Document Sample
             USAID Program and Operations Assessment Report No. 27

                       Center for Development Information and Evaluation

                                                          December 2000


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

       USAID Development Experience Clearinghouse (DEC)
       1611 N. Kent Street, Suite 200
       Arlington, VA 22209–2111
       Telephone: (703) 351–4006
       Fax: (703) 351–4029
       To access CDIE documents from the Internet, key in Click on Publications/Partner
       Resources, then on USAID Evaluation Publications.

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


The views and interpretations expressed in this report are those of the author
and not necessarily those of the U.S. Agency for International Development.
                                 USAID Program and Operations
                                 Assessment Report No. 27

Complex Humanitarian Emergencies
         And USAID’s
    Humanitarian Response

                  Donald G. McClelland
    Center for Development Information and Evaluation


                    Elizabeth Adelski
                       Richard Hill
                       John Mason
                      Robert Muscat
           International Resources Group, Inc.

    Center for Development Information and Evaluation
        U.S. Agency for International Development

                     December 2000
Preface ........................................................ v       Haiti .................................................... 2 5
                                                                         Mozambique ...................................... 26
Summary ................................................ vii             Rwanda .............................................. 2 7
                                                                         Bosnia–Herzegovina ......................... 28
1. Introduction ........................................ 1               Policy Implications ........................... 29
   Number of Emergencies                                                 Conclusion ......................................... 3 2
     And People Affected ....................... 1
   Resource Implications ........................ 2                  5. Results: Economic Effects ............... 33
   Legislative Authority .......................... 4                   Haiti .................................................... 3 4
   Complex Humanitarian                                                 Mozambique ...................................... 35
     Emergencies ....................................... 5              Rwanda ............................................... 36
   General Evaluation                                                   When Is the Emergency Over? ........ 38
     Approach ............................................ 6            Conclusion ......................................... 4 0

2. Country Context ................................. 9               6. Conclusions, Lessons Learned,
   Causes of Complex                                                    And Recommendations .................. 43
    Emergencies ..................................... 9                 Conclusions and
   USAID’s Humanitarian                                                  Lessons Learned ............................... 43
    Response ......................................... 1 2              Recommendations ............................ 4 6

3. Results: Humanitarian Effects ........ 15                         Annex A Humanitarian Emergencies
   Haiti .................................................... 1 5       And Donor Assistance
   Mozambique ...................................... 17              Annex B Evaluation Objectives
   Rwanda .............................................. 1 9          And Constraints and Implications
   Bosnia–Herzegovina ......................... 20                    For Donor Coordination
   Targeting ............................................ 2 1        Annex C Are Complex Emergencies
   Efficiency ............................................ 22         Predictable?
   Conclusion ......................................... 2 3          Annex D Implications for the Kosovo Crisis

4. Results: Political Effects and                                    Bibliography
   Effects on Hostilities ....................... 25

C    OMPLEX HUMANITARIAN         emergen-
      cies are by their nature multifaceted
and involve many actors. The relief inter-
                                              as well—exacerbates an already complex

ventions are often undertaken in a context       Given these circumstances, it is not
beyond the control of the implementing      surprising that humanitarian assistance
agencies. Moreover, the interventions are   has been subjected to less rigorous and
generally conditioned by overall foreign    extensive monitoring and evaluation than
policy considerations, which means that     development assistance. In addition to the
political objectives help define the re-    complexities just noted, this reflects the
sponse.                                     fact that until recently there was no stan-
                                            dard methodology for evaluating humani-
     Operational coordination is compli- tarian assistance. Some have likened this
cated because multiple players are in- situation to “methodological anarchy.”
volved (various U.S. government agencies,
other bilateral and multilateral develop-        Nevertheless, it is possible to assess
ment agencies, nongovernmental and pri- the impact of humanitarian assistance on
vate voluntary organizations, and the host vulnerable populations to some degree
country) and these actors often have di- and to shed light on the relationship be-
vergent approaches to strategic planning, tween emergency assistance and the po-
decision-making, and delivery mecha- litical and development processes at work.
nisms. That relief experts have different This assessment and its lessons learned
views of the purpose of emergency assis- should contribute to formulating more ef-
tance—whether it is for relief only, or for fective policies and interventions in re-
rehabilitation and economic development sponse to complex emergencies.

