of goods and services for food security and the supply of agricultural inputs.

Distribution issues are especially important, including those in the agricultural sector where conflicts over resources are often direct and take place in a context of increasing inequity and scarcity. It is fundamental that competition over scarce resources should take place in an equitable environment, within an enforced legal framework, with commonly accepted standards of commercial behaviour and respect for property rights, be they individual or communal. The enforcement of legislation supporting equitable landownership structures is a major step in this direction. In the short term, for many conflict-affected countries, "adequate attention to distributive policies" might imply moderating the impact of competition on smallholders during a transitional period when normal marketing services, transport facilities and agricultural extension are being re-established.

It has been observed that the most conflict-prone countries include many of those where agriculture is a major component of the economy and where the majority of the population is rural. In these situations, not only does the promotion of agricultural and rural development foster development and food security overall, but it is also a powerful way of reducing the risks of conflict.

The maintenance of strategic food stocks would appear to be an obvious response to the food insecurity situations caused by conflict. Stockholding is, however, notoriously costly and difficult to manage - especially during conflicts, when transportation is disrupted and silos become natural targets of attacks.

The building and maintenance of farm storage facilities in conflict-prone areas is only a viable option if they can be protected and controlled, and if the stored food can be effectively channelled to the areas in need. It should also be noted that it is very difficult and costly to maintain food stocks adequately in tropical and humid zones.

The World Trade Organization (Uruguay Round) Agreement on Agriculture allows countries to stockpile commodities for the purposes of food security and to sell from stockpiles at below market prices to the urban and rural poor. In general, the Agreement on Agriculture is quite flexible towards agricultural policies that are explicitly directed at poverty reduction, and these are of great relevance to conflict situations. Under the Agreement, the Decision on Measures concerning the Possible Negative Effects of the Reform Programme on Least-Developed and Net Food-Importing Developing Countries (the so-called Negative Effects Decision) is of particular relevance. This decision states the intention of parties to establish mechanisms to ensure that the rules stipulated by the Agreement on Agriculture do not adversely affect the food security of the least- developed and food-deficit countries (i.e. almost all conflict-affected countries) by: i) reviewing the adequacy of food aid; ii) ensuring that foodstuffs are provided to these countries on an increasingly concessionary basis; and iii) making a commitment to consider requests for technical and financial assistance to foster agricultural growth and food security. Unfortunately, at the end of the 1990s, no concrete steps had been taken to implement these commitments.

Emergency relief efforts should be linked to longer-term development goals.

Box 10


Disarmament and the resolution of a conflict appear to offer the prospect of shifting public expenditure towards economic growth and social welfare, generally known as the "peace dividend". A number of factors might prevent the realization of the peace dividend:
There are costs associated with the transfer of resources from military to civilian uses. These costs include the retraining of personnel, the funding of resettlement (perhaps including equipping people with agricultural tools) and the conversion of military bases.
Where there is no immediate reallocation of labour from military to civilian employment, there will be unemployment.
The peace dividend depends on a decisive end to conflict, but the nature of civil war is such that there is no clear end to hostilities. Political, economic and physical insecurity often persists after the formal end of a war, and this creates a disincentive for the reallocation of resources to productive purposes, particularly by the private sector. At the same time, continued tension creates a need for continued expenditure on national security.
There are fiscal problems associated with the realization of the peace dividend.
Revenue raising might not increase after the end of war. For example, in Ethiopia the state used coercive revenue-raising measures during the war and, when these were abandoned with peace, there was a decline in government revenue.
Any putative peace dividend can easily disappear when there is a need, and external pressure, for deficit reduction.
There is no automatic mechanism for realizing a peace dividend. It is ultimately a political matter, within a context of extremely sensitive social tensions.

Evidence of a peace dividend is elusive. After eight years of peace, the Ugandan economy was still far below its pre-war peak.1 In South Africa and the southern African region, the economic benefits of the peace dividend have yet to be realized, despite a 45 percent decline in the defence budget between 1989 and 1995.2 Nonetheless, there are examples of possible peace dividend effects. Although the Mozambican economy also remains below the levels it had achieved at independence, since the peace agreement in 1992 the country has been among the fastest growing in the world.

