c humanitarian emergencies and usaidâ„¢s humanitarian response

U.S. AGENCY FOR INTERNATIONAL DEVELOPMENT

USAID Program and Operations Assessment Report No. 27

Center for Development Information and Evaluation

December 2000

PN�ACG�605

COMPLEX HUMANITARIAN EMERGENCIES

And USAID�S HUMANITARIAN RESPONSE

This report and others in the evaluation publication seriesof the Center for Development Information and Evaluation(CDIE) can be ordered from:

USAID Development Experience Clearinghouse (DEC)1611 N. Kent Street, Suite 200Arlington, VA 22209�2111Telephone: (703) 351�4006Fax: (703) 351�4029E-mail: [email protected] access CDIE documents from the Internet, key inwww.usaid.gov. Click on Publications/PartnerResources, then on USAID Evaluation Publications.

The CDIE Evaluation Publications Catalog and notices ofrecent publications are also available from the DEC.


The views and interpretations expressed in this report are those of the authorand not necessarily those of the U.S. Agency for International Development.

USAID Program and OperationsAssessment Report No. 27

Complex Humanitarian EmergenciesAnd USAID’s

Humanitarian Response

By

Donald G. McClellandCenter for Development Information and Evaluation

with

Elizabeth AdelskiRichard HillJohn Mason

Robert MuscatInternational Resources Group, Inc.

Center for Development Information and EvaluationU.S. Agency for International Development

Washington

December 2000

ContentsPreface ........................................................ v

Summary ................................................ vii

1. Introduction ........................................ 1Number of Emergencies

And People Affected ....................... 1Resource Implications ........................ 2Legislative Authority .......................... 4Complex Humanitarian

Emergencies ....................................... 5General Evaluation

Approach ............................................ 6

2. Country Context ................................. 9Causes of Complex

Emergencies ..................................... 9USAID’s Humanitarian

Response ......................................... 12

3. Results: Humanitarian Effects ........ 15Haiti .................................................... 15Mozambique ...................................... 17Rwanda .............................................. 19Bosnia–Herzegovina ......................... 20Targeting ............................................ 21Efficiency ............................................ 22Conclusion ......................................... 23

4. Results: Political Effects andEffects on Hostilities ....................... 25

Haiti .................................................... 25Mozambique ...................................... 26Rwanda .............................................. 27Bosnia–Herzegovina ......................... 28Policy Implications ........................... 29Conclusion ......................................... 32

5. Results: Economic Effects ............... 33Haiti .................................................... 34Mozambique ...................................... 35Rwanda ............................................... 36When Is the Emergency Over? ........ 38Conclusion ......................................... 40

6. Conclusions, Lessons Learned,And Recommendations .................. 43Conclusions and

Lessons Learned ............................... 43Recommendations ............................ 46

Annex A Humanitarian EmergenciesAnd Donor Assistance

Annex B Evaluation ObjectivesAnd Constraints and ImplicationsFor Donor Coordination

Annex C Are Complex EmergenciesPredictable?

Annex D Implications for the Kosovo Crisis

Bibliography

OMPLEX HUMANITARIAN emergen-cies are by their nature multifaceted

and involve many actors. The relief inter-ventions are often undertaken in a contextbeyond the control of the implementingagencies. Moreover, the interventions aregenerally conditioned by overall foreignpolicy considerations, which means thatpolitical objectives help define the re-sponse.

Operational coordination is compli-cated because multiple players are in-volved (various U.S. government agencies,other bilateral and multilateral develop-ment agencies, nongovernmental and pri-vate voluntary organizations, and the hostcountry) and these actors often have di-vergent approaches to strategic planning,decision-making, and delivery mecha-nisms. That relief experts have differentviews of the purpose of emergency assis-tance—whether it is for relief only, or forrehabilitation and economic development

Preface

as well—exacerbates an already complexsituation.

Given these circumstances, it is notsurprising that humanitarian assistancehas been subjected to less rigorous andextensive monitoring and evaluation thandevelopment assistance. In addition to thecomplexities just noted, this reflects thefact that until recently there was no stan-dard methodology for evaluating humani-tarian assistance. Some have likened thissituation to “methodological anarchy.”

Nevertheless, it is possible to assessthe impact of humanitarian assistance onvulnerable populations to some degreeand to shed light on the relationship be-tween emergency assistance and the po-litical and development processes at work.This assessment and its lessons learnedshould contribute to formulating more ef-fective policies and interventions in re-sponse to complex emergencies.

C

SummaryN 1998, some 32 million people neededhumanitarian assistance because they

were caught up in complex emergencies(armed conflicts or civil wars as distinctfrom natural disasters). That is triple thenumber of a typical year from the early1980s. Most of these people are refugeesor internally displaced persons. About 40percent reside in Africa. The value of hu-manitarian assistance worldwide has typi-cally been less than 2.0 percent of officialdevelopment assistance (ODA). But in 1994it jumped to 6.8 percent ($4.3 billion) ow-ing to the crisis in the Great Lakes regionof Africa. It fell to an estimated 5.7 per-cent of ODA in 1998 but is likely to peakagain in 1999 owing to the Kosovo andEast Timor crises. The fact remains that hu-manitarian assistance has more thandoubled since 1990 despite diminishingforeign assistance.

In 1990, U.S. ODA totaled nearly $13.6billion in real terms (1998 dollars); by 1997,it had fallen by half to $7.0 billion, the low-est level since World War II. By contrast,U.S. humanitarian assistance has in-creased. In 1990, it was $263 million (1.9percent of ODA). In 1994 it peaked at $1.2

billion (11.4 percent of ODA). By 1997 it hadgradually decreased to $344 million (4.9percent of ODA), but in 1998 it more thandoubled to $898 million (10.2 percent ofODA). Although U.S. ODA as a percentageof total ODA has been falling steadily sincethe 1970s, the United States continues tobe a generous provider of humanitarianassistance. USAID’s Food for Peace Office,Office of Foreign Disaster Assistance, andOffice of Transition Initiatives are prima-rily responsible for administering U.S.emergency assistance.

This evaluation seeks to assess the ef-fectiveness of U.S. humanitarian assistancein nations afflicted by complex emergen-cies. It addresses three principal questions:Did U.S. emergency assistance save livesand alleviate suffering? Did it affect socialtensions and political hostilities? Did itcontribute to long-term economic devel-opment? The findings are based on field-work carried out in three countries (Haiti,Mozambique, and Rwanda) as well as onevaluation results in other countries.

Evaluating relief programs in the con-text of armed conflict or civil war intro-

I

viii Complex Humanitarian Emergencies and USAID’s Humanitarian Response

duces politically sensitive issues concern-ing sovereignty, international law, theappropriate balance of aid between oppos-ing sides, and donors’ foreign policyinterests. This assessment treats the politi-cal effects of humanitarian assistanceindependently of U.S. foreign policy con-siderations. Nevertheless, these and otherissues peculiar to complex emergenciesmade the evaluation methodologicallymore difficult.

Complex emergencies are typicallypolitical in nature, characterized by vio-lent conflict (often war) and a breakdownof institutions. But their underlying causesvary. Predatory governance was the prin-cipal cause of Haiti’s complex emergency.By contrast, ethnic and ideological factorswere pivotal in Mozambique and Rwanda.In all three countries, poverty was a con-tributory factor. In Haiti, per capita incomewas $250 in 1994; in Mozambique, $80 in1986, the lowest in the world. Poverty wasjust as severe in Rwanda. Moreover, thedistribution of income and wealth washighly skewed in all three countries.Haiti’s 200-year history has been charac-terized by oppressive governments that fa-vored the rich at the expense of the poor.Mozambique was characterized by ahighly dualistic economy. In Rwanda,where the proportion of people living inpoverty increased from 40 percent to 70percent during 1990–93, a winner-take-allmentality has benefited a tiny elite at theexpense of the poor majority.

Civilians in all three countries suf-fered widespread and systematic human

rights abuses. Tens of thousands of refu-gees fled Haiti (often as boat people). Mil-lions fled Mozambique and Rwanda toescape indiscriminate terror. Hundreds ofthousands were the victims of wholesalemassacre or, in the case of Rwanda, geno-cide.

Donors, including USAID, respondedwith increased emergency assistance, bothfood and nonfood (water, seed, farmingtools, medical supplies). Nongovernmen-tal organizations were the main im-plementers of the humanitarian response.In Haiti, the international community wasfeeding 1.3 million people—one in sevenHaitians—each day, with the United Statesproviding 68 percent of the food. InMozambique in 1989, an estimated onethird of the population of 16 million de-pended on food aid for 60 to 70 percent oftheir food needs; again, the United Statesprovided about 60 percent of total food aidduring 1987–95. In Rwanda 1.3 millionbeneficiaries received emergency food aidin 1996–97.

What were the results? The assess-ment concluded that emergency assistanceprograms funded by USAID and imple-mented by U.S. nongovernmental agencies(NGOs) clearly helped save lives and alle-viate suffering—which, after all, is theiroverarching objective. Except for Haiti,though, data collection and monitoringwere not done (or were done poorly), so itis difficult to quantify results. In fact, mostevaluations of humanitarian assistance tella “mission accomplished” story but are

Summary ix

unable to substantiate that story with harddata.

Distributing relief supplies was aproblem to some extent in all three coun-tries. Food aid, in particular, was highlyvalued and became a source of violentcompetition—not only for its value as foodbut also as a source of political power forthose controlling access. There were re-ports of corruption, theft, and political orpersonal favoritism in food aid distribu-tion. And target populations did not al-ways receive timely and sufficient food.NGOs addressed these problems with vary-ing degrees of success. In Haiti they werehighly successful in limiting diversion to5 to 10 percent. In Mozambique, leakagewas typically 30 percent when the govern-ment was in charge of distribution, and atone point reached 50 percent. But after theNGOs took over, losses fell to under 5 per-cent. In Rwanda the military and formerpolitical leaders controlled much of therelief distribution. They were able to di-vert substantial quantities of food (morethan is usually the case in complex emer-gencies) from the intended beneficiariesfor their own purposes.

While no aid is apolitical, humanitar-ian assistance, in particular, can result insubstantial and unpredictable political ef-fects, since it is provided in the context ofconflict. Though designed to relieve suf-fering and promote peace, it sometimes,inadvertently, fuels, sustains, or worsenscomplex emergencies by making more re-sources available to warring parties. Thisis because aid does not just keep people

alive in a political vacuum but affects thelocal power structure and environment inwhich it is given.

In Haiti, massive quantities of emer-gency food aid reduced the probability offood riots during a period of political andeconomic stress and may have had adampening effect on political tensions; butit also may have contributed to a politicalstatus quo that enabled the de facto mili-tary regime to stay in power longer. InMozambique, external military assistanceprovided by the Soviet Union and SouthAfrica fueled the civil war; food aid, bycomparison, had relatively little effect onthe country’s political dynamics, althoughfood diverted to soldiers may have con-tributed to the war effort. In Rwanda,genocidal killers were mixed with legiti-mate refugees in camps; targeting becameproblematic, and substantial quantities offood aid were diverted by Hutu extrem-ists and militia resident in the camps. Thathad the unintended effect of prolongingthe conflict.

The notion that relief assistance canbe made more developmental in the con-text of ongoing armed conflicts is problem-atic. Unlike with natural disasters, duringcomplex emergencies there is no institu-tional framework to provide physical se-curity and political stability—both ofwhich are necessary preconditions for eco-nomic development. On the contrary, com-plex emergencies are often characterizedby a total breakdown of state institutionsand social and economic structures.

x Complex Humanitarian Emergencies and USAID�s Humanitarian Response

Nevertheless, emergency assistanceprograms can help shape the pattern anddirection of subsequent economic devel-opment. In Haiti, Mozambique, andRwanda USAID and the NGOs not only pro-vided immediate relief (food, medicine)but also agricultural inputs (seed, tools)and household goods to encourage refu-gees and internally displaced persons toreturn to their villages, resume food pro-duction, decrease their dependence onfood aid, and maintain their livelihoods.They also implemented food-for-workprograms in all three countries. These andother programs created short-term jobsand helped rehabilitate productive infra-structure (roads, irrigation) needed foreconomic development. The development-oriented objectives were clear: to restartsubsistence agriculture and to restart therural economy.

The assessment offers 4 management-oriented recommendations (summarizedbelow) and 18 recommendations specificto the Kosovo crisis as of May 1999(annex D).

■■■■■ Monitoring and evaluation. Establish acentral monitoring and data collection unitto serve all donors during the early weeks ofa complex emergency. This is needed,among other things, to help managersidentify appropriate kinds of emer-gency relief, target its distribution,evaluate its effectiveness, and enhancedonor coordination.

■■■■■ Adverse political consequences. Be alertto potential undesirable political or social ef-fects that relief aid may cause. Control of thedistribution of food aid, in particular, canreinforce the power of local authoritiesor political factions; it can also facilitatetheir self-aggrandizing, often exploitive,behavior toward the intended noncom-batant beneficiaries.

■■■■■ Reducing dependency. Give refugees in-centives to return home, and impose disin-centives on those remaining outside theircountry of origin. After populations havebeen repatriated and are settled, the ag-ricultural base begins to be reestab-lished, dependency on free food drops,and long-term food security is en-hanced.

■■■■■ Capacity building. Train technocrats tomanage the postconflict economic transition,and train others in skills for which there isemployment demand. Economic recoveryrequires a cadre of high-level techno-crats with management and conceptualskills; it also requires the unemployed(especially demobilized soldiers) to betrained in marketable skills.

Finally, however one assesses the ef-fectiveness of humanitarian assistance inresponse to complex emergencies, onething cannot be emphasized too strongly:it is far better to prevent complex emer-gencies from occurring in the first placethan it is to respond to victims� needs af-terwards.

INCE THE END of the Cold War, the na-ture of international assistance needs

has changed dramatically. Ethnic and na-tional tensions have led to increased civilstrife and an explosion in the number ofcomplex humanitarian emergencies. As aresult, the number of civilian casualties hasincreased, as has the level of emergencyassistance allocated in response to their hu-manitarian needs. In 1998, USAID’s Centerfor Development Information and Evalu-ation (CDIE) initiated an assessment of theeffectiveness of the Agency’s humanitar-ian assistance interventions. The assess-ment examined USAID programs in threecountries afflicted with complex emergen-cies: Haiti, Mozambique, and Rwanda.This report synthesizes the findings of thethree separate country studies.

Number of EmergenciesAnd People Affected

According to the U.S. Mission to theUnited Nations (1997, 5), the number ofhumanitarian emergencies in which atleast 300,000 civilians depended on inter-national humanitarian assistance to avoid

serious malnutrition or death peaked dur-ing 1993–95. With improved situations inseveral countries (including Armenia,Cambodia, and Mozambique), the num-ber of emergencies dropped to 20 in 1996and remained at that level in 1997 (p. 5).In 1998, Russia (Chechnya) was droppedfrom the list (reducing the number to 19),but Colombia and Uganda were added(boosting it to 21) (U.S. Mission to the UN1998, 7, 9). Annex A (table A1) lists ongo-ing humanitarian emergencies in 1996,1997, and 1998 using data from the U.S.Committee for Refugees.

Worldwide, roughly 33 millionpeople needed emergency assistance inJanuary 1996. That increased to 34 millionin January 1997, then decreased to an esti-mated 32 million in April 1998 (table A1).These levels are triple those typical of theearly 1980s. They include both internallydisplaced persons who have remainedwithin their own borders and refugeeswho have fled across international bor-ders. During the 1990s most have been in-ternally displaced persons rather thanrefugees. In 1996, 52 percent of those re-quiring emergency assistance owing

S

Introduction1

2 Complex Humanitarian Emergencies and USAID’s Humanitarian Response

mainly to armed conflict or governmentrepression resided in sub-Saharan Africa;in 1997 that percentage decreased to 48percent, and in 1998 it decreased still fur-ther to 39 percent (table A1).

Resource Implications

The Development Assistance Com-mittee of the Organization for EconomicCooperation and Development monitorslevels of official development assistanceand humanitarian assistance. Table 1 re-ports these data for all donors in 1998 dol-lars during 1988–98. In real dollar terms,official development assistance declined

gradually from a high of $68.5 billion in1991 to a low of $53.4 billion in 1997, a de-crease of 22 percent. In 1998 it increasedfor the first time since 1994; the increasewas 8.1 percent. (The small increase in 1994was probably due to the Rwanda crisis.)

Humanitarian assistance peaked at$4.3 billion in 1994 owing to the crisis inthe Great Lakes region of Africa. After thatit fell to $3.3 billion as of 1998. However,it was expected to peak again in 1999 be-cause of the emergencies in Kosovo andEast Timor and the Turkish and Taiwan-ese earthquakes. Humanitarian assistancehad typically been less than 2.0 percent ofofficial development assistance—until

Source: OECD DAC/o database

Note: Annex A includes two figures that graphically depict dollar levels of foreign assistanceand humanitarian assistance (figure A1), and humanitarian assistance as a percent of overallassistance (A2) during the 30-year period 1969–98.

Year

19881989199019911992199319941995199619971998

Official DevelopmentAssistance

61,57060,17263,79168,50367,79262,65963,17656,96856,53053,42457,774

Humanitarian Assistance

955969

1,2703,5032,8803,8634,3033,4013,2062,9213,288

HA as a Percent Of ODA

1.61.62.05.14.26.26.86.05.75.55.7

Table 1. Official Development Assistance and HumanitarianAssistance, in Millions of 1998 US$, 1988–98

Introduction 3

1991, when it jumped to 5.1 percent (ow-ing to violence in the Balkans). It peakedat 6.8 percent of official development as-sistance in 1994 with the Rwanda crisis,but afterward fell to 5.7 percent of ODA in1998. The fact remains that within a dimin-ishing overall foreign aid budget, humani-tarian assistance has nearly tripled since1990.

What about the United States? In 1990,U.S. official development assistance to-taled nearly $13.6 billion in real terms(1998 dollars); by 1997, it had fallen by halfto $7.0 billion in real terms (see table 2).This was the lowest level of U.S. assistancesince World War II (Miller 1997, 1). U.S.

humanitarian assistance, by contrast, hasincreased. In 1990 it was $263 million (1.9percent of ODA); in 1994, it peaked at $1.2billion (11.4 percent of ODA). Since then ithas gradually decreased to $344 millionin 1997 (4.9 percent of ODA). In 1998 bothU.S. ODA and U.S. humanitarian assistanceincreased, and in 1999 humanitarian as-sistance was expected to increase againgiven current humanitarian needs.

Although U.S. official developmentassistance as a percentage of total ODA hasbeen falling steadily since the 1970s, theUnited States continues to be a generousprovider of humanitarian assistance. In1998, for example, the United States allo-

Source: OECD DAC/o database

Note: Annex A graphically depicts trends in U.S. ODA and U.S. humanitarian assistance in 1998dollars during 1971–98 (figures A3 and A4).

Year

19881989199019911992199319941995199619971998

Official DevelopmentAssistance

13,1419,547

13,58012,95113,14611,09910,6627,7439,6696,9598,786

Humanitarian Assistance

220261263685585733

1,216829603344898

HA as a Percent of ODA

1.72.71.95.34.46.6

11.410.76.24.9

10.2

Table 2. U.S. Development Assistance and HumanitarianAssistance, in Millions of 1998 US$, 1988–98


cated over 10 percent of its ODA to humani-tarian assistance needs. However, to meethumanitarian needs, U.S. policymakershave had to divert resources away fromsustainable development programs(Messer 1998, 15, citing USAID’s FY97 Con-gressional Presentation). Any proliferationof complex emergencies is likely to tightenthe squeeze on sustainable developmentprograms even further. Ironically, devel-opment assistance programs designed tospur economic growth and reduce povertymay help mitigate the need for more ex-pensive responses to complex emergen-cies, since they reduce the likelihood oftheir occurring in the first place.