I   N 1998 , some 32 million people needed      billion (11.4 percent of ODA). By 1997 it had
    humanitarian assistance because they gradually decreased to $344 million (4.9
were caught up in complex emergencies percent of ODA), but in 1998 it more than
(armed conflicts or civil wars as distinct doubled to $898 million (10.2 percent of
from natural disasters). That is triple the ODA). Although U.S. ODA as a percentage
number of a typical year from the early of total ODA has been falling steadily since
1980s. Most of these people are refugees the 1970s, the United States continues to
or internally displaced persons. About 40 be a generous provider of humanitarian
percent reside in Africa. The value of hu- assistance. USAID’s Food for Peace Office,
manitarian assistance worldwide has typi- Office of Foreign Disaster Assistance, and
cally been less than 2.0 percent of official Office of Transition Initiatives are prima-
development assistance ( ODA). But in 1994 rily responsible for administering U.S.
it jumped to 6.8 percent ($4.3 billion) ow- emergency assistance.
ing to the crisis in the Great Lakes region
of Africa. It fell to an estimated 5.7 per-           This evaluation seeks to assess the ef-
cent of ODA in 1998 but is likely to peak fectiveness of U.S. humanitarian assistance
again in 1999 owing to the Kosovo and in nations afflicted by complex emergen-
East Timor crises. The fact remains that hu- cies. It addresses three principal questions:
manitarian assistance has more than Did U.S. emergency assistance save lives
doubled since 1990 despite diminishing and alleviate suffering? Did it affect social
foreign assistance.                             tensions and political hostilities? Did it
                                                contribute to long-term economic devel-
      In 1990, U.S. ODA totaled nearly $13.6 opment? The findings are based on field-
billion in real terms (1998 dollars); by 1997, work carried out in three countries (Haiti,
it had fallen by half to $7.0 billion, the low- Mozambique, and Rwanda) as well as on
est level since World War II. By contrast, evaluation results in other countries.
U.S. humanitarian assistance has in-
creased. In 1990, it was $263 million (1.9            Evaluating relief programs in the con-
percent of ODA). In 1994 it peaked at $1.2 text of armed conflict or civil war intro-
duces politically sensitive issues concern-     rights abuses. Tens of thousands of refu-
ing sovereignty, international law, the         gees fled Haiti (often as boat people). Mil-
appropriate balance of aid between oppos-       lions fled Mozambique and Rwanda to
ing sides, and donors’ foreign policy           escape indiscriminate terror. Hundreds of
interests. This assessment treats the politi-   thousands were the victims of wholesale
cal effects of humanitarian assistance          massacre or, in the case of Rwanda, geno-
independently of U.S. foreign policy con-       cide.
siderations. Nevertheless, these and other
issues peculiar to complex emergencies               Donors, including USAID, responded
made the evaluation methodologically            with increased emergency assistance, both
more difficult.                                 food and nonfood (water, seed, farming
                                                tools, medical supplies). Nongovernmen-
     Complex emergencies are typically          tal organizations were the main im-
political in nature, characterized by vio-      plementers of the humanitarian response.
lent conflict (often war) and a breakdown       In Haiti, the international community was
of institutions. But their underlying causes    feeding 1.3 million people—one in seven
vary. Predatory governance was the prin-        Haitians—each day, with the United States
cipal cause of Haiti’s complex emergency.       providing 68 percent of the food. In
By contrast, ethnic and ideological factors     Mozambique in 1989, an estimated one
were pivotal in Mozambique and Rwanda.          third of the population of 16 million de-
In all three countries, poverty was a con-      pended on food aid for 60 to 70 percent of
tributory factor. In Haiti, per capita income   their food needs; again, the United States
was $250 in 1994; in Mozambique, $80 in         provided about 60 percent of total food aid
1986, the lowest in the world. Poverty was      during 1987–95. In Rwanda 1.3 million
just as severe in Rwanda. Moreover, the         beneficiaries received emergency food aid
distribution of income and wealth was           in 1996–97.
highly skewed in all three countries.
Haiti’s 200-year history has been charac-
terized by oppressive governments that fa-            What were the results? The assess-
vored the rich at the expense of the poor.      ment concluded that emergency assistance
Mozambique was characterized by a               programs funded by USAID and imple-
highly dualistic economy. In Rwanda,            mented by U.S. nongovernmental agencies
where the proportion of people living in        (NGOs) clearly helped save lives and alle-
poverty increased from 40 percent to 70         viate suffering—which, after all, is their
percent during 1990–93, a winner-take-all       overarching objective. Except for Haiti,
mentality has benefited a tiny elite at the     though, data collection and monitoring
expense of the poor majority.                   were not done (or were done poorly), so it
                                                is difficult to quantify results. In fact, most
    Civilians in all three countries suf-       evaluations of humanitarian assistance tell
fered widespread and systematic human           a “mission accomplished” story but are