The implications of the peace dividend for agriculture are ambiguous, and there has been no direct analysis of its impact on the sector. The most obvious mechanism through which a peace dividend can be expected is the return of refugees, internally displaced people and demobilized soldiers, all representing a shift of labour back into agriculture. Some agriculture-related buildings and equipment may be repaired or replaced relatively quickly after a war, and there will be greater security as well as a potentially significant improvement in the efficiency of markets. With the assistance of donors, seed stocks may be replaced quickly. However, a number of problems are likely to hamper the smooth realization of an agricultural peace dividend. Land mines in many conflict areas can take years to remove. In Zimbabwe, some areas were still fenced 20 years after the country's conflict.3

The revival of markets and poverty reduction after a war are likely to be affected by how thoroughly and quickly the rural infrastructure can be rebuilt. This is not just a question of reconstructing infrastructure but also of deciding on priorities in infrastructure investment. FAO's monitoring of the impact on food and agriculture of conflict in Sierra Leone and Liberia stressed the immense damage to infrastructure. Finally, in many post-war situations, institutional change affects agricultural output, access to land and food security. In most of these situations, land tenure disputes continue after peace has returned.

In summary, the peace dividend is not a magic wand. Realizing the scope for a peace dividend is not an automatic process, but one that depends on choices made by relevant actors, not least by foreign donors. Donors have to consider carefully how their interventions might reduce conversion costs and maximize potential peace dividend benefits. Given that peace dividend benefits are likely to emerge in the medium to long term, donors need to make long-term commitments to support the process of realizing the dividend.

1 J.-P. Azam et al. 1994. Some economic consequences of the transition from civil war to peace. Policy Research Working Paper No. 1392. Washington, DC, World Bank.
P. Batchelor and S. Willett. 1998. Disarmament and defence industrial adjustment in South Africa, p. 170. Oxford, UK, Stockholm International Peace Research Institute and Oxford University Press.
S. Willett. 1996. Military spending trends and developments in southern Africa: South Africa, Angola, Zimbabwe, and Mozambique, p. 36. Report for the Development Co-operation Directorate, OECD. 3 January 1996, London, Kings College, University of London. (mimeograph)

Sources: C. André and J.-P. Platteau. 1995. Land relations under unbearable stress: Rwanda caught in the Malthusian trap. Namur, Belgium, Centre de Recherche en Économie du Developpement, Faculty of Economics, University of Namur; P. O'Brien. 1988. The economic effects of the American civil war. Studies in Economic and Social History. Basingstoke, UK, and London, Macmillan; S.J. Stern. 1998. Shining and other paths: war and society in Peru, 1980-95. Durham, UK, and London, Duke University Press.

Policies to aid recovery

Countries' different experiences indicate a tremendous variation in the speed of agricultural recovery after conflict. In Mozambique, recovery was extraordinarily rapid, while in Nicaragua it was frustratingly slow. The policy lessons from these experiences are not always conclusive. Further research is required to determine the extent to which policy (as well as the particular type and combination of policies) facilitates the post-conflict recovery of agriculture. A number of general observations can be made, however.

The public sector has several functions in the recovery process. First, there are the immediate needs for resettlement of refugees and ex-belligerents, rehabilitation of land and infrastructure, and reintegration of communities into marketing networks. In addition, there are programmes that go further towards resolving the structural tensions that generate conflict - through such action as the decentralization of authorities and resources, the promotion of participatory rather than top-down decision-making and clarification of rural property rights. In post-conflict situations, however, the public sector is likely to see its means of intervention considerably weakened, and the process of capacity building may take time. In these circumstances, the speed of rehabilitation is particularly dependent on the dynamism of the private sector and civil society, and on the work of NGOs who often play a crucial role in such activities as relief delivery, advice to communities and support to local planning and programming activities.