Legislative Authority

The United States donates food aid tovictims of floods, earthquakes, droughts,and civil strife under Title II of the Agri-cultural Trade Development and Assis-tance Act of 1954 (PL 480). Other types ofhumanitarian assistance are provided un-der the Foreign Assistance Act of 1961 (asamended). Normally, the duration of bothfood and nonfood emergency assistanceis limited, and only countries that lack re-sources to purchase commodities commer-cially are eligible recipients (GAO 1986, 10).

USAID’s Office of Food for Peace ad-ministers the PL 480 Title II food aid pro-gram. Under this program the UnitedStates provides emergency food aid to co-operating sponsors who in turn distributeit to disaster victims. Food commoditiesare distributed in areas of greatest need

with priority given to people sufferingfrom malnutrition. Cooperating sponsorscan be (1) governments, (2) multilateral or-ganizations such as the World Food Pro-gram, or (3) nonprofit U.S. private volun-tary organizations (PVOs) such as Catho-lic Relief Services, Cooperative for Assis-tance and Relief Everywhere (CARE),Adventist Development and ReliefAgency, and World Vision. Cooperatingsponsors are responsible for establishingdistribution networks to reach disastervictims and for properly storing and ac-counting for commodities. USAID is re-sponsible for the overall administrationand management of the program.

An emergency response normally re-quires not only food but also nonfood as-sistance, including medicine, sanitation,potable water, agricultural inputs, andshelter. USAID’s Office of Foreign DisasterAssistance and Office of Transition Initia-tives are primarily responsible for admin-istering nonfood humanitarian assistance.OFDA coordinates the allocation of fundsappropriated under the International Di-saster Assistance account. The largest per-centage of funds goes to relief and reha-bilitation project grants managed by pri-vate voluntary, nongovernmental, and in-ternational organizations.

OFDA has an internal policy not to ob-ligate funds for longer than 12 months ata time. That allows it to respond to unex-pected crises worldwide. Food for Peacehas a similar policy for emergency foodaid. The Office of Foreign Disaster Assis-tance is widely respected for its respon-

Introduction 5

siveness to disasters. Although it ispermitted to provide responses aimed atrelief, rehabilitation, and reconstruction,OFDA’S lifesaving emphasis is paramount.OFDA is concerned about potential con-gressional criticism if it permits missionsor embassies to use emergency assistancefor longer term developmentally relatedinterventions (Miller 1997, 27).

Complex HumanitarianEmergencies

The term “complex emergency” wasfirst coined in UN circles, probably inMozambique, as a diplomatic euphemismfor a “chronic political” rather than “natu-ral” emergency. The Joint Evaluation ofEmergency Assistance to Rwanda (1996,study 2, 5) points out that complex emer-gencies

tend to have multiple causes, but areessentially political in nature and entailviolent conflict. They typically include abreakdown of legitimate institutions andgovernance, widespread suffering, andmassive population displacements, andthey often involve and require a rangeof responses from the international com-munity, including intense diplomacy andconflict resolution efforts, UN policingactions, and the provision of multilat-eral and bilateral humanitarian assis-tance by official and private agencies.A complex emergency tends to be verydynamic, characterized by rapidchanges that are difficult to predict.

Brandt (1995, 1) suggests that complexhumanitarian emergencies are often wars.

In the post–Cold War period, most havebeen conflicts that have taken place within,not between, countries. These internalstruggles among warring factions are usu-ally defined by ethnicity, religion, or lan-guage. According to Apthorpe (1997, 91),complex emergencies have deep roots anddense branches and cannot be understoodfrom a Western-ethnocentric perspective.Similarly, Kleist (1994, 45) suggests that adisaster may be defined as complex whenits origins are multiple and its effects com-pound one another. As USAID’s 1998 Per-formance Report (131) notes, complex emer-gencies are manifested by “armed conflict,death, displaced populations, hunger, andinjury.”

Humanitarian aid (donations of foodand other commodities and services) is in-tended to save lives in situations wherevirtually everyone is at exceptionally highrisk. These situations require getting as-sistance to where it is needed—urgently.The humanitarian response is primarily anact of rescue. Under the umbrella of hu-manitarian assistance, there is a distinctionbetween relief and rehabilitation (Kleist,47). Relief helps people survive; rehabili-tation helps people get back on their feetso they can reestablish their livelihoods.

Development aid, by contrast, is nor-mally long term and sustainable in nature.In economic terms, it can be characterizedas investment rather than consumption.There is no clear, operational definition ofwhen short-term relief ends and long-termdevelopment begins. For all practical pur-poses, though, relief activities end with the

termination of emergency resources. Andthis occurs when donors decide to ceaseproviding these resources. This suggeststhat the relief-to-development continuummay exist conceptually, but not operation-ally.

General EvaluationApproach

The overall objective of the CDIEassessment was to examine the effective-ness of U.S. humanitarian interventions,especially emergency food aid, in nationsafflicted by a complex emergency. Itaddressed three principal questions:

1. Did U.S. emergency assistancesave lives and alleviate suffering duringthe complex emergency (humanitarianeffects)?

2. Did U.S. emergency assistanceaffect social and political hostilities ortensions associated with the complexemergency (political effects)?

3. Did U.S. emergency assistancecontribute to long-term development(economic effects)?

The assessment examined the resultsof humanitarian assistance. It did not delveinto the various agencies involved inimplementation. In short, it did not evalu-ate the implementing agencies, but ratherthe results of their activities. The findingsare based on fieldwork carried out in threecountries: Haiti (McClelland 1999),

Mozambique (Lieberson 1999), andRwanda (Renison 2000). They also drawon syntheses of related evaluation resultsincluding in particular Apthorpe (1997)covering six evaluations of humanitarianassistance in Africa; Borton and Macrea(1997) covering 28 evaluations worldwide;and the UNHCR/WFP (1998) evaluation ofthe experience in Bosnia and Herzegovina.

CDIE evaluation teams carried out keyinformant interviews with beneficiariesand a broad range of experts who hadmanaged or implemented emergency as-sistance programs. Information was col-lected in both urban and rural areas; sitevisits within each country also producedvaluable insights. Illustrative questionsasked during the interviews included thefollowing:

■ What was the political, economic,and social context in which humanitarianassistance was provided?

■ What were the perceived results ofthe assistance in terms of saving lives,affecting hostilities, and contributing todevelopment?

■ Were the results achieved those thatwere intended (i.e., what was the relation-ship between results and objectives)?

■ Were there unintended effects,positive or negative?

■ Were the interventions sustainable(in the case of rehabilitation assistance asdistinct from relief aid)?

■ What were the key strong and weakpoints of the assistance; (i.e., the majorsuccesses and failures)?

The very nature of complex emergen-cies imposed certain methodological limi-tations on the study. For example, therewas little evidence of long-term develop-ment-oriented effects, since achievingshort-term effects was the principal objec-tive of the humanitarian assistance. Many

of the people involved (beneficiaries, man-agers, implementers) had moved on andcould not be reached, and some of theinstitutional mechanisms had been dis-mantled. Because action came first, paper-work was frequently not given priority, sorelevant data were lacking or conflicting.Finally, the contribution of the UnitedStates to operations that were cofinancedcould not be separated from the contribu-tions of other donors.

Country Context

HE TERM “complex humanitarianemergency” is relatively new in the

American lexicon. Complex emergenciesare generally characterized by a break-down of institutions and governance. Theyalways involve conflict, often war. Whatcauses them, and what humanitarian toolshas USAID used to alleviate widespreadsuffering? This section examines thesequestions in the context of the three coun-try case studies: Haiti, Mozambique, andRwanda.

Causes of ComplexEmergencies

Haiti. Most studies have singled outpredatory governance as the principalcause of Haiti’s complex emergency; eth-nic and ideological factors appear less im-portant. Haiti has almost no history ofdemocratic governance or strong public in-stitutions. Instead, during its nearly 200-year history, oppressive governments havefavored the rich at the expense of the poor.The country’s military has controlled a

subservient police, and both institutionshave engaged in widespread and system-atic human rights abuses with nearly com-plete impunity.

The situation boiled over during1991–94, shortly after a military coup re-moved democratically elected PresidentJean–Bertrand Aristide from office. Al-though never an all-out civil war, this pe-riod bore all the hallmarks of a complexemergency with political, social, and eco-nomic collapse. Human rights violationsswelled to unprecedented levels, prompt-ing a series of UN-backed sanctions in-cluding the U.S.–led international em-bargo. By September 1994, an estimated300,000 of Haiti’s 7 million people weredisplaced internally; another 60,000 to70,000 were refugees, some as the highlypublicized boat people (World Bank1998b); thousands had fled across the bor-der to the Dominican Republic; and 4,000had been killed (Dupuy 1997). Gross do-mestic product fell by 35 percent duringthis period and inflation increased to 50percent by 1994 (Buttari 1997). An esti-

T

2


mated 143,000 jobs were lost in the privatesector (Maguire 1996). The inflation-ad-justed value of the minimum wage wasless than it had been 10 years earlier, andper capita GNP was $250.

Mozambique. In 1975, after a 10-yearwar for independence, Mozambique in-herited a highly dualistic colonialeconomy that lacked schools, health facili-ties, and other public services. With theend of colonial rule, most of the Portu-guese and many skilled Mozambicansfled, leaving the country without the tech-nical skills needed to operate factories orthe transport system, to manage commerceor government, or to provide professionalservices. The Front for the Liberation ofMozambique (Frelimo), the insurgentgroup that had fought for independence,took control after the Portuguese left. TheFrelimo government established a one-party state and a centrally plannedeconomy modeled after those in EasternEurope.

Mozambique’s neighbors—white-ruled Rhodesia and South Africa, whichsupported apartheid—were alarmed by ablack-ruled, antiapartheid socialist coun-try on their borders ready to export revo-lution. In 1976, white Rhodesian militaryofficers opposed to the Marxist-leaningMozambique government formed theMozambican National Resistance(Renamo). Renamo guerrillas sought todisrupt Mozambique’s economy in an ef-fort to keep the new government from sup-porting guerrillas who were trying to over-throw the white Rhodesian government.

They cut railway and power lines, de-stroyed roads and bridges, and sabotagedoil-storage depots. They raided towns andvillages and sometimes engaged in thewholesale massacre of civilians. Mozam-bique’s socialist allies countered by pro-viding the Frelimo government withweapons and financial support.

The result was civil war—nominallybased on ideology but actually supportedby foreign countries in the context of ColdWar politics and fueled by a drive forpower by local military and political lead-ers. It was mainly a low-intensity, hit-and-run guerrilla war fought largely with smallarms and land mines to destroy economicand social infrastructure. Both armies ter-rorized the rural population by seizingfood and killing people.

Over 2 million people fled to neigh-boring countries and 4 to 6 million movedto areas of relative safety withinMozambique. As many as 8 million peoplein a country of 16 million were affected—a reflection of the large-scale human suf-fering and economic dislocations that tookplace. In 1986 the economy hit bottom: percapita GNP was $80, the lowest in theworld; real GNP growth was a negative 2.3percent; inflation was 41 percent. In 1992,after 16 years of fighting, General PeaceAccords were signed. The country’s firstdemocratic, multiparty elections were heldtwo years later. Why did the war end?Among the more important reasons wasthat foreign military support ended. As-sistance to Frelimo dropped sharply withthe collapse of the Soviet Union, and

Country Context 11

Renamo lost its external support whenapartheid ended in South Africa. In addi-tion, foreign (mainly Italian) intermediar-ies helped to bring Frelimo and Renamoto the negotiating table and to facilitatetheir reaching a settlement.

Rwanda. Rwanda’s wholesale geno-cide of 1994 was a desperate attempt bythe government and Hutu extremists toprevent the Tutsi-dominated RwandanPatriotic Front from seizing power. Basedin Uganda, the RPF had already tried totopple President Juvénal Habyarimanaand his Hutu-dominated government in1990. That unsuccessful effort set the stagefor a second attempt in 1993, which endedwith a cease-fire and later a peace agree-ment, the Arusha Accords. However, itsoon became clear that key elements of theaccords, including access to land and po-litical power sharing for the Tutsi livingin both Rwanda and Uganda, would notbe honored. This reignited the militarycampaign, and by July 1994 the RwandanPatriotic Front had defeated the army ofthe government of Rwanda.

In a 100-day period during April–July1994, more than 800,000 people were mas-sacred in a genocide historically un-matched in its intensity. The killing, or eth-nic cleansing, eliminated close to threefourths of the Tutsi population of Rwanda.The international community (includingthe United States) ignored, then acknowl-edged, the genocide, but did little to pre-vent it. The U.S. secretary of state apolo-gized for this failure to act in December1997, as did the president in March 1998.

Rwanda remained in a state of ten-sion and instability throughout 1995 and1996, as génocidaires came from neighbor-ing countries, particularly from Idjwi Is-land in Lake Kivu in the Democratic Re-public of the Congo (formerly Zaire). By1999 the Rwandan Patriotic Front had se-cured its borders and established securitythroughout most of the territory withinthem. (Some killing continues, primarilyin the northwest.) The bulk of the popula-tion in exile or refugee camps has returnedto Rwanda. However, there remains asmall group of Hutu-power extremists, de-termined to overthrow the current govern-ment and finish their work of genocide.Many of these génocidaires have foundsanctuary in the Democratic Republic ofthe Congo, the Central African Republic,and Angola.

Scholars have documented the exist-ence of a culture of political impunity inRwanda—including ethnically based masskilling that is sanctioned or planned bygovernment (Uvin 1998, Prunier 1995, Af-rica Rights 1998). Violence has been partof a winner-take-all mentality that hasdominated Rwanda’s governments duringthe colonial and postcolonial periods. BothHutu and Tutsi have used violence to ob-tain, and then maintain, absolute controlover political and economic decisions.

This mentality has benefited a tinyelite, exclusively and handsomely, at theexpense of the poor majority. During 1990–93, the proportion of Rwanda’s populationliving in poverty increased from 40 per-cent to 70 percent. During 1994–98, eco-


nomic activity declined sharply. As a re-sult, 75 percent of rural households (or 90percent of all households) currently livebelow the poverty line, compared with 53percent five years earlier. Structural adjust-ment programs had not pulled Rwandaout of its economic crisis.

But this is not surprising, since mostelements of these programs were notimplemented. That prompted the WorldBank and the International Monetary Fundto halt disbursements. According to theWorld Bank, “rising poverty undoubtedlyplayed some role in exacerbating socialtensions leading up to the genocide”(World Bank 1998, i). But as Uvin states,“structural adjustment did not cause these[economic] problems; rather, it was irrel-evant to their resolution” (Uvin 1998, 59).

USAID’s HumanitarianResponse

Haiti. The need for humanitarian as-sistance was sharpened by the economicembargo imposed by the internationalcommunity in response to the 1991 coupand the military’s subsequent political re-pression. USAID responded with an ex-panded program that included food aid,potable water, and health and sanitationassistance.

USAID increased Haiti’s PL 480, Title IIprogram by 60 percent, from $15.4 mil-lion in 1993 to $24.6 million in 1994. In1995 it was increased by another 37percent to $33.6 million (USAID/Haiti1992–96). At its peak, the international

community was feeding 1.3 millionpeople—one of seven Haitians—eachday at 3,100 distribution points through-out the country. It was also providingmost of the country’s health services(USAID 1995).

Private voluntary organizationsimplemented the humanitarian responsein Haiti. The Adventist Development andRelief Agency operated 1,100 feeding cen-ters in poor urban neighborhoods in Port-au-Prince and in northern and centralHaiti. CARE worked in the northwest andArtibonite regions through 1,200 schoolfeeding centers as well as hospitals, clin-ics, and other distribution centers. Catho-lic Relief Services operated 800 feedingcenters in the Port-au-Prince area and inthe south and southwest. InternationalLifeline implemented food aid programsfor two years, and the UN World FoodProgram, working through local and Eu-ropean nongovernmental organizations,provided food to vulnerable children,pregnant and lactating mothers, and thedestitute.

USAID’s Office of Foreign Disaster As-sistance awarded seven grants to privatevoluntary organizations totaling $5.4 mil-lion. Catholic Relief Services, the princi-pal grantee, distributed essential drugs,medical supplies, and agricultural inputs(tools, seed, and fertilizer) and contributedto UNICEF’s oral rehydration therapy andmeasles immunization programs. The Of-fice of Foreign Disaster Assistance alsosupported efforts to purify drinking wa-ter throughout Haiti, to purchase equip-

Country Context 13

ment for the Port-au-Prince municipal wa-ter system, and to buy fuel needed to trans-port emergency assistance to beneficiaries.The Pan American Development Founda-tion played a pivotal role in the humani-tarian response by implementing a $38million jobs creation project.

Finally, the Office of Transition Initia-tives provided $17.3 million in 1994–95 tosupport Haiti’s transition to democraticgovernance. It funded the demobilizationof the armed forces (not covered in the casestudy) and over 1,900 microprojects de-signed to bridge the gap between reliefand development.

Mozambique. In the mid-1980s, USAIDassistance to Mozambique was under $50million a year, with emergency aid a smallproportion of total aid. During 1988–91,as the civil war and humanitarian suffer-ing escalated, USAID assistance doubled toan average of $100 million a year. In 1989,an estimated one third of the populationdepended on food aid for 60 to 70 percentof their food needs. In 1992, in the finalthroes of the war, total U.S. aid doubledagain to $200 million annually. During1993–95, after the 1992 peace accords, U.S.assistance averaged $125 million a year, ofwhich emergency assistance was a largepart.

USAID’s relief-to-development pro-gram in Mozambique included several keycomponents: resettlement packages (food,seed, farming tools, household goods); re-building rural transport infrastructure;support for elections and civic education;

demobilization of the two armies; andmine clearance. The number of internallydisplaced emergency food aid beneficia-ries was reduced from 1.5 million in 1993to only 600,000 in 1995. During 1996–97,USAID assistance dropped to about $50 mil-lion a year, and the mission resumed itsemphasis on development.

Rwanda. USAID provided food andother types of emergency assistance toRwanda. According to USAID/Kigali, thevalue of the assistance was almost $118million in 1997 and over $56 million in1998. Most of it was food commodities.Beans, cornmeal, and vegetable oil wereprovided to genocide survivors, includingwidows, orphans, unaccompanied minors,and refugee-returnees. Most of the assis-tance was channeled through the WorldFood Program, Catholic Relief Services,the International Committee of the RedCross, and World Vision Relief and Devel-opment. The most vulnerable (childrenunder 5 and pregnant and lactating moth-ers) were reached through wet feedingprograms in nutrition centers, inpatientfeeding, and feeding programs in centersfor unaccompanied children and orphans.

The Office of Foreign Disaster Assis-tance simultaneously provided potablewater, sanitation, and health services (aswell as emergency food aid) along theroute of returning refugees. Part of thisassistance targeted orphans and unaccom-panied minors. For example, a grant to theInternational Rescue Committee helpedestablish transit camps for such children.


OFDA also assisted over 50,000 vulner-able farm families by providing seed,tools, and food rations for three monthsthrough the World Food Program. The in-tent was to jump-start agricultural produc-tion. The Rwanda Emergency Seeds andTools project, also OFDA funded, helped90,000 families for one month followingrepatriation. Another agricultural project,implemented by Food for the Hungry, In-ternational, distributed seed packages to25,000 vulnerable farmers. The project en-couraged farmers to move onto rehabili-tated marshlands, trained farmers in newpractices, and rehabilitated rural infra-structure. The Seeds of Hope project sup-

ported agricultural experts who identifiedappropriate seed stock, which then wasmultiplied.