viii              Complex Humanitarian Emergencies and USAID’s Humanitarian Response
unable to substantiate that story with hard    alive in a political vacuum but affects the
data.                                          local power structure and environment in
                                               which it is given.
     Distributing relief supplies was a
problem to some extent in all three coun-            In Haiti, massive quantities of emer-
tries. Food aid, in particular, was highly     gency food aid reduced the probability of
valued and became a source of violent          food riots during a period of political and
competition—not only for its value as food     economic stress and may have had a
but also as a source of political power for    dampening effect on political tensions; but
those controlling access. There were re-       it also may have contributed to a political
ports of corruption, theft, and political or   status quo that enabled the de facto mili-
personal favoritism in food aid distribu-      tary regime to stay in power longer. In
tion. And target populations did not al-       Mozambique, external military assistance
ways receive timely and sufficient food.       provided by the Soviet Union and South
NGOs addressed these problems with vary-       Africa fueled the civil war; food aid, by
ing degrees of success. In Haiti they were     comparison, had relatively little effect on
highly successful in limiting diversion to     the country’s political dynamics, although
5 to 10 percent. In Mozambique, leakage        food diverted to soldiers may have con-
was typically 30 percent when the govern-      tributed to the war effort. In Rwanda,
ment was in charge of distribution, and at     genocidal killers were mixed with legiti-
one point reached 50 percent. But after the    mate refugees in camps; targeting became
NGO s took over, losses fell to under 5 per-   problematic, and substantial quantities of
cent. In Rwanda the military and former        food aid were diverted by Hutu extrem-
political leaders controlled much of the       ists and militia resident in the camps. That
relief distribution. They were able to di-     had the unintended effect of prolonging
vert substantial quantities of food (more      the conflict.
than is usually the case in complex emer-
gencies) from the intended beneficiaries
for their own purposes.                            The notion that relief assistance can
                                              be made more developmental in the con-
     While no aid is apolitical, humanitar- text of ongoing armed conflicts is problem-
ian assistance, in particular, can result in atic. Unlike with natural disasters, during
substantial and unpredictable political ef- complex emergencies there is no institu-
fects, since it is provided in the context of tional framework to provide physical se-
conflict. Though designed to relieve suf- curity and political stability—both of
fering and promote peace, it sometimes, which are necessary preconditions for eco-
inadvertently, fuels, sustains, or worsens nomic development. On the contrary, com-
complex emergencies by making more re- plex emergencies are often characterized
sources available to warring parties. This by a total breakdown of state institutions
is because aid does not just keep people and social and economic structures.

Summary                                                                                ix
     Nevertheless, emergency assistance s Adverse political consequences. Be alert
programs can help shape the pattern and             to potential undesirable political or social ef-
direction of subsequent economic devel-             fects that relief aid may cause. Control of the
opment. In Haiti, Mozambique, and                   distribution of food aid, in particular, can
Rwanda USAID and the NGOs not only pro-             reinforce the power of local authorities
vided immediate relief (food, medicine)             or political factions; it can also facilitate
but also agricultural inputs (seed, tools)          their self-aggrandizing, often exploitive,
and household goods to encourage refu-              behavior toward the intended noncom-
gees and internally displaced persons to            batant beneficiaries.
return to their villages, resume food pro-
duction, decrease their dependence on s Reducing dependency. Give refugees in-
food aid, and maintain their livelihoods.           centives to return home, and impose disin-
They also implemented food-for-work                 centives on those remaining outside their
programs in all three countries. These and          country of origin. After populations have
other programs created short-term jobs              been repatriated and are settled, the ag-
and helped rehabilitate productive infra-           ricultural base begins to be reestab-
structure (roads, irrigation) needed for            lished, dependency on free food drops,
economic development. The development-              and long-term food security is en-
oriented objectives were clear: to restart          hanced.
subsistence agriculture and to restart the
rural economy.                                   s Capacity building. Train technocrats to
                                                    manage the postconflict economic transition,
     The assessment offers 4 management-            and train others in skills for which there is
oriented recommendations (summarized                employment demand. Economic recovery
below) and 18 recommendations specific              requires a cadre of high-level techno-
to the Kosovo crisis as of May 1999                 crats with management and conceptual
(annex D).                                          skills; it also requires the unemployed
                                                    (especially demobilized soldiers) to be
                                                    trained in marketable skills.
s Monitoring and evaluation. Establish a
   central monitoring and data collection unit        Finally, however one assesses the ef-
   to serve all donors during the early weeks of fectiveness of humanitarian assistance in
   a complex emergency. This is needed, response to complex emergencies, one
   among other things, to help managers thing cannot be emphasized too strongly:
   identify appropriate kinds of emer- it is far better to prevent complex emer-
   gency relief, target its distribution, gencies from occurring in the first place
   evaluate its effectiveness, and enhance than it is to respond to victims’ needs af-
   donor coordination.                           terwards.