Agricultural policy in all post-conflict situations requires the balancing of relief activities with development efforts. It has been noted, in fact, that a "gap has emerged between relief and development which has implications for the agencies that have traditionally been involved in either activity. The need for interventions that can bridge this gap and facilitate a faster transition from emergency response to sustainable development is compelling."57 As far as possible, the emphasis should be on long-term development objectives. Even agricultural recovery programmes that are initiated very close to the end of hostilities and aim at relief and the repair of infrastructure could be linked to a longer-term development strategy. There are many constraints to the effective promotion of long-term sustainable development in immediate postwar conditions. Not the least of these is the need for political will and institutional capacity. Tensions may arise between the pursuit of long-term capacity building and the immediate maximizing of the benefits of employment-intensive public works schemes, which must be resolved with regard to concrete conditions.

Development efforts can be made during the emergency phase through capacity building for local and national government services that provide agricultural inputs, extension services and marketing advice. This frequently requires the training of civil servants in activities that provide services to rural communities. Training provision can be combined with a strengthening of the links between private and public sector activities, which have proved important for achieving rapid growth in various parts of the world. Fostering these links is, in part, a matter of developing effective rural institutions such as public sector employment promotion offices and networks of local development agencies. It also involves encouraging the relationships among both private and public institutions. Rehabilitation of the agricultural sector after conflict is, by its nature, public sector-driven, although the participation of the private sector needs to be fostered.

Recovery of the agricultural sector requires action by external donors as well as a strong national effort. A number of UN organizations have policies for operations in conflict-affected countries. Participation of international organizations covers a spectrum of activities, from physical reconstruction, institution building and demobilization, to the creation or re-creation of an inclusive, non-discriminatory society. WFP can provide relief, FAO can advise on agricultural development and food security concerns and the World Bank can fund infrastructure. It is important for the speed of recovery, however, that assistance from these and other agencies is provided in close harmonization and collaboration among themselves and with the government. The latter, in turn, has a crucial role to play in creating a political, economic and institutional environment capable of maximizing the benefits of external assistance.

FAO's role in conflict-affected countries covers a broad range of activities. As soon as the security situation permits, FAO mounts missions to carry out assessments aimed at quantifying needs for both food and emergency assistance in agriculture. The types of assessment carried out may include the impact of the conflict on national food supply and demand, the food security and nutrition situations of affected groups, the population's need for international food assistance, the agricultural production capacity in the affected area, the need for agricultural relief to enable production to resume quickly, and the need for longer-term rehabilitation and reconstruction measures. Assessments are used as a basis for FAO's own relief efforts and are disseminated quickly to the international community in order to enable timely and effective donor responses.

FAO also provides "agricultural relief", defined as agricultural rehabilitation assistance provided on an emergency basis. Such relief is aimed at the rapid reduction of dependence on emergency food assistance and at providing a basis for longer-term rehabilitation. Assistance includes the provision of essential agricultural inputs such as seeds, tools, fertilizers, livestock and veterinary supplies to enable affected populations to resume basic productive activities quickly - in time for the next agricultural season where possible.

Agricultural relief is not, however, limited to the supply of agricultural inputs. Special agricultural relief operations also include the provision of services and technical advice that are not available from other UN agencies or from NGOs.

In its relief interventions, FAO also gives specific attention to the food security constraints of at-risk households in the areas affected, as well as to the need for appropriate information that allows affected households to make informed choices regarding food acquisition, preparation and distribution in unfamiliar situations.

As follow-up to immediate agricultural relief, FAO provides assistance in restoring extension, veterinary, plant protection and input supply services and institutions where these have been disrupted, and in the physical reconstruction of agricultural infrastructure such as dams and irrigation systems, markets and crop storage facilities.

FAO also provides policy and strategy support for recovery and development programmes in the food and agricultural sectors. This assistance is geared towards bringing the need for relief to an end and enabling development to proceed. It includes activities that help to make development sustainable by preventing and preparing for the possibility of further disasters and emergencies. A major emphasis is put on strengthening the coordination of locally active emergency and development institutions and on encouraging the participation of the affected population in designing and implementing interventions to promote household food security and nutrition. Priority is given to the needs of food-insecure households and to the promotion of sustainable and healthy livelihoods.


1 This report is based on information available as of March 2000. Current information on the wheat market can be found in FAO's bimonthly Food Outlook report.

2 More detailed statistics on cereal and non-cereal food aid shipments are available at: under Statistical databases and then All databases.