Finally, OFDA provided $26 million tofund rapid-impact activities. These in-cluded a shelter program to help meet theneeds of some of the 1.3 million returningrefugees and to preempt a potentially un-stable security situation in the northwest.At the same time, the Office of TransitionInitiatives funded the Women in Transi-tion program (reaching over 162,000women) and various activities to educatelocal leadership and support local demo-cratic processes.

Results:Humanitarian Effects

AVING LIVES and alleviating sufferingare key objectives of humanitarian as-

sistance. When judged in terms of thesecriteria, most evaluations of humanitarianassistance tell a “mission accomplished”story. For example, Apthorpe’s review ofsix evaluations of humanitarian assistancein Somalia, the Horn of Africa, Rwanda,Liberia–Sierra Leone, and Sudan con-cludes as follows:

. . . despite the horrendous and horren-dously difficult circumstances, what weread in these consultancies on the wholeis that the humanitarian aid does actu-ally get through. Against all the odds,the job of getting it there is actuallydone, if not always at the times sched-uled or as suitably composed asplanned. (Apthorpe 1997, 101–2.)

However, this message of overall suc-cess is highly qualified. Apthorpe writes:

All accounts appear to find that, shallwe say, making our own brave leap intothe quantitative blue yonder, normallymore than probably 90 percent of as-sistance has not failed to get throughand be duly delivered with, say, usually

very much more than probably 60 per-cent of this being duly distributed, if notto the intended beneficiaries, then atleast to their representatives. (Apthorpe1997, 97.)

What about the complex humanitar-ian emergencies in Haiti, Mozambique,and Rwanda?

Haiti

Emergency indicators. One indicator ofthe magnitude of Haiti’s complex emer-gency is the national food supply. Haitihistorically has had a structural food defi-cit that makes the country dependent onimports. The deficit increased during thecrisis years, 1992–94. Domestic productionwas reported at 90 to 94 percent of nor-mal, while commercial imports decreasedby one third. Food aid increased by anaverage of 29 percent during 1993–95, butthis was insufficient to compensate fordecreased domestic production and com-mercial imports. As a result, Haiti’s fooddeficit increased to an estimated 20 per-cent of national food requirements, nearly

S

3


three times the deficit in a normal year(World Bank 1998a, WFP 1998).

Another indicator is food prices. Foodprices in Haiti rose sharply as commodi-ties became scarce. In Port-au-Prince, riceprices increased by 126 percent during1991–94; bean prices, by 167 percent; andcorn prices, by 184 percent (USAID/Haiti1992–94). Although average food pricesdecreased during 1994–96, they were stillmore than twice their 1991 levels.

Malnutrition rates of children under5 is a third key indicator of the severity ofa complex emergency. Most studies agreethat Haiti’s historically high malnutritionrates increased in 1991–94 owing to thecombination of economic stress and a to-tal breakdown of the public health sector(IDB 1994, Ianotti 1997, World Bank 1998a).Data from OFDA’s monitoring reports alsoshow the trend of increased malnutritionbut indicate the changes may not havebeen significant. Nationally, nutritionalstatus (based on weight-for-age) of 50 per-cent of Haitian children was normal in1992. This figure increased to 52 percentin 1993 but then declined to 49 percent in1994 and dropped still further to 47 per-cent in 1995. It then rose to 51 percent in1996 (USAID/Haiti 1992–96).

Of course, these national rates maskregional differences. For example, severemalnutrition at the national level increasedfrom 3.3 percent (1992) to 3.9 percent (1994)to 4.1 percent (1996). In the northwest,though, it increased from 11.7 percent(1992) to 14.4 percent (1994), then de-

creased to 10.8 percent (1996) (USAID/Haiti1992–96).

Humanitarian response. Internationaldonors increased food aid deliveries toHaiti by a third, primarily to address mal-nutrition. The United States contributed anaverage of 68 percent of total food aid(World Bank 1998a). The United States alsoinitiated programs to provide short-termemployment, agricultural inputs, fuel, andmedicine. Three major U.S. nongovern-mental organizations (CARE, Catholic Re-lief Services, and Adventist Developmentand Relief Agency, as previously noted)implemented most of the emergency as-sistance efforts funded by the UnitedStates. According to the NGOs’ figures, ben-eficiary levels nearly doubled during theemergency. In 1995, food aid was reach-ing 1.3 million direct beneficiaries, or 16percent of Haiti’s population (World Bank1998a).

Efforts to target Haiti’s vulnerablepopulations generally worked well, butthe problems of looting and armed theftwere always present. The Adventist De-velopment and Relief Agency, for example,distributed dry rations for only six monthsin the low-income neighborhood of CitéJasmine in Port-au-Prince in 1992 becauseof violence. Catholic Relief Services alsohad difficulties in urban areas. Overall,though, the estimated amount of leakagewas 5 to 10 percent, regarded as normal.

The Jobs Creation project was imple-mented during 1993–96 primarily to off-set the embargo’s economic pressures. It

Results: Humanitarian Effects 17

created almost half a million person-months of short-term employment duringits 34 months of operation, of which 20percent was for women (Brown 1996).Another USAID-funded activity supportedagricultural production and reduceddecapitalization of farm households. Thisproject loaned funds to farmers for seedand fertilizer and sold them tools at halfprice. Approximately 13,000 farminghouseholds and 47 farmers’ associationsparticipated (Naval 1995). USAID’s Officeof Foreign Disaster Assistance funded thepurchase of fuel needed to deliver emer-gency medical supplies, potable water,and food to more than 400 health centersthroughout the country.

Most studies agree that the embargoseriously exacerbated Haiti’s historicallyhigh malnutrition rates. But they also con-clude that emergency food relief alleviatedthat effect. And unlike the six evaluationsreported by Apthorpe, there is quantita-tive evidence to support this conclusion.The data from Haiti are approximate be-cause the sample populations and meth-ods of data collection were not standard-ized during the emergency. Nevertheless,they provide a basis for making an in-formed judgment about the humanitarianeffects of the emergency assistance. Thiswas not the case in Mozambique.

Mozambique

The United States was Mozambique’smajor donor during its complex humani-tarian emergency, contributing a total of

$636 million during 1987–95. Of this, $529million was food aid, which accounted for60 percent of total food aid provided dur-ing this period.

Indicators. The Mozambique evalua-tion team found no valid quantitative datato assess objectively the impact of U.S.emergency assistance. Although estimatedrates of malnutrition, mortality, and mor-bidity at the national level showed someimprovement, first in the late 1980s andthen again in 1994 after the emergency, itwas impossible to attribute these improve-ments to emergency assistance. NGOs re-ported having little empirical basis for tar-geting food aid because of the lack of sys-tematic information about food insecurityand nutritional status in rural areas. Bul-letins issued by Médecins Sans Frontières(Doctors Without Borders) containedhealth and nutrition information, but theywere produced only after 1992. Moreover,their data were based on small samplesand different methodologies and there-fore, according to Médecins SansFrontières, “must be interpreted with cau-tion.”

Notwithstanding the lack of accuratequantitative data, there was consensusamong donors, relief workers, Mo-zambican government officials, andMozambicans who received food aid thatthe assistance aided people’s survival dur-ing the emergency. All agreed that manymore people would have suffered anddied without food aid, although it wasimpossible to estimate the number of livessaved. Consistent with Apthorpe’s six-


evaluation review, “Mission accom-plished.”

Effectiveness. However, the impact ofemergency assistance varied greatly de-pending on (1) where the beneficiariessought refuge (within Mozambique or innearby countries such as Malawi), (2) whodelivered the assistance (the relief unit ofthe Mozambican government or NGOs),and (3) when people were uprooted andreceived assistance (before 1987, during1987–92, or during 1992–95).

Mozambicans who fled to nearbycountries had a hazardous journey butgenerally received adequate food andmedical care once they reached the refu-gee camps. They were also relatively se-cure from Renamo or Frelimo harassment.By contrast, those who were internally dis-placed within Mozambique received lessadequate relief food that was supplied ir-regularly. And they were often threatenedand harmed by Renamo or Frelimo sol-diers.

The internally displaced who re-ceived aid directly from NGOs or the WorldFood Program reportedly received moreadequate and regular supplies than thoseaided by the Mozambican government.According to former internally displacedbeneficiaries, the government did not pro-vide enough emergency food aid, and usu-ally there was a two- to four-month timelapse between distributions. They also re-ported that the government’s distributionswere unfair, often influenced by the recipi-

ents’ political affiliation and social status.Former internally displaced persons re-ported that everybody was hungry dur-ing the war years, so those who receivedless food preyed on those who receivedmore, and when quantities were insuffi-cient, people in some areas knifed opensacks and fought for a share.

Access to assistance by the internallydisplaced also varied depending onwhether they were in Frelimo- or Renamo-controlled areas. U.S. policy was to pro-vide emergency assistance only to govern-ment- (Frelimo-) controlled areas—exceptfor limited quantities provided to Renamoterritory through the International Com-mittee of the Red Cross. Regardless of itssource, food aid attracted both Frelimo andRenamo soldiers. People from several vil-lages said they lost their food aid toFrelimo by day and to Renamo by night.

Finally, the effectiveness of emergencyassistance varied over time. Before 1987,the war was disruptive, but few peoplehad to flee their homes. Relief efforts wererelatively small and localized. The warintensified during 1987–92. People fled tothe relative safety of neighboring countriesor the Beira corridor. (The corridor is aneast–west swath across the country’s waist;it was guarded by Zimbabwean troopsand thus served as a safe haven forMozambican civilians.) This was an espe-cially difficult period for internally dis-placed persons. After the 1992 peace ac-cords were signed, people began return-ing home. Resettlement packages (food,


tools, and services) were provided byNGOs through 1995. This final stage of theemergency was the smoothest.

Rwanda

In late 1996 and early 1997, some 1.3million refugees were repatriated toRwanda from the border camps, eithervoluntarily or by force. Massive starvationand human suffering would have occurredwithout substantial infusions of predomi-nately U.S. emergency food aid. As inMozambique, though, this is difficult toquantify because data were not systemati-cally collected.

Monitoring. Following repatriation ofrefugees, the World Food Program carriedout a six-month general-distribution foodprogram. This was intended to last onlyuntil the harvest in June 1997 but was re-instated in November 1997 in five prefec-tures. In those areas local food prices hadincreased by a factor of three, signaling asignificant food shortage. Starvation anddeath were on a sharp incline, and emer-gency food aid continued to be provideduntil the next harvest in June 1998.

Food insecurity was especially acutein Ruhengeri Prefecture owing to politi-cal instability as well as poor harvests andhigh prices. As many as 573,000 people ofan estimated population of 869,000 weredisplaced, living in camps and awaitingresettlement. The Ministry of Health dida nutritional survey in January 1999 basedon a sample of 900 children under 5 living

in camps in Ruhengeri. It found what Savethe Children Foundation/UK described as“alarmingly high rates of malnutrition, inparticular severe malnutrition among chil-dren 6–59 months.” Specifically, it found8.0 percent of children with acute malnu-trition, 4.7 percent with edema, 40.6 per-cent underweight, and 59.9 percent withchronic malnutrition. These 1999 figureswere substantially higher than those re-flected in a 1996 National Nutrition Sur-vey. Thus, the situation in Ruhengeri, atleast, was not getting any better—despiteemergency food assistance.

Targeting. Targeting assistance to theintended beneficiaries in pre-1996 Rwandawas mixed. In the Bukavu area in south-ern Zaire, government soldiers formedseparate camps from the very beginning.By contrast, in the Goma area the army,militia, and civilian refugees were allmixed together (forming “refugee–warriorcamps”), and the military and former gov-ernment leaders controlled relief distribu-tion. In Tanzania, the military was not asvisible among the refugees, but the mili-tia and former officials were. There waslittle security in these camps, and food andother relief supplies were diverted fromthe intended beneficiaries.

It was painfully obvious that the per-petrators of human rights abuses andgenocide were fed and assisted in thecamps. Médecins Sans Frontières believedthe only alternative was to leave the campsand suspend most services. By contrast,most other NGOs and the UN High Com-missioner for Refugees decided to stay.


They recognized the humanitarian im-perative to protect and assist the vastpopulation of refugees, even if that meantassisting people guilty of crimes againsthumanity (Joint Evaluation 1996, study 2,58–9).

Poor monitoring (which did not be-gin in earnest until 1998) also contributedto ineffective targeting. U.S. governmentofficials reported considerable double-counting of refugees by former militaryand government leadership, particularlyin the cross-border camps. As a result,many experts believe that more food aidwas supplied both inside and outsideRwanda during this period than was nec-essary—and that more food aid was mis-appropriated in Rwanda than is usual inemergency situations.

Interahamwe (Hutu militia respon-sible for the genocide) and former soldiersof the Rwandan army diverted food fromwomen and children for their own pur-poses. There was also evidence that camprosters were sometimes not updated toremove the names of those deceased,which resulted in the accusation that theUN High Commissioner for Refugees“feeds dead people.” Although this is an-ecdotal, there is little doubt that there wereabuses in the feeding program.

Bosnia–Herzegovina

The humanitarian assistance opera-tion in the former Yugoslavia was one of

the largest initiatives ever undertaken bythe international community. AlthoughBosnia–Herzegovina was not included asone of the country case studies for this as-sessment, the UN High Commissioner forRefugees and the World Food Programhad completed a joint evaluation of theassistance program in 1998 covering theentire period from 1992 to June 1997(UNHCR/WFP 1998, 1). The evaluation ex-amined the effectiveness of targeting in re-sponse to beneficiary needs, the impact ofthe emergency operation on the war itself,and the relevance of food aid in a periodof reconstruction—the same issues coveredin the Haiti, Mozambique, and Rwandaevaluations.

According to the joint evaluation, anaverage of 2.6 million people were reachedannually during 1992–96: 1.2 million in-ternally displaced persons and 1.4 million“war affected” (people who had no meansof support, although they were neitherrefugees nor displaced). By September1997, 1.14 million tons of food had beenprovided at a cost of $710 million. An esti-mated 80 percent of the population ofBosnia–Herzegovina had been beneficia-ries of food aid supplied by the WorldFood Program and the UN High Commis-sioner for Refugees at one time or another(UNHCR/WFP 1998, 2).

The evaluation concluded that therewas no widespread hunger or malnutri-tion in Bosnia–Herzegovina. However,security-related problems hindered access


and distribution to isolated communitiesand cities under siege. Air transport wasused when access by land was denied, asin Gorazde, Sarajevo, Srebenica, and Zepa.The Sarajevo airlift was the longest run-ning humanitarian air bridge in history,lasting from 3 July 1992 until 9 January1996. These operations were generally suc-cessful. In fact, it was only in Bihac dur-ing late 1994 and throughout 1995 that air-drops did not succeed in averting hunger(UNHCR/WFP 1998, 6).

Targeting

Efforts to target humanitarian aid tointended beneficiaries often run into ma-jor problems. Several ways to improve tar-geting and the overall effectiveness of hu-manitarian assistance programs involvealternative distribution channels, com-modity selection, and planning.

Commercial channels may offer an al-ternative to other types of distribution. Forexample, the World Food Program con-tracted with Somali merchants inMombasa, Kenya, to transport commodi-ties to targeted sites in Somalia. This in-volved selling the commodities to the mer-chants and then buying them back with a10 percent profit margin. WFP paid a 10percent markup to avoid having 60 per-cent of the food looted. This was becausebags printed GIFT OF THE USA were morelikely to be looted than commoditiesstored and transported by businessmen.

A variation of this mechanism was touse the proceeds from the sale of the com-modities to establish wage-generation pro-grams rather than buy back the commodi-ties. This gave people cash, creating a mar-ket that the merchants then supplied. Butcommercial channels are not always thesolution. In Mozambique, most privatetransporters refused to transport food aidbecause of bandits and land mines andalso because many that did had had theirtrucks stolen.

Military involvement in complexemergencies—both as protector and pro-vider of commodities—has been a mixedblessing. The tremendous costs of militaryoperations are generally disproportionateto the value of the commodities protected.Military humanitarianism can also getwrapped up with geopolitics and foreignpolicy objectives. In Liberia, for example,the Nigerian-dominated regional militaryforce sent to Liberia for peacekeeping pur-poses found itself in conflict with the larg-est rebel force in the country (Prendergastand Scott 1996).

Carefully selecting commodities foremergency assistance can reduce lootingand improve targeting. For example, riceand other high-value commodities aretypically much more attractive to lootersthan sorghum, maize, or blended foods.But this varies by region: substantial quan-tities of maize were looted in Somalia.Similarly, looters are rarely interested incooked food distributed in numerous


kitchens that are widely dispersed. In-stead, they are generally interested in com-modities for which there is market de-mand and which they can turn into cash.

Diversifying entry points for emer-gency supplies can help guard againstempowering a particular authority. InLiberia, for example, all commoditiescame into Monrovia rather than across theborder upcountry. By contrast, substantialfood assistance was delivered to affectedsites in Ethiopia (Eritrea and Tigray) viathe Sudanese border.

There is an important distinction be-tween distributing emergency commodi-ties to affected areas and to affectedpopulations within areas. Food aid was dis-tributed to Rwandan refugees in campsthrough prefectures, communes, and fi-nally cells; but it was not targeted to indi-vidual families. As a result, the emergencyaid perpetuated the authority of the mili-tary and political leadership that hadplanned the genocide.

Humanitarian aid is more easilydiverted when population figures are in-flated. For example, to achieve purely po-litical and economic objectives, warringfactions in Liberia and Rwanda overesti-mated the need for food. But underestimat-ing need may lead to violent competitionfor food. In assessing need, it is importantto understand people’s coping strategiesand their desire to preserve their liveli-hoods. This may be as important as en-suring short-term hunger alleviation(Borton and Macrea 1997, 27).

In planning emergency assistanceprograms, experts need to be sensible andbeware of over-complexity. It is alsoimportant to address capacity-buildingquestions early on. In anticipation of re-habilitating a collapsed health service, forexample, capacity building might take theform of training medical personnel. Ac-cording to Anderson (1996), capacitybuilding should be a central part of anyemergency response.

Finally, Prendergast and Scott (1996)point out it is important to plan up frontfor monitoring and evaluation: “A com-mitment to adequate, independent, andcontinuous monitoring and evaluation ofprograms may reduce aid’s contributionto conflict.” In Rwanda, as a result of moni-toring, diversion of food was reduced from120 tons per month to 5 tons per monthbetween July 1993 and January 1994. “It’smonotonous, boring, but critical in cuttingdown mismanagement” (Prendergast andScott 1996).

Efficiency

In addition to targeting, another mea-sure of the effectiveness of humanitarianassistance is efficiency. Borton and Macrea(1997) synthesized the results of a broadlyrepresentative sample of 28 evaluations,mostly of complex emergencies, under-taken since 1991 by bilateral donors, UNagencies, and the European CommunityHumanitarian Office (1–2, 12). They exam-ined humanitarian assistance in terms ofcost-effectiveness, which, unlike cost–ben-


efit analysis, does not involve the valua-tion of lives in economic terms. What theyfound is instructive: very few of thesestudies even considered the issue of cost-effectiveness—partly for methodologicalreasons and partly because of reservationsabout whether it should be a criterion forproviding humanitarian aid. According toKleist (1994, 301–02), the reality is thatmonetary costs are less important as anevaluative criterion than the number oflives saved and the security of the person-nel delivering the humanitarian assistance.

Transportation (rail, road, air) is amajor cost of emergency operations thatvaries enormously. For example, commer-cial air transport within the Great Lakesregion was approximately 4 to 5 timesmore expensive than road transport, and10 to 20 times more expensive than railtransport (Joint Evaluation 1996, study 3).Therefore, on efficiency grounds it is bet-ter to transport food to conflict areas byroad or rail rather than by plane. On theother hand, using ground rather than airtransportation could involve negotiatingwith rebel groups, thereby granting thema degree of legitimacy they otherwisewould not enjoy (Hallam 1998, 21).