  x                 Complex Humanitarian Emergencies and USAID’s Humanitarian Response
                           1       Introduction

S   INCE THE END of the Cold War, the na-
     ture of international assistance needs
has changed dramatically. Ethnic and na-
                                              serious malnutrition or death peaked dur-
                                              ing 1993–95. With improved situations in
                                              several countries (including Armenia,
tional tensions have led to increased civil   Cambodia, and Mozambique), the num-
strife and an explosion in the number of      ber of emergencies dropped to 20 in 1996
complex humanitarian emergencies. As a        and remained at that level in 1997 (p. 5).
result, the number of civilian casualties has In 1998, Russia (Chechnya) was dropped
increased, as has the level of emergency      from the list (reducing the number to 19),
assistance allocated in response to their hu- but Colombia and Uganda were added
manitarian needs. In 1998, USAID’s Center     (boosting it to 21) (U.S. Mission to the UN
for Development Information and Evalu-        1998, 7, 9). Annex A (table A1) lists ongo-
ation (CDIE) initiated an assessment of the   ing humanitarian emergencies in 1996,
effectiveness of the Agency’s humanitar-      1997, and 1998 using data from the U.S.
ian assistance interventions. The assess-     Committee for Refugees.
ment examined USAID programs in three
countries afflicted with complex emergen-          Worldwide, roughly 33 million
cies: Haiti, Mozambique, and Rwanda. people needed emergency assistance in
This report synthesizes the findings of the January 1996. That increased to 34 million
three separate country studies.               in January 1997, then decreased to an esti-
                                              mated 32 million in April 1998 (table A1).
                                              These levels are triple those typical of the
Number of Emergencies                         early 1980s. They include both internally
And People Affected                           displaced persons who have remained
                                              within their own borders and refugees
      According to the U.S. Mission to the who have fled across international bor-
United Nations (1997, 5), the number of ders. During the 1990s most have been in-
humanitarian emergencies in which at ternally displaced persons rather than
least 300,000 civilians depended on inter- refugees. In 1996, 52 percent of those re-
national humanitarian assistance to avoid quiring emergency assistance owing
mainly to armed conflict or government              gradually from a high of $68.5 billion in
repression resided in sub-Saharan Africa;           1991 to a low of $53.4 billion in 1997, a de-
in 1997 that percentage decreased to 48             crease of 22 percent. In 1998 it increased
percent, and in 1998 it decreased still fur-        for the first time since 1994; the increase
ther to 39 percent (table A1).                      was 8.1 percent. (The small increase in 1994
                                                    was probably due to the Rwanda crisis.)

Resource Implications                             Humanitarian assistance peaked at
                                             $4.3 billion in 1994 owing to the crisis in
     The Development Assistance Com- the Great Lakes region of Africa. After that
mittee of the Organization for Economic it fell to $3.3 billion as of 1998. However,
Cooperation and Development monitors it was expected to peak again in 1999 be-
levels of official development assistance cause of the emergencies in Kosovo and
and humanitarian assistance. Table 1 re- East Timor and the Turkish and Taiwan-
ports these data for all donors in 1998 dol- ese earthquakes. Humanitarian assistance
lars during 1988–98. In real dollar terms, had typically been less than 2.0 percent of
official development assistance declined official development assistance—until

   Table 1. Official Development Assistance and Humanitarian
          Assistance, in Millions of 1998 US$, 1988–98
        Year      Official Development      Humanitarian Assistance       HA as a Percent Of ODA