3 Expressed in local currency.

4 Unless otherwise indicated, economic estimates and forecasts in this section are from IMF. World Economic Outlook, October 1999.

5 UN. 1999. The World Economy in 1999. Washington, DC.

6 These forecasts were prepared for FAO by the Institute for Policy Analysis, University of Toronto, Canada, associated with Project LINK.

7 "Low-income food-deficit countries with the lowest capacity to finance food imports" is a subgroup of the FAO category of all LIFDCs. The subgroup comprises 31 countries (17 in Africa, 3 in Latin America and the Caribbean, 8 in Asia and the Pacific and 3 in the Near East) for which food imports represent 25 percent or more of total export earnings. The group of EHDAEs comprises 47 countries (24 in sub-Saharan Africa, 18 in Latin America and the Caribbean and 5 in Asia) for which agricultural, fishery and forestry exports are equivalent to at least 20 percent of their total exports, or 20 percent of their total imports.

8 Failures in rural credit delivery have been extensively documented. See FAO/GTZ. 1998. Agricultural finance revisited: Why? Agricultural Finance Revisited No. 1. Rome (see also other publications in this series); World Bank. 1975. Agricultural credit. Washington, DC; and World Bank.1993. A review of Bank lending for agricultural credit and rural finance, 1948-1992. Washington, DC.

9 Officially registered as members of the Microcredit Summit Campaign.

10 Members of the Consultative Group to Assist the Poor (CGAP), major donors of microcredit. The CGAP consortium was created in 1995 and has its headquarters at the World Bank.

11 Microfinance organizations (MFOs) range in size, expertise, funding base, clientele, purpose, geographical coverage and other characteristics. They may be run by the state or local government, private bankers, local or international NGOs or community organizations. Microcredit may be a primary activity of an MFO, or part of the overall development strategy of a government or NGO.

12 Other estimates of borrower numbers are lower, including 6.9 million estimated by the United Nations Development Fund for Women (UNIFEM, 1999) and 16 million estimated by the World Bank (1999).

13 S. Khandker. 1998. Fighting poverty with microcredit, p.11. Washington, DC, World Bank.

14 UNCDF. 1999. Working paper on microfinance. New York.

15 Ibid.

16 An MFO generally includes a savings component. This paper focuses only on the credit aspect of microcredit.

17 Khandker (p. 150, op. cit., note 13) reports that marginal farmers receive 72 percent of the microcredit loans made to agriculture, compared with 3 percent of the agricultural loans made by agricultural development banks.

18 FAO. 1998. The State of Food and Agriculture 1998, p. 290. Rome.

19 IFPRI. 1998. Rural finance and poverty alleviation. Washington, DC.

20 These issues have been extensively discussed, among others, by: World Bank. 1993. op. cit., note 8; and K. Hoff and J. Stiglitz. 1990. Introduction: imperfect information and rural credit markets: puzzles and policy perspectives. The World Bank Economic Review, 4(3).

21 J. Morduch. 1998. Does microfinance really help the poor? New evidence from flagship programs in Bangladesh. HIID, Harvard University (unpublished document).

22 Although J.D. von Pishke (1999. Poverty, human development and financial services. UNDP Occasional Paper No. 25) has found that these programmes do not affect repayment performance, they do seem to affect earnings ability.

23 Khandker, p. 145, op. cit., note 13.

24 Women account for 95 percent of the Grameen Bank's borrowers.

25 Khandker, p. 155, op. cit., note 13.

26 J.Morduch.1999.TheGrameenBank:afinancialreckoning(unpublishedmimeo).Availableat:

27 Real interest rates charged on microcredit loans range from 2.5 to 6 percent per month, compared with 7 to 40 percent per month among moneylenders, according to sources in various countries.

28 Khan et al., cited in H. Zaman. 1999. Assessing the poverty and vulnerability impact of microcredit in Bangladesh: a case study of BRAC.paper for the WDR 2000/2001. Washington, DC, World Bank. The World Bank (2000, Part 5.35) also reports that, when microcredit loans are used as an ex ante coping device by borrowers, the proceeds are usually devoted to building assets rather than direct consumption.