Cost also varied according to distri-bution channel and type of commodity.Military channels were estimated to befour to eight times more expensive thancivilian channels (Borton and Macrea 1997,2, 23). Moreover, replacing high-value ricewith alternative less expensive cereals incoastal West Africa (bulgur wheat in Si-erra Leone and Liberia, and maize meal

in Guinea and Côte d’Ivoire) improvedcost-effectiveness (Apthorpe 1996). Asnoted above, using lower value commodi-ties also reduced the likelihood of diver-sion as well as the disincentive effect onlocal production (Borton and Macrea1997, 24).

Conclusion

The international community is gen-erally unable to assess with any degree ofcertainty the number of lives savedthrough humanitarian assistance providedin a complex emergency. The case of Haiti,where data were available, is the excep-tion; the case of Mozambique, whichlacked data, is the norm. Only 1 of the 28evaluations reviewed by Borton andMacrea (1997) attempted to estimate thenumber of lives actually saved by inter-national assistance interventions. This wasan analysis of the 1990–94 response to thecrisis in Somalia (Hansch 1994). It foundthat 330,000 Somalis were at imminent riskof death in 1992 and 1993. An estimated110,000 of these were sustained (that is,their deaths were averted) by health, food,and other interventions. At least 70 per-cent (154,000) of the famine-related deathsthat did occur in 1992 probably could havebeen prevented had primary health strat-egies been implemented earlier and morewidely (Hansch 1994, cited in Borton andMacrea 1997, 2, 25).

However, another analysis of the ef-fectiveness of humanitarian assistance inSomalia reports results with far less quan-


titative precision. It concludes that “sig-nificant numbers of lives were saved, se-vere malnutrition and vulnerability to in-fectious disease declined, the suffering ofthe displaced was eased, and refugeemovement was slowed—though these effectsare all difficult to quantify” (Kleist 1994, 305;emphasis added). According to Kleist, theinability to measure results with any de-gree of certainty reflects the fact that sav-ing lives is of the utmost priority. Requir-ing answers to detailed questions abouthumanitarian assistance provided duringa complex emergency could cause delayand cost human lives. Therefore, little up-front planning or data collection is done.In Somalia, for example, many proposalsfrom NGOs lacked such basic informationas who the target group was and where itwas located (Kleist 1994, 294–95). Notwith-standing inadequate information, donorstypically allocate resources to relief orga-

nizations to meet urgent humanitarianneeds.

This raises a question about the de-gree of planning and data collection thatis both feasible and desirable when animmediate humanitarian response isneeded to save lives. USAID’s Bureau forHumanitarian Response has determinedthat one of its strategic objectives is to en-sure that “critical food needs of targetedgroups are met.” Two indicators are speci-fied to determine if the objective has beenachieved: (1) the percentage of targetpopulations reached by food aid programsand (2) the impact of the assistance on thenutritional status of beneficiaries (USAID1998b, 4). However, data must be collectedfor these two indicators, and data collec-tion in the throes of a complex humanitar-ian emergency takes time when time is ofthe essence.

Results:Political Effects andEffects on Hostilities

UMANITARIANS TRADITIONALLYhave tried to remain impartial andthus apolitical. There are at least two

good reasons for this. The first is pragmatic:providing relief is often facilitated whenindigenous political actors perceive hu-manitarian agents to be without political,religious, cultural, or other agendas. Im-partiality helps humanitarians gain accessto victims. The second reason to remainimpartial is principled: each society has theexclusive prerogative and responsibility toshape its own destiny. Outsiders shouldnot interfere except in nondisruptive waysto save lives. By helping everyone and re-fusing to take sides, humanitarians placethemselves above the fray (Pasic and Weiss1997, 198–99).

Nevertheless, the increasingly obvi-ous reality is that humanitarian relief andits consequences are inevitably political,often in ways that are not self-evident. Thatis because aid does not just keep peoplealive in a political vacuum but also affectsthe local power structure and changes theenvironment in which it is given. Aid israrely neutral.

Each of the six evaluations reviewedby Apthorpe asked whether humanitarianaid had had the perverse effect of prolong-ing the war or contributing to the wareconomy; that is, whether such aid in ef-fect feeds conflict as well as its victims.Each concluded that although food andother humanitarian assistance are notmeant to feed conflict, they often do(Apthorpe 1997, 95). In short, humanitar-ian assistance can have adverse politicaleffects. Was this the case in Haiti,Mozambique, or Rwanda?

Haiti

The large quantities of aid-financedfood injected into resource-starved Haitiwere conspicuous and highly valued—ei-ther for direct consumption or as a politi-cal tool for those who controlled their dis-tribution. Control over access to food aidbecame a new source of tension andpower, and violent elements—local gangsor groups connected to political factions—hijacked food supplies. Fighting some-times erupted among beneficiaries whenfood was dropped off in urban neighbor-

H

4


hoods for distribution by volunteers. Somemunicipal mayors used their access to foodaid to favor supporters of one political fac-tion or another or to promote their per-sonal aggrandizement.

To their credit, the NGOs limited leak-age and diversion (estimated at less than10 percent) through regular monitoring,convoy protection, and timely adjustmentof their stocking and distribution methods.They also stopped distributions in someneighborhoods in Port-au-Prince, such asCité Jasmin. Violence, political exploita-tion, and local tensions consequently werereduced to manageable if not minor pro-portions.

Did the emergency assistance help thede facto regime in Haiti withstand diplo-matic pressures and the effects of the eco-nomic embargo—before the internationalcommunity finally resorted to militaryforce? Many Haitians believe this is thecase. In their view, exempting humanitar-ian aid from the embargo worked at cross-purposes with the policy of economicisolation. By reducing food distress theemergency assistance dampened publicpressure that might otherwise have risento uncontrollable levels against the regime.Thus, food aid may have permitted a de-lay in the intervention by external forces,intervention that finally proved unavoid-able in order to eject the Haitian militaryand return President Aristide to power.

This view implies that humanitarianaid, by extending the duration and extentof the emergency and the consequences for

its victims, ironically could have causedmore humanitarian distress than it allevi-ated. The point, of course, is speculative,and one cannot conclude with any degreeof certainty that humanitarian assistanceprolonged the conflict in Haiti.

Mozambique

The Reagan and Bush administrationsdebated whether to support the Frelimogovernment or the Renamo rebels whenMozambique’s civil war broke out in themid-1980s. Many American conservativesviewed the war as an ideological battleover communism and believed that theUnited States should therefore supportRenamo. This view was buttressed by thefact that the Frelimo government had in-stalled a socialist system and was receiv-ing support from its socialist allies. Oth-ers, however, believed the United Statesshould assist the Frelimo government inrecognition of its support of the antiapart-heid movement in South Africa. In the fi-nal analysis, the United States providedlimited humanitarian assistance toMozambique’s socialist government.

By the late 1980s, the Frelimo govern-ment had abandoned most of its socialistideology and initiated a program of mar-ket-based economic reforms supported byUSAID and the World Bank. In response,the United States greatly expanded its hu-manitarian assistance, much of it in theform of food aid. This U.S. assistance wasprovided on a government-to-governmentbasis, which meant it went only to inter-nally displaced persons in Frelimo-con-

Results: Political Effects and Effects on Hostilities 27

trolled territories. The only U.S. assistanceprovided to Renamo populations was sup-plied indirectly through the InternationalCommittee of the Red Cross.

What was the impact of emergencyfood aid on the length of the civil war inMozambique? Expatriates and Mo-zambicans alike consistently reported thatforeign political and military support forFrelimo and Renamo—rather than hu-manitarian assistance—was the primaryresource that fueled Mozambique’s 16-year civil war. U.S. humanitarian assis-tance had relatively little influence on thecourse of the war when compared with themilitary assistance provided by the SovietUnion (to Frelimo) and by South Africa (toRenamo).

Nevertheless, both Frelimo andRenamo soldiers tried to steal food aid byintimidating PVO workers and hijackingtrucks. Food aid distributions in rural vil-lages were a magnet for looting by bothmilitaries. Thus, food aid helped supportthe military forces to some extent, but theeffect was relatively small. As in Haiti,emergency food aid in Mozambique wassometimes politicized. Politicians at boththe national and local levels reportedlyused their influence over food aid distri-butions to favor particular factions andreinforce their political power.

Rwanda

The massive influx of Rwandans intorefugee camps in former Zaire was notonly a movement of people but also a

transplantation of a well-organized politi-cal, social, and security structure. TheHutu extremists (ex-Rwandan army regu-lars, former government officials, and al-lied militia) planned to use the refugeecamps as a staging area for their eventualreturn to political power throughRwanda’s northwest. They assumed, cor-rectly, that the international humanitarianrelief agencies (and the national govern-ment and regional authorities of Zaire)would not separate them from bonafiderefugees. This meant they could consoli-date both military and political controlover most of the camp population. In De-cember 1994 a new government ofRwanda in exile was declared, and incur-sions from the camps into Rwanda began.

Mixing bonafide refugees with thosewho were probably guilty of genocide andother high crimes was seen by the UNHigh Commissioner for Refugees, the In-ternational Committee of the Red Cross,and the international humanitarian com-munity as a conundrum, difficult to re-solve in the context of maintaining neu-trality and without military intervention.As a result, UNHCR did nothing. It was im-portant to maintain stability in an inher-ently unstable situation, and separatingthe refugees from the Hutu extremists wasconsidered risky. Moreover, UNHCR didnot believe the new government ofRwanda would welcome the refugees backhome, certainly not those involved in thegenocide. The Hutu refugees, themselves,understandably feared retribution (“re-verse genocide”) if they returned, whetheror not they were guilty of crimes and


atrocities committed in Rwanda. Thesefears and concerns helped create a dead-lock that lasted over two years.

The inability of the international com-munity to resolve the deadlock gave theHutu extremists in the refugee camps afalse sense of enhanced legitimacy. Theyused this opportunity to regroup, rearm,and revitalize themselves with food ra-tions intended for refugees—all in the rela-tive safety of the camps. As indicated ear-lier, once the camps became militarized,targeting became problematic. Beneficiaryfigures were significantly inflated. Foodaid was diverted to and consumed by theex–Rwandan army regulars and inter-ahamwe militia resident in the camps.That had the unintended effect of prolong-ing the conflict. According to the U.S. am-bassador to the United Nations, BillRichardson, “the failure of the interna-tional community to respond adequatelyto both genocide and the subsequent mix-ing of genocidal killers with the legitimaterefugee population in the former easternZaire only served to prolong the crisis”(October 1996). It was not until 1996 thatUSAID became sufficiently concerned andceased providing food aid to the WorldFood Program for use in the camps. Theidea of providing humanitarian aid to theplanners and implementers of genocidewas seen as inconsistent with the statedobjectives of humanitarian aid.

The crisis is not yet over. Rwandanex-political and ex-military leadership isusing former Zaire as a staging ground to

destabilize and overthrow the present gov-ernment of Rwanda. Their objective is tocomplete the unfinished work of genocide,using a campaign of propaganda and ter-ror to destroy the political and economicstructures of the northwest (and beyond)and to gain support of the local Hutupopulation. In 1997, 30,000 to 40,000 sol-diers began arriving in the northwest,while several thousand remained in Zaireto maintain the camps as a base of opera-tions. In 1998, several commune offices inthe northwest were looted and burned,and the officials were murdered or terror-ized. By March 1999, many services hadground to a halt. Water sources had beendestroyed and health problems multiplied.The government of Rwanda requestedmassive food assistance from the WorldFood Program. In response, USAID is pro-viding both development assistance andhumanitarian assistance to the northwest.

Bosnia–Herzegovina

The conflict in former Yugoslavia alsoraises concerns about emergency aid en-abling protagonists to prolong the conflict.As early as 1994, some evidence, thoughinconclusive, suggested that aid supportedtroops, thereby releasing the authoritiesfrom responsibilities they might have hadto civilians (Prendergast and Scott 1996,11, citing Minear 1994a). This earlier evi-dence was examined by the recent UNHigh Commissioner for Refugees/WorldFood Program joint evaluation of theBosnia experience (UNHCR/WFP 1998).


This joint evaluation offers two pos-sible arguments supporting the thesis thathumanitarian aid to Bosnia helped pro-long the conflict (UNHCR/WFP 1998, 3).Both arguments continue to be a source ofcontroversy. The first suggests that by giv-ing generous support in the form of foodaid, donors were able to defend them-selves against the charge of inaction. Butdoing so in effect postponed the militaryintervention that ultimately was neededto end the conflict. This reasoning paral-lels that of the Haiti evaluation, and asnoted there, is purely speculative.

The second argument suggests thathumanitarian assistance prolonged thewar in Bosnia because it was diverted tothe combatants and thus supported theirmilitary efforts. Even if humanitarian sup-plies were not diverted to combatants, theaid still would have allowed resources oth-erwise needed to sustain the noncomba-tant population to be used instead to sup-port the war effort. (Of course, combatantorganizations in some complex emergen-cies have shown little concern over the con-dition of “their” noncombatants.)

It is possible that without food aid theconsequent civilian suffering might havehastened the cessation of hostilities inBosnia. However, while the conflict mighthave been shorter, the suffering wouldlikely have been greater. Moreover,UNHCR/WFP suggests that the outcomewould likely have been unsatisfactory: aworld without Bosnia, with the countryinstead divided among its neighbors. Ac-

cording to UNHCR/WFP (1998, 7), there isno convincing evidence that the increasedsuffering that would likely have occurredin the absence of humanitarian supportwould have been justified by an increasedprospect for a swifter and satisfactory out-come of the conflict.

The provision of aid did have someunavoidable negative political effects.First, since the authorities on the groundcontrolled distribution of the assistance,bargaining with them and agreeing to usethe channels they controlled inevitablyreinforced their authority. Anderson (1996,3–4, 16–17) suggests that this is not unusualand that aid agencies often must negoti-ate with army leaders to gain access tocivilian populations or to hire armedguards to protect the goods they bring.Second, in some cases military authoritieslevied food taxes to allow convoys to passand seized food when they were notpaid, sometimes at gunpoint (UNHCR/WFP1998, 7).

Policy Implications

This assessment of the political effectsof humanitarian assistance raises key ques-tions: Should aid be given if some of it isbeing diverted to armed participants in theconflict? Should aid be distributed throughlocal structures if these are consideredpredatory or biased? USAID clearly wouldnot operate a development assistance pro-gram if security deteriorated to the degree


it did in Somalia, for example. Yet theUnited States can nearly always becounted on to provide emergency assis-tance, even under such volatile conditions.

Prendergast and Scott (1996) are sen-sitive to the possibility that humanitarianaid designed to relieve suffering and pro-mote peace often, inadvertently, fuels, sus-tains, or exacerbates such conflicts by mak-ing more resources available to warringparties. They like others recognize thathumanitarian aid may be given without apolitical agenda, but it rarely escapes hav-ing political consequences. Moreover, itcan be deliberately manipulated to serveas an instrument of war by providing ameans for sustaining the conflict. This canoccur by (1) manipulating access to the aid,(2) manipulating population movements,and (3) diverting or looting the aid. Ex-amples:

1. Manipulating access. Warring partiesoften manipulate humanitarian aid to en-hance their power over civilian popula-tions or weaken their opponents by deny-ing them food. Throughout the 1980s theEthiopian and Sudanese governments lim-ited the amount of aid going to rebel-heldareas. Warring factions also have manipu-lated access to aid in Bosnia, Liberia,Mozambique, Rwanda, and Somalia.

Relief agencies generally abide by theprinciples of neutrality and impartiality.They don’t take sides, and instead giveboth sides equal access to relief aid. Butproviding aid in rebel areas necessarilyhelps legitimize the rebel organizations,

because it enables them to feed the popu-lations they seek to control. The same istrue of the government and the agenciesthey mandate to distribute relief. As a re-sult, organizations (often NGOs) that at-tempt to provide humanitarian aid neu-trally may support forces that carry outviolence against civilian populations, as inMozambique. But the alternative may beno better: to subject civilian populationsto the double punishment of violence andhunger.

2. Manipulating population movements.Warring factions have used civilians asshields or “vehicles” to obtain food andother types of humanitarian aid. They po-sition civilians near airstrips to enhance theability of their troops or militia to remainin areas they otherwise would abandon forlack of supplies or difficulty in defending.For example, aid supplied to the refugeecamps in Zaire helped maintain the formerRwandan government’s control over apopulation that otherwise might have dis-persed or returned home.

3. Diverting or looting aid. Warring fac-tions tax, steal, or divert humanitarian as-sistance—especially food and drugs, giventheir easy monetization—for their ownconsumption, for barter, or for sale. Thisis a principal means to buy arms. Liberianrebels looted relief supplies and stole re-sources, especially vehicles and fuel to usefor hit-and-run campaigns. In Somalia,food distribution operations were lootedso often that as little as 12 percent of inter-national food aid destined for refugees in1986 reached the intended recipients (GAO


1986). Much of the remainder was takenby the Somali army and associated mili-tia. Agencies had to negotiate for their ownsecurity because there was no national se-curity force, and this led to diversion ofgoods, especially food (Kleist 1994, 298).Similar occurrences took place during thefamine in northern Ethiopia in the 1980sand the wars in Mozambique and south-ern Sudan (de Waal 1993).

Given this negative and widespreadexperience, under what conditions, if any,should humanitarian aid be halted?Prendergast and Scott suggest three fac-tors to consider in deciding whether tocease providing humanitarian assistance:lack of progress in peace negotiations; sup-port of undesirable political factions andhuman rights abusers; and danger to aidpersonnel.

During the civil war in Ethiopia in thelate 1980s, Lutheran World Relief urgedagencies seriously to consider withhold-ing aid if peace efforts failed (Prendergastand Scott 1996, 44). Should we pay end-less millions for humanitarian aid, theyasked, when this may only exacerbate andprolong the conflict? Perhaps we shoulddisavow the principle that food should notbe used as a political weapon and insteaduse it to force peace negotiations. How-ever, others have argued that political de-cisions should be separate from the basichuman right to humanitarian aid, and thatrelief should not be used as a politicalweapon. The rationale underlying the lat-ter point of view is summarized inPrendergast and Scott (1996, 44) in these

terms: “The people who don’t care and arenot affected are the rulers.”

Food and medical aid are particularlyvaluable to combatants. But aid agenciesrarely withhold such assistance despitehuman rights abuses and looting, andmost will not completely withdraw froman area unless the emergency is completelyover or the security situation is untenable.In Liberia during 1990–93, aid agencieswere subjected to harassment and robberyby warring factions. Diversion rates wereclose to 50 percent, according to some re-ports, and debate raged within the aidcommunity about how much aid was toomuch. All agencies eventually withdrewoperations until security improved. How-ever, it remains unclear whether the with-drawal contributed to the 1995 peaceagreement. And even if it did, was it atthe cost of more human suffering? Thereis simply no way to measure the effect ofstopping the aid.

De Waal points out that war has be-come synonymous with famine in muchof Africa. In fact, war is often designed tocreate famine. War, according to de Waal,has received far less attention than it de-serves—compared with other contributorycauses of famine such as drought, environ-mental degradation, and inappropriate de-velopment strategies. This brings intoquestion the tendency for donor govern-ments to fund humanitarian assistance butnot to address the underlying causes ofwar (Hallam 1998, 5). Humanitarian agen-cies are sometimes needed less than po-litical or military actors. In Rwanda in


1994, for example, a well-armed UNpeacekeeping force may have been able toprevent or mitigate the genocide. Themultitude of NGOs responding to humani-tarian needs once the genocide was overwas much less effective. The issue of con-flict prevention, as distinct from cure, isbriefly introduced in annex C.