 1988                     61,570                       955                           1.6
 1989                     60,172                       969                           1.6
 1990                     63,791                     1,270                           2.0
 1991                     68,503                     3,503                           5.1
 1992                     67,792                     2,880                           4.2
 1993                     62,659                     3,863                           6.2
 1994                     63,176                     4,303                           6.8
 1995                     56,968                     3,401                           6.0
 1996                     56,530                     3,206                           5.7
 1997                     53,424                     2,921                           5.5
 1998                     57,774                     3,288                           5.7
 Source: OECD DAC/o database

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

  2                 Complex Humanitarian Emergencies and USAID’s Humanitarian Response
1991, when it jumped to 5.1 percent (ow-          humanitarian assistance, by contrast, has
ing to violence in the Balkans). It peaked        increased. In 1990 it was $263 million (1.9
at 6.8 percent of official development as-        percent of ODA); in 1994, it peaked at $1.2
sistance in 1994 with the Rwanda crisis,          billion (11.4 percent of ODA). Since then it
but afterward fell to 5.7 percent of ODA in       has gradually decreased to $344 million
1998. The fact remains that within a dimin-       in 1997 (4.9 percent of ODA). In 1998 both
ishing overall foreign aid budget, humani-        U.S. ODA and U.S. humanitarian assistance
tarian assistance has nearly tripled since        increased, and in 1999 humanitarian as-
1990.                                             sistance was expected to increase again
                                                  given current humanitarian needs.
     What about the United States? In 1990,
U.S. official development assistance to-               Although U.S. official development
taled nearly $13.6 billion in real terms          assistance as a percentage of total ODA has
(1998 dollars); by 1997, it had fallen by half    been falling steadily since the 1970s, the
to $7.0 billion in real terms (see table 2).      United States continues to be a generous
This was the lowest level of U.S. assistance      provider of humanitarian assistance. In
since World War II (Miller 1997, 1). U.S.         1998, for example, the United States allo-

 Table 2. U.S. Development Assistance and Humanitarian
          Assistance, in Millions of 1998 US$, 1988–98
        Year     Official Development     Humanitarian Assistance      HA as a Percent of ODA

 1988                     13,141                     220                            1.7
 1989                      9,547                     261                            2.7
 1990                     13,580                     263                            1.9
 1991                     12,951                     685                            5.3
 1992                     13,146                     585                            4.4
 1993                     11,099                     733                            6.6
 1994                     10,662                   1,216                           11.4
 1995                      7,743                     829                           10.7
 1996                      9,669                     603                            6.2
 1997                      6,959                     344                            4.9
 1998                      8,786                     898                           10.2
 Source: OECD DAC/o database

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

Introduction                                                                                     3
cated over 10 percent of its ODA to humani-      with priority given to people suffering
tarian assistance needs. However, to meet        from malnutrition. Cooperating sponsors
humanitarian needs, U.S. policymakers            can be (1) governments, (2) multilateral or-
have had to divert resources away from           ganizations such as the World Food Pro-
sustainable development programs                 gram, or (3) nonprofit U.S. private volun-
(Messer 1998, 15, citing USAID’s FY97 Con-       tary organizations (PVOs) such as Catho-
gressional Presentation). Any proliferation      lic Relief Services, Cooperative for Assis-
of complex emergencies is likely to tighten      tance and Relief Everywhere (CARE ),
the squeeze on sustainable development           Adventist Development and Relief
programs even further. Ironically, devel-        Agency, and World Vision. Cooperating
opment assistance programs designed to           sponsors are responsible for establishing
spur economic growth and reduce poverty          distribution networks to reach disaster
may help mitigate the need for more ex-          victims and for properly storing and ac-
pensive responses to complex emergen-            counting for commodities. USAID is re-
cies, since they reduce the likelihood of        sponsible for the overall administration
their occurring in the first place.              and management of the program.