29 Some research suggests that control of the loan and productive activity may rest with males in the household (e.g. L. Mayoux. 1999. Women's empowerment and microfinance programmes: approaches, evidence and ways forward. Milton Keynes, UK, Development Policy and Practice Working Paper No. 41.), but the point is debated in the literature.

30 S. Schuler, S. Hashemi and A. Riley. 1997. The influence of women's changing roles and status in Bangladesh's fertility transition. World Development, 25(4).

31 E. Ostrom, R. Gardner and J. Walker. 1994. Rules, games and common- pool resources. Ann Arbor, USA, University of Michigan Press.

32 Khandker (p. 7, op. cit., note 13) reports a higher return and more cost-effectiveness of Grameen Bank loans compared with other microcredit programmes, targeted food programmes and agricultural development anks.

33 In one document, the World Bank estimated that more than 50 countries had been involved in civil conflicts between 1980 and 1995 - see World Bank. 1997. A framework for World Bank involvement in post-conflict reconstruction, p. 3. Washington, DC. In this paper, the count used is 33 (see Box 5).

34 J. Carter. 1999. First step toward peace is eradicating hunger. International Herald Tribune, 17 June 1999.

35 See R.H. Green. 1987. Killing the dream: the political and human economy of war in sub-Saharan Africa. IDS Discussion Paper. Brighton, UK, Institute of Development Studies; R.H. Green. 1994. The course of the four horsemen: the costs of war and its aftermath in sub-Saharan Africa. In J. Macrae and A. Zwi, eds. War and hunger: rethinking international responses in complex emergencies. London, Zed Books; F. Stewart. 1993. War and underdevelopment: can economic analysis help reduce the costs? Journal of International Development, 5(4): 357-380; and F. Stewart. 1998. The root causes of conflict: evidence and policy implications. Paper prepared for conference on War, Hunger and Displacement: the Economics and Politics of the Prevention of Humanitarian Emergencies, Stockholm, 15-16 June 1998. UN University/World Institute for Development Economics Research.

36 For a critical discussion on the costs of war, see C. Cramer. 1999. The economics and political economy of conflict in sub-Saharan Africa. Paper presented to the Standing Committee on University Studies of Africa (SCUSA) conference, Norwich, UK, September 1999. For a historical parallel for the First World War, see: A.S. Milward. 1984. The economic effects of the two world wars on Britain, second edition. Studies in Economic and Social History. Basingstoke, UK, and London, Macmillan.

37 The method followed specifies a simple equilibrium model in which the level of agricultural output in the absence of conflict is determined by relative prices and a trend variable. The former seek to capture the short-term output-optimizing decisions of farmers, while the latter approximates the growth of the labour force and technical change. Conflict losses are estimated within this framework by a binary ("dummy") variable that takes the value of unity for conflict years. In this study, conflict years for each country were identified through the compilation of information from a number of sources that specialize in conflict monitoring. Figure 13 shows the number of developing countries for which there were statistically significant conflict losses. While other countries were also conflict-affected, either the impact on agriculture was not statistically significant or data for estimation were not available. The country count shows that, after increasing for almost 15 years, the number of output-reducing conflicts peaked in 1990 and then declined.

38 P. Svedberg. 1990. Undernutrition in sub-Saharan Africa: a critical assessment of the evidence. In J. Dreze and A. Sen. The political economy of hunger. Oxford, UK, Clarendon Press.

39 FAO. 1996. Food assistance and food security. World Food Summit Technical Background Documents, Vol. 3, No. 13. Rome. This estimate is based on the assumption, albeit unrealistic, that food assistance can be targeted perfectly.

40 Losses from conflict in Central America are estimated by a similar method in J. Weeks. 1997. Trade liberalization, market deregulation and agricultural performance in Central America. Journal of Development Studies, 35(5).

41 L. Luckham, I. Ahmed and R. Muggah. 1999. The impact of conflict on poverty in sub-Saharan Africa. Background Paper for World Bank Poverty Status Assessment for sub-Saharan Africa. Brighton, UK, Institute of Development Studies, University of Sussex, p. 21-22.

42 A good example of this was the conflict in Angola, in which at least two of the anticolonial movements were at war with each other and with the Portuguese army.