Though war may be a principal causeof famine (as de Waal suggests), Sen (1993)provides a compelling argument that de-mocracy and a free press are great forcesin preventing famine. Sen points out thata government cannot ignore famine con-ditions if (1) it has to face reelection, (2) itcannot censor the terrible facts of starva-tion, disease, and death that accompanyfamines, and (3) it has to face criticismfrom opposition parties and newspapers.“It is not surprising that even though fam-ines have happened in colonial economiesand in modern authoritarian states, neverhas a famine occurred in a democraticcountry with a relatively free press” (Sen1993, 88).

Conclusion

While no aid is apolitical, humani-tarian assistance provided during complexemergencies can result in substantial andunpredictable political effects, since it is

provided in the context of conflict (Hallam1998, 12–13). The political effects ofhumanitarian assistance in Haiti, Mo-zambique, and Rwanda were mixed.According to most accounts, the assistanceclearly prolonged the conflict in Rwanda.In Mozambique, external military assis-tance, rather than humanitarian assistance,fueled the civil war for more than a de-cade. In Haiti, evidence of the political ef-fects of the assistance is inconclusive. InBosnia, airlifting food to Sarajevo probablyprolonged the war, but this does not meanthat the increased suffering that wouldhave occurred by withholding food aidwould have been justified by the possibil-ity of a shorter war.

Perhaps what is most important tokeep in mind is the underlying principlethat, at the very minimum, aid that is in-tended to help victims in war settingsshould not cause additional harm (Ander-son 1996, 6). The challenge, therefore, is tospecify, on a case-by-case basis, clear ob-jectives and to monitor closely the extentto which the humanitarian assistance isachieving those objectives. Allowing flex-ible implementation and encouraging ef-fective communication and coordinationamong other involved authorities (diplo-matic, military) is equally important.

Results:Economic Effects

HE RELATIONSHIP between short-termrelief and long-term development has

been viewed as a continuum in which re-lief operations, in response to a humani-tarian crisis, are followed by rehabilitationand then development activities (USAID1998a, 18–19). More recent literature ques-tions the utility of this concept of a relief-to-development continuum. According tothe Department of State’s Bureau for Popu-lation, Refugees, and Migration, relief anddevelopment assistance have significantlydifferent aims, and implementers of eachtype of assistance should address the ap-propriate aims. Messer (1998, 15) notes that“although relief officials try to make relieffunction as development assistance, the‘relief-to-development continuum’ theytalk about appears to be more wishfulthinking than fact. The bulk of emergencyfood assistance is devoted to meeting ba-sic human welfare needs.” Miller (1997, 15)believes too much emphasis and attentionhave been given to the concept of relief-to-development, given the relatively briefperiod when relief assistance overlapswith longer term development assistance.

None of the six evaluations reviewedby Apthorpe (1997, 92) concludes that re-lief is poorly done if it is not specificallyforwardly linked to development. Two ofthe six explicitly reject the linear linkageas too simplistic and do not see it as thebest guide to what is needed in relief aid.A third notes that the continuum conceptderives from natural disasters and there-fore has only limited application to com-plex emergencies, which are often politi-cal in nature. Apthorpe concludes thatmuch of the literature on the relief–devel-opment continuum has little operationalvalue.

Nevertheless, emergency programscan have an important effect on shapingthe pattern and direction of subsequenteconomic development. Societies recover-ing from disastrous conflict are in the pro-cess of remaking themselves, and the eco-nomic opportunities that emerge from thisprocess can be influenced by how emer-gency assistance is designed and deliv-ered. What was the experience in Haiti,Mozambique, and Rwanda?

T

5


Haiti

In Haiti, USAID emergency assistancesupported economic development in twomain ways: employment generation andagricultural production. In addition,USAID’s Office of Transition Initiatives pro-vided small grants to various organiza-tions to fund numerous microprojects.

Employment generation. The 1991–94crisis worsened rural poverty in a coun-try where poverty levels already wereamong the highest in the world. When theassembly plants in Port-au-Prince closedbecause of the U.S.–led economic em-bargo, 400,000 urban poor returned to thecountryside. That increased pressure onrural households’ scarce resources. USAID’sJobs Creation project was implementedduring 1993–96 primarily to generate em-ployment needed to maintain householdincomes and thereby offset the embargo’seconomic pressures. Its secondary objec-tive was to rehabilitate productive infra-structure.

The project created half a million per-son-months of short-term employmentduring 34 months. More than 120 indi-vidual projects were carried out, resultingin the repair of 1,000 miles of roads, 2,000miles of irrigation canals, and 4,500 milesof soil conservation barriers (PADF). Theproject achieved its main objective—em-ployment creation. However, maintenanceand long-term sustainability of the infra-structure was not a project objective, andover time the infrastructure has deterio-rated. Although more durable infrastruc-

ture could have been built, that wouldhave required purchasing materials (ratherthan hiring labor), and the primary objec-tive of generating employment may havebeen compromised.

Agricultural production. Another emer-gency activity, funded by USAID’s Officeof Foreign Disaster Assistance and admin-istered by Catholic Relief Services, was de-signed to support agricultural productionand reduce decapitalization of Haitianfarm households. The project loaned agri-cultural inputs (seed and fertilizer) tofarmers who were otherwise being forcedto sell their productive assets to buy food;it also sold them tools at half price. Some13,000 farming households (less than 10percent of all farmers) and 47 farmer as-sociations participated (Naval 1995). Bysupplying inputs needed for food produc-tion, the program provided emergencyassistance in a way that helped maintainbeneficiaries’ incomes and livelihoods andat the same time reduced their dependenceon short-term relief.

Microprojects. The Office of TransitionInitiatives typically funds programs de-signed to bridge the gap between short-term relief (often managed by the Officeof Foreign Disaster Assistance) and long-term development (typically managed byresident USAID missions). These programsoften inject cash into an economy to gen-erate employment or provide commodi-ties to meet peoples’ most pressing needsquickly. They are meant to create precon-ditions for development and at the sametime facilitate the phaseout of emergency

Results: Economic Effects 35

assistance. This is important, because suc-cessful rehabilitation is more difficult thelonger relief is provided (Kleist 1994,300, 307).

The Office of Transition Initiativesfunded an $11 million program in Haitithat supported more than 1,900 micro-projects over a 27-month period during1994–96. (This means that, on average,more than two microprojects were initi-ated each day during this period.) Theyranged from rehabilitation and construc-tion of community schools, roads, markets,canals, and bridges to the organization andimplementation of literacy, public health,sanitation, reforestation, and civic educa-tion activities. The projects were explicitlydesigned to deliver assistance rapidly,have high visibility, provide tangible ben-efits, and support the legitimacy of localgrass-roots organizations. According to amidterm evaluation, “there is absolutelyno doubt that this program has had un-precedented success in mobilizing highlyvalued resources to tens of thousands ofneedy beneficiaries all over Haiti” (Chan-dler 1996). The evaluation also notes, how-ever, that the program emphasized instal-lation more than maintenance.

In sum, employment generation ac-tivities provided short-term benefits butnot permanent, off-farm sources of in-come. Rehabilitated infrastructure contrib-uted to increased economic activity in theshort term, but links to long-term eco-nomic development were tenuous at best.This is understandable. Relief agencieswere working in an environment of social

and economic chaos in which the goal oflong-term sustainable development waseclipsed by the immediate short-term needfor relief. Haitians themselves were con-cerned primarily with physical securityand survival rather than development.Therefore, emergency assistance programsgenerally were designed with only inci-dental links to economic development, andthey had minimal developmental impact.Although it is desirable to incorporatelong-term development objectives whendesigning short-term emergency re-sponses, the Haiti experience highlightsthe difficulty of doing both well.

Mozambique

Apart from the Beira and Tete corri-dors, which were secured by the Zimba-bwean Army, there was no functioninginfrastructure (roads, bridges, or rail lines)or rural markets in Mozambique in 1990.They had been destroyed by years of civilwar. During the war USAID providedemergency assistance to save lives andalleviate suffering. When the war ended,assistance programs turned toward reha-bilitation and reconstruction. They hadtwo objectives: to restart the rural economyand to restart subsistence agriculture.Roads were demined, rebuilt, and re-opened. That helped restore the privatetransport sector and facilitated freemovement of goods (especially food) fromsurplus to deficit areas, which spurredmarket development. At the same time,assistance was provided to displacedfarmers to help them resume agricultural


production and reduce their dependenceon food aid.

Seeds and tools. Farmers needed seedsand tools to recapitalize their farms. Thetotal cost was small in absolute terms, lessthan $50 per household. But it was largerelative to per capita income. PVOs locatedseeds and tools (generally not available inlocal markets), purchased them, and madethem available to farmers. Food for theHungry International also provided agri-cultural inputs, conducted field trials toidentify higher yielding varieties withshorter growing seasons, and introducedimproved farming practices. Both pro-grams helped restart subsistence produc-tion.

Food for work. USAID graduallystopped providing relief food and begansupporting food-for-work projects. Doingso helped break the dependency mental-ity. The program supported labor-inten-sive rural road construction and rehabili-tation, construction of schools and healthclinics, and rehabilitation of small-scale ir-rigation works. The quality of constructionwas generally satisfactory, but mainte-nance was questionable. As economic re-covery continued, the food-for-workprojects evolved into cash-for-workprojects. That helped create sustainablemarket mechanisms for supplying foodand other consumer goods as the casheconomy developed.

Economic liberalization. In 1990 Mo-zambique’s rural markets were function-

ing poorly, and trade was limited. Civilwar and socialist economic policies hadtaken their toll. USAID and the World Banksupported efforts to reduce state controlof markets and prices and to promoteprivatization of state-owned enterprises.These measures helped establish the foun-dation for rapid growth of small marketsand increased activity of private traders.They also encouraged refugees and inter-nally displaced persons to resettle. At thesame time, USAID funded a commodityimport program that supplied importsneeded to support economic liberalization.These initiatives together with the pro-gram to demine and rehabilitate roadshelped open up trade in rural areas.

Rwanda

Approximately one million Huturefugees returned to Rwanda within a one-month period in late 1996. The humani-tarian community immediately was facedwith the difficult task of helping this massof humanity put their lives back together.The task of linking relief to developmentassumed new dimensions. USAID ad-dressed not only development needs in ag-riculture, health, education, and commercebut also the preconditions for develop-ment: political stability, physical security,justice, and legitimacy of the new govern-ment.

Seeds and tools. The Office of ForeignDisaster Assistance supported the distri-bution of seeds and tools in almost everyregion of Rwanda in 1995 and 1996. The


largely untargeted distribution was report-edly successful. In 1997 the size of the pro-gram was reduced, and by 1999 it was ter-minated, except in the two northwest pre-fectures where continued high levels ofinsecurity had led to widespread lootingof farm supplies. Although livestock tra-ditionally had been used to maintain soilfertility and provide an important sourceof nutrition to farm families, relief organi-zations (including OFDA) were slow to in-clude livestock in emergency packages.The conventional wisdom was that ani-mals represented a level of assistance farbeyond “emergency” requirements. How-ever, most families are now acquiring ani-mals, either through loan programs orwith their own savings.

Seeds of Hope. Agricultural expertsfrom the international agricultural re-search centers recognized that productivecropping in the various microclimates ofRwanda required adapted seed. Theyidentified appropriate seed and rootstockfrom their own seed banks as well as inRwanda where adapted seed still existed.Seed stocks were multiplied and madeavailable in 1995. Because the experts hadalerted nongovernmental organizationsabout the importance of planting adaptedseed, more local seed was used than oth-erwise would have been the case. Seedmultiplication is now being expandedunder Seeds of Hope II. At the same time,food-for-work programs were imple-mented to reclaim wet lowland farmingareas (marais) and to improve terracing andland productivity.

Capacity building. Most complex hu-manitarian emergencies have occurred incountries where the government was veryweak (as in Somalia). This meant nongov-ernmental organizations could operate inan environment relatively free from gov-ernment intervention. Rwanda was an ex-ception. The government preferred min-istry-administered programs rather thanNGO-administered programs, and it foundit difficult to incorporate NGOs into its pro-grams. Communication between the twodeteriorated, and in late 1995, 16 of about60 humanitarian NGOs were expelled orasked to suspend operations.

USAID was an early and strong sup-porter of the government’s effort to claimcontrol of relief and development pro-grams. Believing that a stable, fair, com-petent government was key to Rwanda’ssuccessful agricultural and economic de-velopment, USAID helped strengthen thegovernment’s capacity. Addressing justicein response to the genocide had high pri-ority. USAID funded the Rwandan-initiatedInternational Genocide Conference in1995; trained court clerks in the Ministryof Justice; supported a media campaign onthe genocide trial process; and developeda central database for genocide prosecu-tors. USAID also supported decentraliza-tion of the Ministry of Health and helpedthe ministry establish an emergency re-sponse unit. Finally, USAID provided ba-sic equipment to 10 ministries includingjustice, health, interior, and the president’soffice. Most NGOs now work more closelywith government officials and seek oppor-


tunities to help build government capac-ity.

Democratic initiatives. The Office ofTransition Initiatives has funded two ac-tivities designed to support decentraliza-tion and educate local leadership. One isthe Women in Transition program, whichhas reached over 160,000 women. It en-courages commercial interaction amongdifferent ethnic groups. As part of a broad-based program to support women inpostconflict situations, it also assistswomen’s groups in the northwest, wherethere has been an increase in the numberof women farmers. The other democraticinitiative is administered by Africare in theMinistry of Interior. It supports electioneducation by building local decision-mak-ing processes and grass-roots organiza-tions. The program operates in 15 com-munes in four prefectures.

All the activities described abovewere designed to meet immediate needswhile preparing for follow-on programs.In contrast to long-term development-ori-ented programs whose success depends ontheir being sustainable, this was not a cri-terion for success in Rwanda. As the evalu-ation states, “relief and transition pro-grams . . . do not have to be sustainable.”

When IsThe Emergency Over?

At some point the emergency endsand development resumes. But the demar-cation between the two is not always clear.

In Haiti, emergency food assistance (dryrations) was still being distributed in thenorthwest as recently as July 1998—yearsafter the end of the crisis. (However, it wasscheduled to terminate in September1998.)

In 1998, four years after the emer-gency in Mozambique ended, most of thevillage groups interviewed by the evalua-tion team in the Beira corridor were stillasking NGOs for free seed, tools, food, andeven tractors. Although the NGOs had in-formed recipients that free food would endby a specific date, many did not expect thatto happen. They remained in the refugeecamps until free food was actually termi-nated.

Donor pressure increased after theDayton Peace Accords in 1995 to reduceemergency assistance to Bosnia. The logicwas that peace would bring stability andeconomic recovery, making large reduc-tions in food aid possible. There were alsothe usual arguments about avoiding foodaid dependency and disincentives to ag-ricultural production. In response to do-nor demands, the joint UN High Commis-sioner for Refugees/World Food Programmission recommended reducing the num-ber of direct beneficiaries from 1.6 millionto 600,000 (from over 50 percent of thepopulation to about 20 percent as esti-mated by WFP in 1996) (UNHCR/WFP 1998,8). That would help ensure that food aidwas not seen as an alternative to a socialwelfare system. However, the joint evalu-ation also pointed out the importance ofnot scaling down too rapidly because an


effective social welfare system was not yetin place.

In 1996–97, as many as 118,000 farmfamilies in Liberia received food rationsas well as seed and tools under a programfunded by the Office of Foreign DisasterAssistance, the European Union, and theFood and Agriculture Organization tosupport the transition from war to recov-ery. This ensured that rice seed wasplanted rather than consumed and alsothat farmers had the energy to work. As aresult, rice production increased from ap-proximately 30 percent of prewar levels in1996 to 60 percent of prewar levels in 1997(USAID/Liberia 1998, 13). At the beginningof 1997, 350,000 beneficiaries were receiv-ing emergency food aid in IDP camps; bythe end of the year, 150,000 had been per-manently resettled in rural areas. The do-nors developed a plan to end general foodaid distribution in IDP camps after Febru-ary 1998 and to rechannel these resourcesto targeted activities in rural areas. Theseincluded rural resettlement of refugeesand internally displaced persons, agricul-tural recovery, school feeding, and food forwork.

In Somalia, as elsewhere, humanitar-ian assistance and related relief operationsgenerated substantial local employmentand purchasing power. Over 50,000 Soma-lis found cash or food-for-work employ-ment. But this all ended when the reliefended, and the relief ended with termina-tion of the donor-funded contracts withNGOs that provided the relief (Kleist 1994,295). This suggests that the amount of food

aid provided in a complex emergency isdriven not only by an assessment of needsbut also by the availability of donor re-sources. The implication is that complexemergencies may end too quickly (withhumanitarian needs still unmet) or notquickly enough—depending on what thevarious political interests stand to gain orlose.

Hill (1997) reports that many conflictsare unreconciled and have not ended. Thisis true, for example, in Afghanistan, Iraq–Kurdistan, Lebanon, and Somalia. Whena conflict does subside, it is for a reason.Sometimes it is because one side wins. Inthe post–Cold War era, though, this hasrarely been the case. More often it is be-cause the war is no longer profitable. Ofcourse, economic issues are not the onlymotives perpetuating complex emergen-cies. Religious or ethnic hegemony haveoften driven conflict (in Bosnia and SriLanka, for example). But whatever itscauses, war is expensive and must be fi-nanced. And for those who invest in thewar, it must be seen as profitable, at leasteventually; otherwise, they would ceasetheir support.

Economies in post–Cold War con-flicts, as with economies in all wars, re-volve around scarcity. With the disruptionof outside trade, loss of incomes, and sev-ered or restricted corridors for delivery ofgoods, food and other essentials becomevery expensive. Profits are enormous forthose who have access to scarce resourcesand can deliver them to areas where theyare needed. The combatants and their al-


lies are in the best position to manipulateand profit from this trade.

Relief organizations are not. On thecontrary, relief organizations attempting tooperate in conflict situations provide a richsource for exploitation. There are severalreasons. They have to operate with largeamounts of cash, which can be stolen. War-ring factions can exact exorbitant fees fromrelief organizations in return for provid-ing them protection and ensuring theiraccess to insecure areas (as in Somalia).And they can charge rents for warehousesthat are higher than market value. Someof the most obvious spoils of war derivefrom the blatant looting of infrastructure.In Bosnia, for example, an entireVolkswagen production plant nearSarajevo was dismantled and sold. In So-malia, almost every phone line, electriccable, and water line was taken or rippedup, put on a ship, and sold in some porton the Indian Ocean (Hill 1997, 5). Thus,peace can threaten a very profitable situa-tion for the combatants.

Can donors help end wars by dem-onstrating that more can be gainedthrough peace? According to Hill, the ba-sic elements of reconstruction include (1)identifying and creating markets formanufactured and agricultural goods; (2)reestablishing the rule of law, especiallyin economic issues; (3) increasing oppor-tunities for entrepreneurs; (4) supportingself-sustaining lending institutions; and (5)reestablishing acceptable levels of water,power, heat, sanitation, and other basicservices. Education, often overlooked, is

also important. External assistance poten-tially has an important role to play in allof these areas.

Conclusion

The notion that relief assistance canbe made more “developmental” or that itcan be linked to development activities ishighly problematic in the context of ongo-ing armed conflicts owing to the frequentlack of local social and economic structuresthat might legitimately be strengthened(Borton and Macrea 1997, 8). Unlike thecase of natural disasters, with complexemergencies there is no institutionalframework to provide security and justice,both of which are necessary preconditionsfor successful development activities. Onthe contrary, complex emergencies are of-ten characterized by (1) a total breakdownof state institutions (Somalia, Liberia, Af-ghanistan); (2) large areas of territory heldfor prolonged periods of time by rebelmovements (Eritrea, Tigray, south Sudan);or (3) a situation in which the occupyingregime had not received international rec-ognition or was subject to internationalsanctions (Sudan, Rwanda, Cambodia)(Borton and Macrea 1997, 30).