                                                      An emergency response normally re-
Legislative Authority                            quires not only food but also nonfood as-
                                                 sistance, including medicine, sanitation,
      The United States donates food aid to      potable water, agricultural inputs, and
victims of floods, earthquakes, droughts,        shelter. USAID’s Office of Foreign Disaster
and civil strife under Title II of the Agri-     Assistance and Office of Transition Initia-
cultural Trade Development and Assis-            tives are primarily responsible for admin-
tance Act of 1954 (PL 480). Other types of       istering nonfood humanitarian assistance.
humanitarian assistance are provided un-         OFDA coordinates the allocation of funds
der the Foreign Assistance Act of 1961 (as       appropriated under the International Di-
amended). Normally, the duration of both         saster Assistance account. The largest per-
food and nonfood emergency assistance            centage of funds goes to relief and reha-
is limited, and only countries that lack re-     bilitation project grants managed by pri-
sources to purchase commodities commer-          vate voluntary, nongovernmental, and in-
cially are eligible recipients (GAO 1986, 10).   ternational organizations.

      USAID’s Office of Food for Peace ad-           OFDA has an internal policy not to ob-
ministers the PL 480 Title II food aid pro-      ligate funds for longer than 12 months at
gram. Under this program the United              a time. That allows it to respond to unex-
States provides emergency food aid to co-        pected crises worldwide. Food for Peace
operating sponsors who in turn distribute        has a similar policy for emergency food
it to disaster victims. Food commodities         aid. The Office of Foreign Disaster Assis-
are distributed in areas of greatest need        tance is widely respected for its respon-

  4               Complex Humanitarian Emergencies and USAID’s Humanitarian Response
siveness to disasters. Although it is           In the post–Cold War period, most have
permitted to provide responses aimed at         been conflicts that have taken place within,
relief, rehabilitation, and reconstruction,     not between, countries. These internal
OFDA’S lifesaving emphasis is paramount.        struggles among warring factions are usu-
OFDA is concerned about potential con-          ally defined by ethnicity, religion, or lan-
gressional criticism if it permits missions     guage. According to Apthorpe (1997, 91),
or embassies to use emergency assistance        complex emergencies have deep roots and
for longer term developmentally related         dense branches and cannot be understood
interventions (Miller 1997, 27).                from a Western-ethnocentric perspective.
                                                Similarly, Kleist (1994, 45) suggests that a
                                                disaster may be defined as complex when
Complex Humanitarian                            its origins are multiple and its effects com-
Emergencies                                     pound one another. As USAID’s 1998 Per-
                                                formance Report (131) notes, complex emer-
     The term “complex emergency” was gencies are manifested by “armed conflict,
first coined in UN circles, probably in death, displaced populations, hunger, and
Mozambique, as a diplomatic euphemism injury.”
for a “chronic political” rather than “natu-
ral” emergency. The Joint Evaluation of               Humanitarian aid (donations of food
Emergency Assistance to Rwanda (1996, and other commodities and services) is in-
study 2, 5) points out that complex emer- tended to save lives in situations where
gencies                                         virtually everyone is at exceptionally high
                                                risk. These situations require getting as-
     tend to have multiple causes, but are      sistance to where it is needed—urgently.
     essentially political in nature and entail The humanitarian response is primarily an
     violent conflict. They typically include a act of rescue. Under the umbrella of hu-
     breakdown of legitimate institutions and   manitarian assistance, there is a distinction
     governance, widespread suffering, and      between relief and rehabilitation (Kleist,
     massive population displacements, and
     they often involve and require a range     47). Relief helps people survive; rehabili-
     of responses from the international com-   tation helps people get back on their feet
     munity, including intense diplomacy and    so they can reestablish their livelihoods.
    conflict resolution efforts, UN policing
    actions, and the provision of multilat-    Development aid, by contrast, is nor-
    eral and bilateral humanitarian assis-
                                          mally long term and sustainable in nature.
    tance by official and private agencies.
    A complex emergency tends to be very  In economic terms, it can be characterized
    dynamic, characterized by rapid       as investment rather than consumption.
    changes that are difficult to predict.There is no clear, operational definition of
                                          when short-term relief ends and long-term
   Brandt (1995, 1) suggests that complex development begins. For all practical pur-
humanitarian emergencies are often wars. poses, though, relief activities end with the