43 On occasion, insurgents can obtain a degree of international recognition by being granted "belligerent" status, as was the case with the insurgentsin El Salvador in the 1980s.

44 In Angola, the death and destruction wrought in the two years after the 1992 peace accord was greater than that suffered between 1961, the start of the anticolonial conflict, and 1992. See the Preface to E. Medi. 1997. Angola: study of vocational rehabilitation, training and employment programmes for persons disabled by the conflict: experiences and issues. Geneva, ILO.

45 For a review of the history of agricultural disruption during conflict from the twelfth century onwards, see PRIO. 1999. To cultivate peace: agriculture in a world of conflict. PRIO Report No. 1/99. Oslo. International Peace Research Institute.

46 E. Messer, M.J. Cohen and J. D'Costa. 1998. Food from peace: breaking the links between conflict and hunger. Food, Agriculture and the Environment Discussion Paper No. 24. Washington, DC, IFPRI.

47 For example, before independence Angola was the world's fourth largest coffee exporter. Intermittent drought, policy neglect, the direct effects of war and political insecurity combined to reduce annual export production from 220 000 tonnes in 1973 to 3 000 tonnes in 1993. See S.H. McCormick. 1994. The Angolan economy: prospects for growth in a postwar environment. Washington, DC, Center for Strategic and International Studies. On a similar war and its effects on cashew production in Mozambique, see C. Cramer. 1999. Raising agricultural output capacity and productivity in low-income countries: with special reference to Mozambican cashew production. Background paper for UNCTAD. Least Developed Countries Report, 1999. London, School of Oriental and African Studies, University of London.

48 I. Ahmed and R.H. Green. 1999. The heritage of war and state collapse in Somalia and Somaliland: local-level effects, external interventions and reconstruction. Third World Quarterly, 20(1): 113-128.

49 T. Bruck. 1997. Macroeconomic effects of the war in Mozambique. QEH Working Paper No. QEHWPS11. Oxford, UK, Queen Elizabeth House.

50 On the economic implications of infrastructure provision, see Y. Hayami and J.-P. Platteau. 1997. Resource endowments and agricultural development: Africa versus Asia. Cahiers de la Faculté des Sciences Économiques, Sociales et de Gestion - Namur, Série Recherche No.192, 1991/12. Namur, Belgium, Centre de Recherche en Économie du Developpement; R. Ahmed and N. Rustagi. 1984. Marketing and price incentives in African and Asian countries: a comparison. In E. Dieter, ed. Agricultural marketing strategy and pricing policy. Washington, DC, World Bank.

51 Wartime diversion of transport vehicles and equipment from rural trading activities has long been acknowledged as one of the features that determines a strong relationship between war (even where this is not a civil war) and famine. See: J. Illiffe. 1987. The African poor: a history. African Studies Series No. 58. Cambridge, UK, Cambridge University Press.

52 M. Chingono. 1995. The state, violence and development: the political economy of war in Mozambique. Aldershot, UK, Avebury.

53 P. Collier. 1999. Doing well out of war. Paper prepared for the Conference on Economic Agendas in Civil Wars, London, 26-27 April 1999.

54 W. Reno. 1998. Humanitarian emergencies and warlord economies in Liberia and Sierra Leone. Paper presented at the conference on War, Hunger and Displacement: the Economics and Politics of the Prevention of Humanitarian Emergencies, Stockholm, 15-16 June 1998. United Nations University/World Institute for Development Economics Research (UNU/WIDER). p. 17-18.

55 D. Keen. 1994. The benefits of famine: a political economy of famine and relief in Southwestern Sudan, 1983-89. Princeton, New Jersey, USA, Princeton University Press. See also Luckham, Ahmed and Muggah, op. cit., note 41. For an analysis of the role of agriculture in the origins of conflict, see: PRIO, op. cit., note 45; Stewart, 1998, op. cit., note 35; Collier, op. cit., note 53; and Cramer, op. cit., note 36.

56 World Bank, op. cit., note 33.

57 E. Muehlhoff. and M. Herens. 1997. Household food security and nutrition in agricultural relief and rehabilitation programmes. Food, Nutrition and Agriculture No. 19, p. 5.

Previous PageTop Of PageNext Page