In the absence of physical security andpolitical stability, support of long-termeconomic development can be risky. InSudan, for example, water and health in-frastructure that had been rehabilitatedwas later destroyed by military action. InSri Lanka, the assets rebuilt by a WorldBank–led emergency rehabilitation and


reconstruction program in 1987–90 at a costof $125 million were destroyed when hos-tilities resumed (Borton and Macrea 1997,32). These examples reflect the unpredict-ability of warfare. They also cautionagainst prematurely supporting long-terminvestments in economic developmentbefore political stability exists.

This does not suggest that relief agen-cies blindly ignore development-oriented

opportunities that may arise. For example,many NGOs providing emergency assis-tance also routinely provide funds for ba-sic agricultural inputs, food-for-work pro-grams, and housing construction. But ex-perience suggests that “large rehabilitationfinancing may be more appropriately pro-vided after resolution of the political frame-work, rather than during the process ofpolitical transition itself” (Borton andMacrea 1997, 30–31).

Conclusions,Lessons Learned, andRecommendations

OME OBSERVERS CONSIDER each com-plex humanitarian emergency unique.

The implication is that past experience isnot applicable to future crises. By contrast,evaluation analysts tend to look for com-mon themes—even when assessing the ef-fectiveness of interventions in response tocomplex emergencies. Their underlyingpremise is that past experience is, in fact,relevant for future crises. On the basis ofthe three country case studies (Haiti,Mozambique, and Rwanda) and evalua-tions of other complex emergencies, atleast six common themes and four recom-mendations emerge. Implications specificto the Kosovo crisis are summarized inannex D.

Conclusions andLessons Learned

1. Saving lives. Emergency assistanceprograms funded by USAID and implementedby American nongovernmental organizationsappear to deliver sufficient assistance to ensurethe survival of a country’s vulnerable poor,

though inadequate monitoring makes it diffi-cult to quantify results.

One-half million to 1.3 million Hai-tians (as many as one in seven) receivedfood aid during 1991–96. In Mozambiquean estimated one third of the populationof 16 million depended on food aid for 60to 70 percent of their food needs in 1989.In late 1996 and early 1997, 1.3 millionrefugees were repatriated to Rwanda fromneighboring countries and received foodaid. Without massive infusions of pre-dominately U.S. emergency assistance,more Haitians would have fled Haiti seek-ing refuge in the United States. Massivestarvation and human suffering wouldhave occurred in Mozambique andRwanda. Emergency assistance clearlyhelped save lives and alleviate suffering.However, except in Haiti, data collectionand monitoring were not done (or weredone poorly), so it is difficult to quantifyresults.

2. Relief distribution. Effective distri-bution of emergency assistance requires orga-

S

6


nization and control to limit theft, minimizeabuse, guard against political manipulation,and protect beneficiaries.

Distributing relief supplies was aproblem to some extent in all three coun-tries. The large quantity of food aid, inparticular, became a source of violent com-petition—not only for its value as food forconsumption but also as a source of po-litical power for those controlling access.In Haiti, fighting among beneficiariessometimes erupted when food was distrib-uted. Distribution points used to stockpilefood supplies were looted and supplieswere hijacked. Local authorities some-times used food to favor certain politicalfactions or for their personal aggrandize-ment. In Mozambique as well there werereports of corruption, theft, and politicalor personal favoritism in food aid distri-bution. Target populations did not alwaysreceive timely and sufficient food aid. InRwanda the military and former govern-ment leaders controlled much of the reliefdistribution. Thus they were able to divertfood from the intended beneficiaries fortheir own purposes.

NGOs were mainly in charge of reliefdistribution in Haiti. They addressed theseproblems by stocking and distributingfood aid in neutral settings (schools, fac-tory yards), using ration cards to track thereceipt of food aid, and having NGO per-sonnel and occasionally police present tomonitor distribution. These measures lim-ited diversion to less than 10 percent andhelped reduce violence. In Mozambiquethe government emergency relief agency

lacked the technical expertise to plan, or-ganize, and manage the distribution ofmassive supplies of relief aid. Leakage wastypically 30 percent, and at one point 50percent was lost, stolen, or diverted. Inresponse, donors, NGOs, and the privatesector took over much of the distribution,and losses dropped to under 5 percent. Incamps in Tanzania and Zaire, more foodaid was supplied than was necessary, andmore than usual was misappropriated.Some NGOs suspended their operationsbecause they knew they were assistingpeople guilty of crimes against humanity.

3. Political and social unrest. Emer-gency assistance can help maintain social calmand mitigate political instability. Conversely,it can exacerbate political tensions. Rarely is itpolitically neutral.

The international community pro-vided massive quantities of emergencyassistance to Haiti, Mozambique, andRwanda. The political effects of the assis-tance varied. In Haiti, food aid reducedthe probability of food riots during a pe-riod of political and economic stress andmay have had a dampening effect on po-litical tensions; but it also may have re-sulted in a political status quo that enabledthe de facto military regime to stay inpower longer. In Mozambique, externalmilitary assistance provided by the SovietUnion and by South Africa fueled the war.Food aid, by comparison, had relativelylittle effect on the country’s political dy-namics, although food diverted to soldiersmay have contributed to the war effort. InRwanda, where genocidal killers were

Conclusions, Lessons Learned, and Recommendations 45

mixed with legitimate refugees in camps,humanitarian assistance served to prolongthe emergency.

4. Demobilization . Demobilization,disarmament, and reintegration of armed forcesis vital in ending a complex emergency andbeginning a period of recovery.

Demobilization of Haiti’s armedforces removed one source of violence inthe country. However, many of the demo-bilized soldiers retained their arms, andbecause most were unemployed owing tothe weak economy, they are believed tohave caused at least part of the post-1994rise in theft and street violence. InMozambique, demobilization of Renamoand Frelimo armed forces and their rein-tegration into civilian life was essential forthe transition from relief to recovery. Asin Haiti, though, many weapons were notturned in and that contributed to a rise incrime. In Rwanda, soldiers and militialoyal to the former government remainarmed. They are still trying to destabilizethe present government of Rwanda by us-ing a campaign of propaganda and terrorto destroy the political and social struc-tures of the country, beginning in thenorthwest.

5. Relief to development. Emergencyassistance that enables people to protect theirlivelihoods (as well as meet immediate needs)helps reduce dependency and contributes tolong-term economic development.

In Haiti many urban factory workerslost their jobs, and some farmers were

obliged to sell their agricultural and house-hold assets to survive. Numerous farmersin Mozambique and Rwanda also lost theirproductive resource base when they fledtheir villages. These people became depen-dent on emergency relief. USAID and theNGOs responded—not only with food as-sistance but also with agricultural inputs(seeds and tools) and household goods.That assistance encouraged refugees andinternally displaced persons to return totheir villages. It enabled them to resumefood production and decreased their de-pendence on food aid. NGOs in all threecountries also implemented food-for-workprograms that created short-term jobs andhelped rehabilitate productive infrastruc-ture (roads, irrigation) needed for eco-nomic development. Often, though, theinfrastructure was not maintained.

6. Donor coordination. A clearly des-ignated, agreed-upon central authority canmake the delivery of humanitarian assistancemore effective.

In Haiti the United Nations officiallydesignated the Pan American Health Or-ganization as the coordination point foroverall health planning and services dur-ing the U.S.–led embargo. That enablednumerous NGOs to deliver medical sup-plies and food to vulnerable populationsmore effectively. In Mozambique, by con-trast, donor efforts at times overlapped orworked at cross-purposes. One donor wasgiving free seed while another was sell-ing it; one donor was shifting to develop-ment assistance while another was stillproviding grant relief. That confused ben-


eficiaries and undermined efforts to re-duce dependency. Similarly, lack of donorcoordination was a serious problem inRwanda.

Recommendations

Four key recommendations emergefrom these six conclusions and lessonslearned. They are mainly managementoriented.

1. Monitoring and evaluation. Estab-lish a central monitoring and data-collectionunit to serve all donors during the early weeksof a complex emergency.

Baseline data for socioeconomic indi-cators (e.g., malnutrition rates, food prices,population displacement) can help man-agers identify appropriate kinds of emer-gency relief, target its distribution, andsubsequently measure and evaluate its ef-fectiveness. Close monitoring enhancesdonor coordination and is essential forassessing aid needs, avoiding work atcross-purposes, identifying recipientgroups no longer needing emergency aid,shifting from relief to reconstruction anddevelopment, and designing and adjust-ing economic policies.

2. Adverse political consequences. Bealert to potential undesirable political or socialeffects that relief aid may cause.

Emergency food distribution, in par-ticular, can have unintended and undesir-

able political consequences. Control overfinal distribution often has reinforced thepower of local authorities or political fac-tions. It has strengthened their relativeposition during or after the conflict andfacilitated their self-aggrandizing, oftenexploitive, behavior toward the intendednoncombatant beneficiaries. Decisions tocontinue, withdraw, or modify aid distri-bution should be made as a matter of de-liberate policy on a regular basis by eachindividual donor.

3. Reducing dependency. Give refu-gees incentives to return home and impose dis-incentives on those remaining outside theircountry of origin.

Generally, the longer encampment ortemporary foreign residence lasts, the lesswilling refugees are to return home. Acombination of “push” factors (such as ter-minating free food distribution) and“pull” factors (such as including seeds andtools in resettlement packages) is likely toaccelerate the repatriation process. But forreasons of political and bureaucratic self-interest, local governments may not re-move from the rolls those no longer need-ing relief. Therefore, donors must moni-tor each situation closely, recognizing thatboth relief and development assistancemay be needed if some areas remain inemergency status while others stabilizemore quickly. After populations have beenrepatriated and are settled, the agriculturalbase begins to be reestablished, depen-dency on free food distribution drops, andlong-run food security is enhanced.

Conclusions, Lessons Learned, and Recommendations 47

4. Capacity building. Train technocratsto manage the postconflict economic transition,and train others in skills for which there isemployment demand.

Complex emergencies seriouslyweaken the capacity of governments toprovide basic public services. Economicrecovery requires a cadre of high-leveltechnocrats with both management andconceptual skills, especially in macroeco-nomic and sectoral policy formulation.

Such skills are likely to be in short supply,especially if preconflict professionals andthe intelligentsia were targeted for delib-erate elimination or have permanently leftthe country. Recovery also needs to bedovetailed with postconflict economic re-alities. Job training is fruitless if unemploy-ment in the depressed economy remainshigh. Training is especially critical for de-mobilized soldiers who, because they of-ten remain unemployed, tend to turn todestabilizing criminal activity.

Annex A

HumanitarianEmergencies andDonor Assistance

HE TABLE BELOW indicates the num-ber of people affected by humanitar-

ian emergencies in recent years. The fourfigures in this annex show changing

levels of official development assistanceand humanitarian assistance provided bythe donor community and the UnitedStates over the past 30 years.

Source: U.S. Mission to the United Nations, April 1997 and September 1998. Figures are based ondata provided by the U.S. Committee for Refugees.

Note: The table includes only major emergencies, those in which at least 300,000 people requiredinternational humanitarian assistance to avoid severe malnutrition or death.

January 19973.52.50.783.11.0

0.480.72.50.3

>0.52.12.0

>5.00.350.631.51.00.854.40.63

33.82

April 19984.12.50.771.50.751.00.450.300.850.30.51.71.1

>7.4

0.3>1.0

1.00.74.40.90.4

31.92

Table A1. Ongoing Humanitarian Emergencies and Number ofPeople Affected (Millions), 1996, 1997, 1998

AfghanistanAngolaAzerbaijanBosnia and HerzegovinaBurundiColombiaCroatiaEritreaEthiopiaGeorgiaHaitiIraqLiberiaNorth KoreaRussia (Chechnya)RwandaSierra LeoneSomaliaSri LankaSudanTajikistanUganda

Total

January 19964.02.50.953.70.8

0.51.03.51.01.12.651.5

0.31.01.81.00.854.01.0

33.15

T

Complex Humanitarian Emergencies and USAID’s Humanitarian ResponseA2

Figure A1Official Development Assistance and Emergency/Distress Relief, All Donors, 1969–98

(billions $US 1998)

Source: OECD/DAC

Figure A2Emergency and Distress Relief as a Percent of Official Development Assistance,

All Donors, 1969–98

Source: OECD/DAC

0

10

20

30

40

50

60

70

80

1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997

OD

A

0

1

2

3

4

5

E/D

R

ODA

E/DR

0

1

2

3

4

5

6

7

8

1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997

perc

ent

Annex A: Humanitarian Emergencies and Donor Assistance A3

Figure A3U.S. Official Development Assistance and Emergency/Distress Relief, 1971–98

(billions $US 1998)

Source: OECD/DAC

Figure A4Trends in U.S. Official Development Assistance and Emergency/Distress Relief,

1971–98

Source: OECD/DAC

0

2

4

6

8

10

12

14

16

1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997

OD

A

0.0

0.5

1.0

1.5

2.0

E/D

R

U.S. ODA

U.S. E/DR

0

10

20

30

40

50

60

1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997

perc

ent

U.S. E/DR % of all E/DR

U.S. ODA % of all ODA

U.S. E/DR % of U.S. ODA

Evaluation ObjectivesAnd Constraints andImplications forDonor Coordination

Evaluation Objectives

HE OVERALL OBJECTIVE of the CDIEevaluation was to assess the effective-ness of U.S. emergency assistance in

response to complex humanitarian emer-gencies.* It covered all types of relief as-sistance including (1) food, (2) water andsanitation supplies, (3) medical servicesand health care, and (4) clothing, shelter,and resettlement assistance. The evalua-tion in each country examined the politi-cal and historical events that caused theemergency as well as the effectiveness ofthe U.S. humanitarian response. Conclu-sions and lessons learned are cast in termsof management recommendations to helpguide USAID’s future policy, program, andbudget decisions in humanitarian assis-tance.

Evaluation Methodology

The evaluation looked at three maintopics framed in the form of these ques-tions:

1. Did U.S. emergency assistance savelives and alleviate suffering during thecomplex emergency?

2. Did U.S. emergency assistance af-fect social and political hostilities (or ten-sions) associated with the complex emer-gency?

3. Did U.S. emergency assistance con-tribute to economic development?

The first question derives directlyfrom legislation that gives USAID primaryresponsibility within the U.S. governmentfor responding to overseas disasters. Ques-tions 2 and 3 are not directly related to theoverarching mandate to “save lives andalleviate suffering.” The second questionconcerns the potential political impact ofemergency assistance and whether suchassistance has had the unintended effectof prolonging complex emergencies. Thethird question addresses the potential eco-nomic impact of the assistance and the so-called relief-to-development continuum.

Background work for the evaluationwas carried out in Washington, and field-work was conducted in Haiti, Mo-

Annex B

T

*The “Emergency Assistance” concept paper(11 December 1997) provides the underlyingrationale for the assessment.

Complex Humanitarian Emergencies and USAID’s Humanitarian Response

zambique, and Rwanda. During the firstphase, desk studies for each of the threecountries were completed, a workshopwas convened to review the desk studies,and two topical guides were developed tohelp structure key informant interviewsand field observations. One topical guidewas for implementers and experts and theother was for beneficiaries.

The second phase involved fieldworkin each of the three countries. Evaluationteams sought answers to the three ques-tions above through careful analysis ofsecondary sources not available in theUnited States (reports, evaluations, agri-cultural production surveys, health statis-tics), interviews with key informants andfocus groups, and site visits. The resultsof the three case studies were synthesizedduring the third phase.

During the second phase, evaluatorstalked with a broad range of national andexpatriate experts (field technicians, pro-gram directors, administrators) who hadmanaged or implemented emergency as-sistance programs. These people typicallyincluded donor agency staff, NGO partners,host government officials (both nationaland provincial), UN agencies, and someacademics and journalists. Beneficiary-level information was collected in bothurban and rural areas. Site visits producedvaluable insights and helped corroborateinformation from other sources and toground-truth the teams’ interpretations.The process of interviewing and listeningproduced answers and also generated sec-ondary questions.

This mainly qualitative methodologi-cal approach does not produce statisticallyvalid proof of impact. Rather, it allows aninterpretation of the links between a USAIDintervention and various effects that plau-sibly can be associated with that interven-tion. Given the characteristics of humani-tarian assistance, a more scientific ap-proach is rarely feasible (Hallam 1998, 28).One can judge the validity of the interpre-tations on the basis of several criteria: thelogic and consistency of the argumentssubstantiating them, the strength and qual-ity of the evidence, triangulation, and thereputations of those involved. This issometimes called the common-senseschool of evaluation: impact is deducedfrom a combination of information fromkey informants and from the evaluators’own sense of how the world works.

Evaluation Constraints

Donor agencies that provide relief inthe context of conflict or civil war mustconsider factors they normally would notneed to. These factors include sovereignty,international law, the appropriate balanceof aid between opposing sides, and per-haps national foreign policy interests. Thepolitical and legal questions associatedwith humanitarian assistance in conflictareas make the evaluation of relief pro-grams far more sensitive than that of de-velopment programs (Borton 1994, 11–12).

Moreover, difficulties typically asso-ciated with evaluating development assis-tance programs seem to be magnified

B2

Annex B: Evaluation Objectives and Constraints and Implications for Donor Coordination

when evaluating relief programs. Theseinclude (1) lack of adequate baseline data;(2) difficulty in identifying control groups;(3) the dynamic context of relief programs,which makes it difficult to isolate the ef-fect of the relief intervention; (4) the lim-ited utility of cost–benefit analysis; (5) thelarge number of agencies involved; and (6)the high political and media profile of re-lief programs (Borton 1994, 12).

There is yet another difficulty withevaluating relief programs: the prevalentattitude among many relief agencies thatassessments of their programs are unnec-essary. To paraphrase their view: our mo-tives were well intentioned, we did ourbest under difficult circumstances, whyshould we now subject ourselves to a criti-cal examination? (Borton 1994, 1). Reliefagencies also may believe that “all disas-ters are different, so what is the point intrying to learn the lessons of our responseto this particular disaster?” (Borton 1994).

According to Larry Minear (1998),four characteristics of humanitarian orga-nizations make them resistant to change:

1. As just noted, humanitarian orga-nizations tend to approach every crisis asunique. “As long as every crisis is per-ceived as wholly without precedent orparallel, there will be little scope for insti-tutional learning (10).” Correcting this ten-dency will require greater institutionalmemory, greater attention to comparativeanalysis, and more support for in-houseevaluation.

2. Humanitarian organizations areaction oriented. As a result, evaluationsgenerally are not read and lessons are notlearned (11).* Although the action orien-tation of NGOs is a positive attribute, itshould be informed by past experience aswell as current political, military, and so-cial realities. Unfortunately, NGOs some-times ignore this experience rather thanuse it as a basis for more strategic inter-vention.

3. Humanitarian organizations areoften defensive about criticism, even con-structive criticism. As such, they are un-likely to use it to improve their effective-ness (12). Nevertheless, constructive criti-cism can be valuable, which highlights theneed for independent research and evalu-ation.

4. Humanitarian organizations oftenlack accountability. They are not held re-sponsible for their actions: everybody—and thus nobody—is responsible. This lackof accountability will be difficult to rem-

*The multidonor Joint Evaluation of Emergency As-sistance to Rwanda demonstrates the importance ofevaluations but also demonstrates their limitedability, in and of themselves, to produce institu-tional change (6). This comprehensive evaluationreviewed aid programs that cost $1.4 billion fromApril through December 1994. It was done by 37institutions (governmental, intergovernmental,and nongovernmental), enlisted 52 consultants,produced a five-volume report, and cost $2 mil-lion. The broadest of its 64 recommendations wereignored; those stressing “coordination by com-mand” were rebuffed; and only the least radicaloptions were acted on to some degree.