Introduction                                                                             5
termination of emergency resources. And      Mozambique (Lieberson 1999), and
this occurs when donors decide to cease      Rwanda (Renison 2000). They also draw
providing these resources. This suggests     on syntheses of related evaluation results
that the relief-to-development continuum     including in particular Apthorpe (1997)
may exist conceptually, but not operation-   covering six evaluations of humanitarian
ally.                                        assistance in Africa; Borton and Macrea
                                             (1997) covering 28 evaluations worldwide;
                                             and the UNHCR/WFP (1998) evaluation of
General Evaluation                           the experience in Bosnia and Herzegovina.
                                                  CDIE evaluation teams carried out key
     The overall objective of the CDIE        informant interviews with beneficiaries
assessment was to examine the effective-      and a broad range of experts who had
ness of U.S. humanitarian interventions,      managed or implemented emergency as-
especially emergency food aid, in nations     sistance programs. Information was col-
afflicted by a complex emergency. It          lected in both urban and rural areas; site
addressed three principal questions:          visits within each country also produced
                                              valuable insights. Illustrative questions
     1. Did U.S. emergency assistance asked during the interviews included the
save lives and alleviate suffering during following:
the complex emergency (humanitarian
effects)?                                          s What was the political, economic,
                                              and social context in which humanitarian
     2. Did U.S. emergency assistance assistance was provided?
affect social and political hostilities or
tensions associated with the complex               s What were the perceived results of
emergency (political effects)?                the assistance in terms of saving lives,
                                              affecting hostilities, and contributing to
     3. Did U.S. emergency assistance development?
contribute to long-term development
(economic effects)?                                s Were the results achieved those that
                                              were intended (i.e., what was the relation-
     The assessment examined the results ship between results and objectives)?
of humanitarian assistance. It did not delve
into the various agencies involved in              s Were there unintended effects,
implementation. In short, it did not evalu- positive or negative?
ate the implementing agencies, but rather
the results of their activities. The findings      s Were the interventions sustainable
are based on fieldwork carried out in three (in the case of rehabilitation assistance as
countries: Haiti (McClelland 1999), distinct from relief aid)?
    s What were the key strong and weak     of the people involved (beneficiaries, man-
points of the assistance; (i.e., the major  agers, implementers) had moved on and
successes and failures)?                    could not be reached, and some of the
                                            institutional mechanisms had been dis-
      The very nature of complex emergen- mantled. Because action came first, paper-
cies imposed certain methodological limi- work was frequently not given priority, so
tations on the study. For example, there relevant data were lacking or conflicting.
was little evidence of long-term develop- Finally, the contribution of the United
ment-oriented effects, since achieving States to operations that were cofinanced
short-term effects was the principal objec- could not be separated from the contribu-
tive of the humanitarian assistance. Many tions of other donors.
                           2       Country Context