B3


edy because of the accepted approach ofcoordination by consensus (or default)rather than coordination by command (12–14).

Implications forDonor Coordination

This raises the issue of donor coordi-nation. Coordination is a concept ap-proved by all but defined by few(Prendergast and Scott 1996). A majoremergency assistance operation can in-volve numerous bilateral and multilateraldonors, hundreds of NGOs, a range of UNagencies, military contingents, and na-tional governments. An effective divisionof labor among these and other actors isneeded to maximize the comparative ad-vantage and impact of each.

Donor coordination is often perceivedas a role for the United Nations. That bodyexpanded its peacekeeping operations atthe end of the Cold War when complexemergencies proliferated. But peacekeep-ing operations required endorsement bythe Security Council. This meant the Per-manent Five could control UN peacekeep-ing and enforcement operations giventheir veto power and control over finances.In the case of Rwanda, the major powerson the Security Council (except France)made clear they were not interested in asmall African country that was marginalto their economic or political concerns andperipheral to international strategic rival-ries. This experience suggests that the

United Nations is not always bestequipped to coordinate an internationalresponse to a complex emergency.

The only country with a demon-strated ability to energize the United Na-tions and the Security Council in a crisis isthe United States. But in the case ofRwanda, even the United States, hauntedby the memories of Somalia, was deter-mined not to get involved in another Afri-can conflict. Not crossing the “Mogadishuline” became the guiding principle. Wash-ington was also preoccupied with criseselsewhere, especially in Bosnia and Haiti,and the potential financial burden ofRwanda was a major concern. The UnitedStates at the time was assessed 31 percentof the costs of all UN peacekeeping op-erations. (Joint Evaluation 1996, study 2,11). In short, even the United States is notalways well positioned to coordinate anappropriate humanitarian response to allcomplex emergencies.

Lack of donor coordination in provid-ing humanitarian assistance is well docu-mented. This is true within a single orga-nization as well as among several organi-zations. For example, within the UN sys-tem, UNICEF has responsibility for womenand children; but UNHCR has responsibil-ity for refugees, including women andchildren. Similarly, WFP has responsibil-ity for food assistance; but UNICEF andWHO may be involved in actual programoperations. Lack of coordination can re-sult in inefficiency. For example, in mostsituations each agency arranges local

B4

Annex B: Evaluation Objectives and Constraints and Implications for Donor Coordination

transport for its own supplies, but this pro-cess only bids up prices. Another issuehaving cost-related implications is that hu-manitarian organizations are heavilyweighted toward mop-up operations; theyare reactive, not proactive. Too little atten-tion is given to investments in conflict pre-vention (Minear 1994b, 4, 6); see annex C.

The creation of the UN Departmentof Humanitarian Affairs (and subse-quently, the Office for the Coordination ofHumanitarian Affairs) is generally seen asa positive step. Nevertheless, the UN isoften criticized. For example, each of thesix evaluations reviewed by Apthorpe callsfor some type of reform in the United Na-tions. James Ingram (former head of WFP)is more draconian: he believes there is noreason that a coordinated international re-

sponse to future complex emergenciesshould necessarily be built around theUN—and suggests there are a variety ofreasons why it should not. He recom-mends the International Committee of theRed Cross or a new organization outsidethe UN system (Apthorpe 1997, 96, 98).

In disaster after disaster, local insti-tutions and people provide the first lineof response. By contrast, the world’s hu-manitarian system relies heavily on out-side resources that marginalize local re-sources and expertise. The idea seems tobe fixed that in complex emergencies thereis simply no alternative to using donorresources delivered by NGOs. This may betrue. If so, donor resources should be pro-vided to support existing livelihoods orcoping strategies as much as possible.

B5

Are ComplexEmergenciesPredictable?

N REFERENCE to the Rwanda crisis,James Kunder notes that “an ounce of

prevention is worth 25,000 tons of foodaid” (Minear 1994, 9). The point is welltaken. It is far better to prevent complexemergencies from occurring in the firstplace than it is to respond to victims’ needsafterwards. What factors seem to be re-sponsible for igniting complex emergen-cies? Are complex emergencies predict-able? Regions that are particularly proneto seasonal natural disasters (for example,cyclones in India, hurricanes in Florida,drought in the Sahel) have developedearly-warning systems designed to predictthe next natural disaster and to mitigateits effects. Such is not the case with com-plex emergencies. However, studies havebeen undertaken to help explain the causesof civil wars and complex emergencies,and that may be the first step in predict-ing their occurrence. This annex summa-rizes the results of several of these stud-ies.*

The ethnic model is the prevailing ap-proach to explaining complex humanitar-

ian emergencies. It postulates thatethnicity (based on differences of lan-guage, race, tribe, religion, national origin,or some other cultural sense of identity) isthe primary factor underlying a complexhumanitarian emergency. By contrast, theeconomic model views complex humani-tarian emergencies and ethnic conflict inthe context of economic development andstructural change. In this model, economicfactors are pivotal in shaping conflicts,though these conflicts may be triggered bypolitical or ethnic causes. That is, economicfactors create the conditions for ethnic orpolitical explosions that in turn lead tocomplex emergencies (Nafziger 1996,3–4).

Paul Collier and Anke Hoeffler (1998)are among those who explain the phenom-enon in economic terms using economicanalysis. They identify four variables thatthey hypothesize are associated with civilwars: per capita income, size of the natu-ral resource base, population size, andethnolinguistic fractionalization. Using asample of 98 countries (of which 27 hadcivil wars during 1960–92), they analyzethe relative importance of these four vari-ables. Their results show that all four vari-

*A substantial body of literature has developed onthis subject. This annex only scratches the surface.

I

Annex C


ables are significant determinants of boththe probability of civil wars occurring andtheir duration. Specifically:

1. Per capita income. The study showsthat civil war is overwhelmingly a phe-nomenon of low-income countries. Otherthings being equal, the probability of civilwar is substantially greater in countrieswith a very low per capita income than incountries with a relatively high per capitaincome. Moreover, the predicted durationof civil war is much shorter in countrieswith a higher per capita income. This isbecause a high-income population hasmore to lose during a conflict, and the costsof rebellion increase with its duration.Conversely, the opportunity cost of beinga rebel and prolonging a conflict is low fora low-income population.

2. Natural resource base. The effect ofnatural resource endowments is not asstraightforward. Initially, increased natu-ral resources increase the risk and dura-tion of civil war. This is because the tax-able base of the economy constitutes an at-traction for rebels wishing to capture thestate. But at a high level, natural resourcesstart to reduce the risk of civil war. This isdue to the government’s greater financialcapacity to defend itself through militaryexpenditures.

3. Population. Countries with largerpopulations have a higher risk of civil warand wars that are likely to last longer. Thisreflects the greater likelihood that coun-tries with larger populations will have alarger number of diverse groups, which

in turn suggests there would be a greaterlikelihood of various ethnic groups want-ing to secede owing to cultural and lin-guistic disparities. (Note, however, that theeffect of population size is ambiguous be-cause potentially it could be inconsistentwith point 4, next.)

4. Ethnolinguistic fractionalization. Thischaracteristic is an index that ranges from0 (complete homogeneity) to 100 (maxi-mum fractionalization). The index wouldbe 100 when each individual in a countrywas in a different ethnolinguistic group.Conversely, the index would be 0 in a so-ciety with a single ethnolinguistic group.Contrary to conventional wisdom (and topoint 3, above), the authors find that morefractionalized societies are no more proneto civil war than highly homogeneousones. Rather, the danger of civil war ariseswhen a society is polarized into twogroups. Polarized societies have a muchhigher probability of civil war than eitherhomogeneous or highly fractionalizedones. Thus, a country with two similar-sizeethnolinguistic groups could reduce therisk of civil war either by partition or byunion with other countries.

Not surprisingly, Collier and Hoefflerconclude that the “ideal society” (one en-dowed with the most favorable of each ofthese characteristics) has less risk of civilwar than the “catastrophic society” at theother end of the spectrum (one with theleast favorable of each characteristic). Theyalso conclude that poverty is the maincause of civil war. Table C1 lists the sampleof 98 countries in their analysis. Countries

C2

Annex C: Are Complex Emergencies Predictable?

AlgeriaBurundiChadDominican RepublicEl SalvadorEthiopiaGuatemalaIndiaIndonesiaIraqLiberiaMauritaniaMoroccoMozambiqueMyanmarNicaraguaNigeriaPakistanPeruPhilippinesSomaliaSri LankaSudanTurkeyUgandaZaireZimbabwe

ArgentinaAustraliaAustriaBarbadosBeninBoliviaBrazilBurkina FasoCameroonCanadaCent. African Rep.ChileCongoCosta RicaDenmarkEcuadorEgyptFinlandFranceGabonGambiaGermanyGhanaGreeceGuyanaHaitiHonduras

Hong KongIcelandIrelandIsraelItalyIvory CoastJamaicaJapanKenyaKoreaMadagascarMalawiMalaysiaMaliMaltaMauritiusMexicoNepalNetherlandsNew ZealandNigerNorwayPanamaPapua New GuineaParaguaySaudi ArabiaSenegal

Sierra LeoneSingaporeSouth AfricaSpainSwedenSwitzerlandSyriaTanzaniaThailandTogoTrinidad and TobagoTunisiaUnited KingdomUnited StatesUruguayVenezuelaZambia

With a Civil War Without a Civil War

Table C1. Countries With and Without a Civil War, 1960–92

Source: Collier and Hoeffler 1998, 573.

C3

in column 1 experienced civil war during1960–92; the rest did not.

Nafziger also identifies economic fac-tors as primarily responsible for complexemergencies (Nafziger 1996, v). He speci-fies four, some of which parallel thoseidentified by Collier and Hoeffler:

1. Prolonged stagnation. The majorityof countries with humanitarian emergen-

cies have experienced several years (oreven decades) of negative or stagnant eco-nomic growth. Below a given threshold, aprotracted decline in incomes is likely totrigger increasingly fierce competition forscarce resources, jobs, and opportunities(5).

2. Unequal growth. The situation islikely to deteriorate more rapidly if incomeand asset distribution worsen. Skewed

Complex Humanitarian Emergencies and USAID’s Humanitarian ResponseC4

economic growth increases the relativedeprivation of substantial sections of thepopulation, even if it does not cause abso-lute deprivation (6).

3. Population pressure on resources .Rapid population growth coupled with en-vironmental degradation and resourcedepletion can contribute to diminishingreturns to agricultural land. Declining ag-ricultural returns, often exacerbated bymaldistribution of land and water, are asource of conflict (8).

4. Distributional shifts owing to adjust-ment programs. Large and abrupt shifts inthe distribution of income and wealth dur-ing stabilization and liberalization pro-grams can affect the distribution of powerwithin a country (9).

According to Nafziger, the way inwhich elites react to these four factors in-fluences the probability of political con-flict and humanitarian disasters occurring(Nafziger 1996, 10).

De Soysa and Gleditsch of the Inter-national Peace Research Institute, Oslo,analyzed 103 armed conflicts that occurredduring 1989–97, since the end of the ColdWar. They found that political instabilitythat led to violent conflict has sprungmainly from economic concerns, ratherthan political or ideological differences.They also found that most contemporaryarmed conflicts have occurred in impov-erished countries where agriculture wasthe mainstay of the economy. Armed con-

flicts were often fought over issues relatedto agriculture, such as land ownership,environmental change, water scarcity, andfood shortages. However, divisions overthese issues often fall along ethnic lines,obscuring the fundamental causes.

The researchers conclude that condi-tions in poor countries that undermine therural economy can generate political griev-ances that result in endemic armed con-flict. Agriculture is the dominant economicsector in most poor countries. Poor coun-tries that invest in their agricultural sec-tors provide livelihoods to people andthereby lower the incidence of conflict.People do not need to turn to violent move-ments as a means of survival. These find-ings, like those of Collier and Hoeffler andNafziger, stress the importance of eco-nomic factors in explaining the outbreakof complex emergencies—and in particu-lar, the importance of agriculture.

Finally, Mary Anderson also identi-fies economic considerations among thekey underlying factors that contribute tocivil war. She suggests that some engagein civil war because they have little to lose;others, because they have something togain. Those with few economic alterna-tives (the poor) are often the rebels whohave little to lose. They are often supportedby those who have something to gain(arms merchants and other profiteers).Others who stand to gain include thosewhose employment depends on continu-ation of the war and the funding it gener-ates.

Annex C: Are Complex Emergencies Predictable? C5

Although many leaders claim to beengaged in a struggle against past injus-tices, the evidence available suggests thatjustice and fairness are neither the funda-mental motives nor the likely outcomes oftheir wars. Rather, these leaders seekpower. However, once a civil war starts,the war itself creates a spiral of atrocitiesand reprisals and hatred that become thelegitimate root cause for its continuation(Anderson 1996, 10–12). As reported inone anthropological study, vengeance is

the major reason for civil war in 75 per-cent of the cases.

In sum, complex emergencies clearlyare not predictable. Moreover, the incon-sistencies among the findings of the stud-ies just noted illustrate that there is no gen-erally accepted theory of conflict. It seemsequally clear, though, that economic fac-tors are of major importance in generat-ing these crises, and they may well havean important role in preventing them.

Implications forThe Kosovo Crisis

N MAY 1999 the Kosovo crisis embracedmuch of the Balkans in one way or an-

other. Various proposals were beingmooted for international and regional ac-tion to secure a stable peace when the con-flict ended. The length of the conflict, theextent of casualties and population dis-placement, the amount of physical de-struction, the postconflict sovereignty con-figuration, and the lingering political andsocial effects on the states most affectedwere all unpredictable at that time.

This annex identifies lessons from thisand other evaluations of complex humani-tarian emergencies that might be useful incoping with the Kosovo crisis. Applicabil-ity of any individual lesson depends, ofcourse, on how the crisis develops, how itis resolved, and what postconflict politi-cal architecture emerges. The lessons andobservations have been culled from docu-ments prepared by the following: CDIE, the

Department for International Develop-ment of the UK, the Directorate Generalof International Cooperation of Nether-lands, the Danish International Develop-ment Agency, the Organization for Eco-nomic Cooperation and Development/Development Assistance Committee, theUN High Commissioner for Refugees/World Food Program, the InternationalBank for Reconstruction and Develop-ment/Operations Evaluation Department,the International Crisis Group, and thisauthor.

Crisis Management

1. Coordination. The Kosovo crisis de-mands an effective coordination mechanism.

Virtually all evaluations dealing withcomplex humanitarian emergencies citeinadequate donor coordination as causingserious problems—inefficiencies, waste,lost leverage. Complex emergencies typi-cally face unusually complex coordinationproblems: large numbers of multi- andbilateral donors and nongovernmentalorganizations working in a context of con-

Robert J. Muscat prepared the original version ofthis annex (25 May 1999). In anticipation of theend of the Kosovo crisis, Muscat synthesized les-sons from this and other evaluations of complexhumanitarian emergencies for possible applicationin Kosovo.

I

Annex D


tested or collapsed governance, sometimeswith external military organizations in-volved (Bosnia, Cambodia, Kosovo, Soma-lia). Coordination has ranged from volun-tary, committeelike structures to more hi-erarchical, or command, systems involv-ing lead agencies or even a UN or othersanctioned authority with powers of di-rection (Bosnia, Cambodia).

Postconflict Kosovo is likely to havean interim international administrationunder which a centrally guided coordina-tion system would be appropriate. Of allrecent complex humanitarian emergen-cies, the Bosnian and Cambodian experi-ences have the most relevant lessons in thisregard, especially concerning the security–political–economic interfaces and theexercise of authorized powers. (Extensiveindependent and internal agency evalua-tions of the international administrationexperience of these two cases are avail-able.)

2. Monitoring. A central monitoringand data collection unit should be set up toserve all donors.

Complex emergency information-sharing and monitoring systems have alsobeen evaluated negatively. A central unitto collect and analyze socioeconomic dataand program information, one that servesall donors (again, as in Bosnia or Cambo-dia), is needed to ensure donor coordina-tion. Close monitoring is essential for as-sessing needs, avoiding work at cross-pur-poses, identifying recipient groups nolonger needing emergency aid,

transitioning from relief to reconstruction,and designing and adjusting economicpolicies. The Rwanda experience demon-strated that good information on the con-cerns and expectations of encamped refu-gees, and on the power and leadershipstructures in the camps, is critical for main-taining orderly relations and confidenceamong refugees, assistance authorities,and host governments.

3. Food distribution. Local food distri-bution organizations commonly need closemonitoring to avoid factional diversion orpoliticization.

Judging by the internationalcommunity’s generally successful hu-manitarian response record, malnutritionor disease is unlikely to become a signifi-cant complication of the current crisis, withtwo important caveats. First, given inter-nal displacement, exposure to ethniccleansing, and the bombing of supply in-frastructure, those remaining inaccessibleinside Kosovo could suffer severe priva-tion, depending on how long the conflictlasts. Second, Kosovo, Albania, andMacedonia each have deeply divided po-litical factions and parties. In similar cases(Ethiopia, Sudan), factions that controlledfood aid distribution have withheld foodfrom entitled beneficiaries loyal to rivalfactions. Thus far, the political dynamicsin Albania and Macedonia have reportedlyfacilitated rather than hindered aid distri-bution to refugees. Given the fragile poli-tics in both countries and the stresses gen-erated by the crisis, the integrity of emer-gency aid distribution should be closely

D2

Annex D: Implications for the Kosovo Crisis

monitored. If postconflict arrangementsretain Kosovo’s ethnic Serb minority, theproblem could arise there, unless distri-bution is well controlled by external agen-cies.

4. Refugee repatriation. Refugees mayrequire incentives to return home.

Generally, the longer encampment ortemporary foreign residence lasts, the lesswilling refugees are to return home(Mozambique). A combination of incen-tives for returning to Kosovo and disin-centives for remaining outside may be re-quired if NATO’s expressed optimismabout early return does not materialize.

5. World Bank coordination of recon-struction. Several mechanisms can be used tostrengthen World Bank performance.

The international community has re-cently relied on the Bank to take the leadin reconstruction, monitoring, and othernonpolitical and nonmilitary functions. Onthe basis of the Bank’s own evaluation, itseffectiveness in complex humanitarianemergencies has varied, depending to aconsiderable extent on senior manage-ment’s level of interest and commitmentas reflected by (a) authorities and size ofstaff of the in-country resident represen-tative’s office, (b) speed of headquarters’processing and willingness to cut bureau-cratic corners, and (c) size of administra-tive budget.

The Bank’s performance in some caseshas been enhanced by (a) receipt of (grant)

trust funds from donors interested in in-ducing and supporting a larger Bank ef-fort than might otherwise be the case, and(b) strong advocacy, by one or more boardmembers, for maximum Bank perfor-mance.

Relief to Development:Transition and Links

1. Funding discontinuities. USAIDshould avoid interruptions in funding andoperations during the transition phase.

Relief funding affords considerableflexibility compared with the proceduralcomplexity and slower pace of long-termdevelopment operations. Thus, USAID andNGOs have experienced funding gaps andprogram interruptions between the phase-out of relief assistance and the phasein ofdevelopment assistance (Mozambique).The two funding systems may need tooperate simultaneously in countries wheresome areas remain in emergency statuswhile other areas have stabilized. The ex-istence of land mines in Kosovo could ne-cessitate operating in both modes at once.Projects undertaken as rapid transitionresponses should not ignore longer runreconstruction and development objec-tives. This may require close coordinationamong USAID’s units responsible for dif-ferent funding sources and operations.

2. Resettlement planning. Realisticplanning for resettlement in Kosovo should al-ready be under way.

D3


The needs of returning refugees dur-ing the initial stages of resettlement mayseem obvious. In practice, though, resettle-ment planning has often been deficient,based on unrealistic assumptions, inad-equate information, and poor analysis ofthe conditions prevailing in repatriationdestinations (Cambodia, Haiti, Mo-zambique). If repatriation begins as earlyas autumn 1999 (as NATO spokesmen haveasserted), then planning, preparation, andfinancial provision should already be un-der way, including mobilization for com-munity and home demining.

Since repatriation will occur at theonset of winter, the repatriation packagewill need to include housing reconstruc-tion (fraught with problems of finance andimplementation), food aid until the nextharvest, agricultural assets and inputs fornext season’s cultivation, and householditems including fuel, livestock restocking,and perhaps cash (if press accounts arecorrect that departing Kosovars have beenstripped of their money and valuables). Ifearly return proves infeasible, contingencyplanning for winterizing the camps shouldalso be undertaken immediately.

3. Agenda for negotiating settle-ment. Economic dimensions should constitutea key element of future negotiations to settlethe conflict.

Conflict-settlement negotiations andarrangements should address not onlypolitical and security dimensions but alsoeconomic implications, with due attentiongiven to the policy framework and man-

agement requirements. Otherwise, reha-bilitation may be impeded. Developmentwas delayed and opportunities missed inCambodia because the interim UN admin-istration interpreted its mandate—againstthe advice of its economics unit—as limit-ing its authority in economic development;the intent was to leave such decisions tothe successor national authorities yet to beestablished. The economic recovery of theresettled Kosovar population should notbe delayed by making the interim inter-national administration’s economic scopetoo restrictive.

4. Economywide distortions. Moni-tor the effects of external interventions to de-tect inflationary and other economywide dis-tortions.

Local expenditures made by largenumbers of international military, civilian,and NGO personnel can have major posi-tive and negative effects on fragile crisisand postconflict economies. Such expen-ditures can help stimulate the recovery oflocal production and service sectors, butthey can also have inflationary and otheradverse effects. For example, they can af-fect local housing markets, wage levels ofscarce local professional and technical per-sonnel, and wages and perks (includingsupplements) of civil servants seconded asaid-project staff. Well-positioned officialsand other elite may capture much of thisexpenditure, producing new, large incomedisparities and consequent resentment.The obverse may also be a problem: eco-nomic downturn and employment losswhen stabilization allows major reduc-

D4


tions in external military and other per-sonnel. Monitoring such effects should beinitiated right from the start.

5. NGO experience. Benefit from the ex-perience of nongovernmental organizations.

NGOs play a major role in complex hu-manitarian emergencies, and lessons fromtheir experience should be identified, es-pecially in Bosnia and Croatia. Amongthose lessons:

a. NGOs emerging in response to theunprecedented availability of funds forcivil society often have only shallow ex-perience; they should be encouraged withsmall grants for an initial testing period.

b. Technical assistance is often essen-tial; it can bring even small NGOs to thepoint of effective management.

c. Most Bosnian and Croatian NGOswere founded and managed by women—teachers, mental health workers—re-sponding to the emergency need to assistdisplaced families and then to promoteresettlement, normalization, and commu-nity reconciliation.

d. Youth and women often appearmore ready than adult males to reconcileacross ethnic lines. If this is also true inKosovo (assuming postconflict Kosovo isnot monoethnic), the bias toward reach-ing youth and women and neglectingadult males needs correction for reconcili-ation to work.

e. NGOs dedicated to specific benefi-ciaries and discrete activities (traumatizedwomen, youth voluntarism, sports, fam-ily therapy, interethnic reconciliation, cul-tural revival) should not be pressed intonew activities in which they lack exper-tise (e.g., microenterprise) merely becausedonor priorities have shifted. Instead, do-nors should encourage the creation ofgroups of NGOs (“strategic” rather than adhoc) with complementary capabilities andobjectives suitable for the array of prob-lems that need attention in specific com-munities.

f. Financial support for local NGOsfrom the ethnic Albanian Diaspora can fa-cilitate the gradual phaseout of aid depen-dence, but political capture of such remit-tances can create disillusionment andchoke off these funds, as demonstrated inCroatia.

g. Local staffs of the International Res-cue Committee and other strong NGOs inBosnia and Croatia could be valuable re-sources for aid-funded projects utilizing(or assisting) Kosovar (or other) localNGOs. For example, the experience of agen-cies helping reconstruct housing in Bosnia(such as Mercy Corps International) alsomay be valuable in Kosovo. Finally,Kosovar NGOs that may be intact in refu-gee status, outside Kosovo, should be sup-ported to ensure their survival pendingreturn.

6. Safety nets. Anticipate a need for aid-ing widow-headed households.

D5


When the fate of the large number ofmissing Kosovar males becomes known,Kosovo may join the ranks of other coun-tries that have emerged from complexemergencies with tragically distorted de-mography. A major loss of adult malescould create a severely disadvantagedgroup of widow-headed households, es-pecially for Kosovo’s large rural popula-tion. The need for effective safety nets andthe need to restore such households to eco-nomic viability should be anticipated.Other disadvantaged groups needing tai-lored support may include the elderly,widower heads of households, singlemothers, and the disabled.

7. Employment-oriented training.Train people in skills for which there is em-ployment demand and link training withstartup capital.

Recovery activities need to matchpostconflict economic realities. Job train-ing for youth, women, and demobilizedsoldiers (who have often turned to desta-bilizing criminal activity when they re-mained unemployed) is virtually fruitlessif unemployment in the depressedeconomy remains high. Although oppor-tunities for self-employment in microen-terprise may also be limited, they are likelyto be enhanced if the training is linked tothe provision of seed capital.

8. Commodity procurement. Maxi-mize regional procurement to help stimulateeconomic recovery.

Purchasing aid-funded goods andservices internally and from the affectedneighboring countries (with an eye to thepossible price effects, as noted above) cancontribute to regional employment, eco-nomic recovery, and the restoration ofcross-border economic relations. This mayrequire flexibility in the application of pro-curement regulations in the affected coun-tries.

9. High-level technocrat training.Training technocrats to manage thepostconflict economic transition should beginimmediately.

Reconstruction and development inKosovo (under non-Yugoslav administra-tion) is likely to be associated with a newset of economic institutions and rules ofthe game that mirror Western Europeannorms. As with many complex humani-tarian emergencies, overall macroeco-nomic and sectoral policy formulationduring recovery-cum-transition is likely torequire broad conceptual and manage-ment skills. Those skills are likely to be inshort supply, especially if the preconflictprofessional and intellectual cadre havebeen targeted for deliberate elimination.

Although Kosovo’s civil society haddeveloped considerable institutional expe-rience (mainly in social and cultural sec-tors) during the autonomy period of 1974–89, the fate of many civil society leaders isunknown. If possible these cadre shouldbe found, perhaps among the refugees,and readied for return to Kosovo to help

D6


revive the social infrastructure and preparefor economic recovery. A balance will beneeded between foreign implementingagencies and contractors on the one handand Kosovar organizations on the other.Otherwise, the recovery may be jeopar-dized either by overburdening theKosovars or by sinking them with over-bearing disregard.

Political Effects

1. Potential political consequences.Monitor emergency aid distribution for pos-sible unintended and undesirable political con-sequences.

Emergency aid (especially food com-modities) are a source of power. A flaweddistribution process can have unintendedand undesirable political consequences. Inseveral cases (especially Rwanda), controlover final food distribution reinforced thepower of local authorities or factions,strengthened their relative position, andfacilitated their self-aggrandizing, oftenabusive, behavior toward the intendednoncombatant beneficiaries. Conse-quences of this sort would not be surpris-ing under prevailing conditions in Kosovoand neighboring countries. Reconstructionaid in the Balkans will far exceed emer-gency food aid in amount and potentialpolitical consequences. It also will needcareful monitoring.

2. Peace conditionality. Use aid as le-verage to enforce adherence to the peace accordsand responsible governance.

In an evaluation of its ownpostconflict experience, the World Bankasserted that donors could have andshould have exerted more forcefully theleverage they had during the initialpostconflict period. This is typically a pe-riod of almost total dependence on exter-nal financing and security. The study re-fers to governance functions within thenormal scope of the Bank’s mandate (e.g.,fiscal practices). The point applies as well,though, to the notion of peace condition-ality—that is, making provision (or with-holding) of aid flows dependant on localauthorities’ adhering to the political com-mitments in the peace accords (or compa-rable instruments). Lack of donor consis-tency in this regard can undermine suchpotential (Cambodia), which reinforces theneed to design an effective coordinationstructure. The Bosnian experience withpeace conditionality is perhaps the mostpertinent for postconflict Kosovo.

3. Regional destabilization. Non-project aid for budget support can help front-line states maintain domestic stability.

There is concern that the Kosovo cri-sis may destabilize Albania and Mace-donia. The presence of Kosovar refugeesand the burdens they impose jeopardizethe ability of the two governments to con-tinue to finance and sustain precrisis lev-els of service delivery to their citizens. Aidin the form of nonproject budget supportcould help them sustain domestic budgetoutlays. It could also help diminish the po-tential for ethnic polarization in

D7

Complex Humanitarian Emergencies and USAID’s Humanitarian ResponseD8

Macedonia (e.g., by enabling the govern-ment to maintain pensions and civil ser-vice salaries) and the deepening of the so-called left–right polarization in Albanianpolitics. Fast-disbursing aid flows mayalso be critical for sustaining the positionsof these governments regarding the con-flict itself, a political point beyond thescope of this note. Understandings andcommitments regarding allocation of thelocal currency counterpart should be ex-plicit and should be monitored to ensurecompliance. Disbursements in tranchesshould be considered to encourage com-pliance.

4. Resentment of refugees. To avoidresentment, keep refugee support standardsmodest in relation to host-population stan-dards.

Political problems have arisen in hostcountries when encampment extends forsome considerable period. This may wellbe the case with the Kosovar refugeesowing to mines, housing destruction, andother problems. (One recent estimate as-sumed 3–5 years, security considerationsaside.) Evaluations have stressed that refu-gees should not be supported at a stan-dard that evokes resentment among thehost population. Resentment can also stemfrom refugees competing for scarce localemployment. After resettlement, resent-ment could arise if country allocations ofreconstruction aid are perceived as unfairor unjust. Aid to refugees should thereforeconform with regional living standards.

Bibliography

Africa Rights. 1998. Rwanda: The Insurgencyin the Northwest. London: Africa Rights.

Anderson, Mary B. 1996. Do No Harm: Sup-porting Local Capacities for Peace ThroughAid. Cambridge: Collaborative for De-velopment Action.

Apthorpe, Raymond. 1997. “Some Relieffrom Development: HumanitarianEmergency Aid in the Horn of Africa(Including Sudan), Rwanda, andLiberia.” The European Journal of Devel-opment Research, vol. 9, no. 2 (Decem-ber): 83–106.

Borton, John. 1994. “Assessing the Impactand Effectiveness of Relief Assistance.”Paper prepared for the 1995 WorldDisasters Report.

Borton, John; and Joanna Macrea. 1997.“Evaluation Synthesis of EmergencyAid.” London: Department for Inter-national Development.

Brandt, Donald P. 1995. “War Work: Hu-manitarian Assistance in Complex

Human Emergencies.” Journal of Hu-manitarian Assistance (23 October).

Brown, D.G.; P.W. Bartholomew; J–C.Lucas; and M. Jolivert. 1996. “FinalEvaluation: USAID/GOH-funded JOBS

Creation Program.” Prepared for PanAmerican Development Foundationand USAID/Haiti.

Buttari, Juan J. 1997. “Haiti: Intervention,Reform, and Politics.” Draft report.

Chandler, Dennis; and others. 1996.“Evaluation of the Haiti CommunalGovernance Program.” ManagementSystems International. Washington:USAID.

Collier, Paul; and Anke Hoeffler. 1998. “OnEconomic Causes of Civil War.” OxfordEconomic Papers 50. Oxford UniversityPress.

de Soysa, Indra; and Nils Petter Gleditsch.“To Cultivate Peace: Agriculture in aWorld of Conflict.” Oslo: InternationalPeace Research Institute.


de Waal, Alex. 1993. “War and Famine inAfrica.” IDS Bulletin, vol. 24, no. 4.

Duffield, Mark. 1998. “Aid Policy andPost-Modern Conflict: A Critical Re-view.” School of Public Policy. Bir-mingham: University of Birmingham.

Dupuy, Alex. 1997. Haiti in the New WorldOrder: The Limits of the Democratic Revo-lution. Boulder, Colo.: Westview Press.

Government Accounting Office. 1985.“The United States’ Response to theEthiopian Food Crisis.” Washington.

———. 1986. “Famine in Africa: Improv-ing Emergency Food Relief Programs.”Washington.

Hallam, Alistair. 1998. “Evaluating Hu-manitarian Assistance Programmes.”Prepared for the DAC Expert Group onAid Evaluation and reviewed by theDAC Task Force on Conflict, Peace, andDevelopment Cooperation, February2–3.

Hansch, Steve; and others 1994. “LivesLost, Lives Saved: Excess Mortalityand the Impact of Health Interventionsin the Somalia Emergency.” Washing-ton: Refugee Policy Group.

Hill, Richard. 1997. “Post–Cold War Re-construction Programs: ObservationsFrom the Field on the Sword and thePloughshare.” Paper presented at aConference on Urban Triumph or Ur-

ban Disaster?: Dilemmas of Contem-porary Post-War Reconstruction. Cam-bridge, Mass.

Ianotti, L. 1997. “Nutrition in Haiti: Con-straints and Solutions.” Washington:Inter-American Development Bank.

Inter-American Development Bank. 1994.“Emergency Economic Recovery Pro-gram: Haiti.” Report of the Joint As-sessment Mission. Washington.

Joint Evaluation of Emergency Assistanceto Rwanda. 1996. “Early Warning andConflict Management.” Study 2. TheInternational Response to Conflict andGenocide: Lessons from the Rwanda Ex-perience.

———. 1996. “Humanitarian Aid and Ef-fects.” Study 3. The International Re-sponse to Conflict and Genocide: Lessonsfrom the Rwanda Experience.

Kitay, Michael. N.d. “Land Tenure Issuesin Postconflict Countries.” Washing-ton: USAID.

Kleist, Ted. See Netherlands DevelopmentCooperation. 1994.

Lautze, Sue; Bruce D. Jones; and MarkDuffield. 1998. Strategic HumanitarianCoordination in the Great Lakes Region1996–97. Office for the Coordination ofHumanitarian Affairs. New York:United Nations.

Biblio2

Bibliography

Lieberson, Joseph; and others. 1999. “Pro-viding Emergency Aid to Mozam-bique.” Center for Development Infor-mation and Evaluation. Washington:USAID.

Maguire, Robert; and others. 1996. HaitiHeld Hostage: International Responses tothe Quest for Nationhood, 1986 to 1996.Occasional Paper no. 23. Providence:Watson Institute.

McClelland, Donald G.; and others. 1999.“Providing Emergency Aid to Haiti.”Center for Development Informationand Evaluation. Washington: USAID.

McHugh, Heather S.; and Joshua M.Kaufman. N.d. “A Summary of Prin-cipal Lessons Learned from USAID Ex-perience in Humanitarian Assistance.”Washington: USAID.

Messer, Ellen; Marc J. Cohen; and JashintaD’Costa. 1998. “Food from Peace:Breaking the Links Between Conflictand Hunger.” Washington: Interna-tional Food Policy Research Institute.

Miller, John M. 1997. “USAID/Washingtonand Field Collaboration in the Use ofCentral Resources.” Washington.

Minear, Larry. 1994a. “Humanitarian Ac-tion in the Former Yugoslavia: TheUN’s Role 1991–93.” Providence R.I.:Watson Institute.

———. 1994b. “The International ReliefSystem: A Critical Review.”

———. 1998. “Learning to Learn.” Discus-sion Paper prepared for a seminar onLessons Learned in Humanitarian Co-ordination. Office for the Coordinationof Humanitarian Affairs and the Min-istry of Foreign Affairs of Sweden.

Nafziger, E. Wayne. 1996. “The Econom-ics of Complex Humanitarian Emer-gencies: Preliminary Approaches andFindings.” UNU/WIDER.

Naval, G. 1995. “Evaluation of the CrisisProgram of the Catholic Relief Servicesin Haiti.” Final report.

Netherlands Development Cooperation.1994. Humanitarian Aid to Somalia.Evaluation Report directed by TedKleist. Operations Review Unit.

Organization for Economic Cooperationand Development. 1999. “Guidance forEvaluating Humanitarian Assistancein Complex Emergencies.” Develop-ment Assistance Committee. Paris.

Organization for Economic Cooperationand Development. DAC/o Database(http://www.oecd.org/dac/).

Pan American Development Foundation.N.d. “Jobs Creation Project: Final Re-port.” Washington.

Biblio3

Complex Humanitarian Emergencies and USAID’s Humanitarian ResponseBiblio4

Pasic, Amir; and Thomas G. Weiss. 1997.“Humanitarian Recognition in theFormer Yugoslavia: The Limits of Non-State Politics.” Security Studies 7, no. 1(autumn 1997): 194–227.

Prendergast, John; and Colin Scott. 1996.“Aid with Integrity: Avoiding the Po-tential of Humanitarian Aid to SustainConflict: A Strategy for USAID/BHR/OFDA in Complex Emergencies.”Occasional Paper no. 2. Washington.

Prunier, Gérard. 1995. The Rwanda Crisis:History of a Genocide. New York: Colum-bia University Press.

Renison, Bill; and others. 2000. “Humani-tarian Assistance in Rwanda: HardChoices, No Exits.” Draft Report.Washington: USAID.

Sen, Amartya K. 1993. “The PoliticalEconomy of Hunger.” Overcoming Glo-bal Hunger . Ismail Serageldin andPierre Landell–Mills, eds. Proceedingsof a Conference on Actions to ReduceHunger Worldwide. Washington.

UNHCR/WFP. 1998. “Joint Evaluation ofEmergency Food Assistance to Return-ees, Refugees, Displaced Persons, andOther War-affected Populations inBosnia and Herzegovina.” Rome.

USAID. Congressional Presentation, FY 1990–FY 2000. Various years. Washington.

———. 1995. “Survey Report on Losses ofPublic Law 480 Commodities in HaitiDuring the Political Transition PeriodSeptember 15, 1994, to January 15,1995.” Audit Report No. 1–521–95–008.

———. 1998a. Agency Performance Report,1997. Center for Development Infor-mation and Evaluation. Washington.

———. 1998b. “Results Review and Re-source Request: Fiscal Year 2000.” Bu-reau for Humanitarian Response. Pre-liminary draft.

———. 1999. Agency Performance Report,1998. Center for Development Infor-mation and Evaluation. Washington.

USAID/Haiti. 1992–96. “USAID MonitoringReport” 1–5. Port-au-Prince.

USAID/Liberia. 1998. “Fiscal Year 2000 Re-sults Review and Resource Request.”Bureau for Humanitarian Response.

U.S. Mission to the United Nations. 1997.“Global Humanitarian Emergencies,1997.”

———. 1998. “Global HumanitarianEmergencies, 1998–99.”

Uvin, Peter. 1998. Aiding Violence: The De-velopment Enterprise in Rwanda. WestHartford: Kumarian Press.

Bibliography

World Bank. 1998a. “Haiti: The Challengesof Poverty Reduction,” vols. I and II.Technical papers. Washington.

———. 1998b. “Haiti: The World Bank’sExperience With Postconflict Recon-struction,” vol. v. Desk Reviews ofCambodia, Eritrea, Haiti, Lebanon,Rwanda, and Sri Lanka. Report no.17769. Washington.

Biblio5

———. 1998c. “Rwanda: Poverty Update.”Washington.

World Food Program. 1998. “The Food AidMonitor.” 1997 Food Aid Flows. Spe-cial issue. Rome: International FoodAid Information System.

c humanitarian emergencies and usaidâ„¢s humanitarian response

Documents