T   HE TERM     “complex humanitarian
     emergency” is relatively new in the
American lexicon. Complex emergencies
                                               subservient police, and both institutions
                                               have engaged in widespread and system-
                                               atic human rights abuses with nearly com-
are generally characterized by a break-        plete impunity.
down of institutions and governance. They
always involve conflict, often war. What            The situation boiled over during
causes them, and what humanitarian tools       1991–94, shortly after a military coup re-
has USAID used to alleviate widespread         moved democratically elected President
suffering? This section examines these         Jean–Bertrand Aristide from office. Al-
questions in the context of the three coun-    though never an all-out civil war, this pe-
try case studies: Haiti, Mozambique, and       riod bore all the hallmarks of a complex
Rwanda.                                        emergency with political, social, and eco-
                                               nomic collapse. Human rights violations
                                               swelled to unprecedented levels, prompt-
Causes of Complex                              ing a series of UN-backed sanctions in-
Emergencies                                    cluding the U.S.–led international em-
                                               bargo. By September 1994, an estimated
     Haiti. Most studies have singled out      300,000 of Haiti’s 7 million people were
predatory governance as the principal          displaced internally; another 60,000 to
cause of Haiti’s complex emergency; eth-       70,000 were refugees, some as the highly
nic and ideological factors appear less im-    publicized boat people (World Bank
portant. Haiti has almost no history of        1998b); thousands had fled across the bor-
democratic governance or strong public in-     der to the Dominican Republic; and 4,000
stitutions. Instead, during its nearly 200-    had been killed (Dupuy 1997). Gross do-
year history, oppressive governments have      mestic product fell by 35 percent during
favored the rich at the expense of the poor.   this period and inflation increased to 50
The country’s military has controlled a        percent by 1994 (Buttari 1997). An esti-
mated 143,000 jobs were lost in the private   They cut railway and power lines, de-
sector (Maguire 1996). The inflation-ad-      stroyed roads and bridges, and sabotaged
justed value of the minimum wage was          oil-storage depots. They raided towns and
less than it had been 10 years earlier, and   villages and sometimes engaged in the
per capita GNP was $250.                      wholesale massacre of civilians. Mozam-
                                              bique’s socialist allies countered by pro-
      Mozambique. In 1975, after a 10-year    viding the Frelimo government with
war for independence, Mozambique in-          weapons and financial support.
herited a highly dualistic colonial
economy that lacked schools, health facili-         The result was civil war—nominally
ties, and other public services. With the     based on ideology but actually supported
end of colonial rule, most of the Portu-      by foreign countries in the context of Cold
guese and many skilled Mozambicans            War politics and fueled by a drive for
fled, leaving the country without the tech-   power by local military and political lead-
nical skills needed to operate factories or   ers. It was mainly a low-intensity, hit-and-
the transport system, to manage commerce      run guerrilla war fought largely with small
or government, or to provide professional     arms and land mines to destroy economic
services. The Front for the Liberation of     and social infrastructure. Both armies ter-
Mozambique (Frelimo), the insurgent           rorized the rural population by seizing
group that had fought for independence,       food and killing people.
took control after the Portuguese left. The
Frelimo government established a one-              Over 2 million people fled to neigh-
party state and a centrally planned           boring countries and 4 to 6 million moved
economy modeled after those in Eastern        to areas of relative safety within
Europe.                                       Mozambique. As many as 8 million people
                                              in a country of 16 million were affected—
      Mozambique’s neighbors—white-           a reflection of the large-scale human suf-
ruled Rhodesia and South Africa, which        fering and economic dislocations that took
supported apartheid—were alarmed by a         place. In 1986 the economy hit bottom: per
black-ruled, antiapartheid socialist coun-    capita GNP was $80, the lowest in the
try on their borders ready to export revo-    world; real GNP growth was a negative 2.3
lution. In 1976, white Rhodesian military     percent; inflation was 41 percent. In 1992,
officers opposed to the Marxist-leaning       after 16 years of fighting, General Peace
Mozambique government formed the              Accords were signed. The country’s first
Mozambican National Resistance                democratic, multiparty elections were held
(Renamo). Renamo guerrillas sought to         two years later. Why did the war end?
disrupt Mozambique’s economy in an ef-        Among the more important reasons was
fort to keep the new government from sup-     that foreign military support ended. As-
porting guerrillas who were trying to over-   sistance to Frelimo dropped sharply with
throw the white Rhodesian government.         the collapse of the Soviet Union, and

 10              Complex Humanitarian Emergencies and USAID’s Humanitarian Response
Renamo lost its external support when                 Rwanda remained in a state of ten-
apartheid ended in South Africa. In addi-        sion and instability throughout 1995 and
tion, foreign (mainly Italian) intermediar-      1996, as génocidaires came from neighbor-
ies helped to bring Frelimo and Renamo           ing countries, particularly from Idjwi Is-
to the negotiating table and to facilitate       land in Lake Kivu in the Democratic Re-
their reaching a settlement.                     public of the Congo (formerly Zaire). By
                                                 1999 the Rwandan Patriotic Front had se-
      Rwanda. Rwanda’s wholesale geno-           cured its borders and established security
cide of 1994 was a desperate attempt by          throughout most of the territory within
the government and Hutu extremists to            them. (Some killing continues, primarily
prevent the Tutsi-dominated Rwandan              in the northwest.) The bulk of the popula-
Patriotic Front from seizing power. Based        tion in exile or refugee camps has returned
in Uganda, the RPF had already tried to          to Rwanda. However, there remains a
topple President Juvénal Habyarimana             small group of Hutu-power extremists, de-
and his Hutu-dominated government in             termined to overthrow the current govern-
1990. That unsuccessful effort set the stage     ment and finish their work of genocide.
for a second attempt in 1993, which ended        Many of these génocidaires have found
with a cease-fire and later a peace agree-       sanctuary in the Democratic Republic of
ment, the Arusha Accords. However, it            the Congo, the Central African Republic,
soon became clear that key elements of the       and Angola.
accords, including access to land and po-
litical power sharing for the Tutsi living            Scholars have documented the exist-
in both Rwanda and Uganda, would not             ence of a culture of political impunity in
be honored. This reignited the military          Rwanda—including ethnically based mass
campaign, and by July 1994 the Rwandan           killing that is sanctioned or planned by
Patriotic Front had defeated the army of         government (Uvin 1998, Prunier 1995, Af-
the government of Rwanda.                        rica Rights 1998). Violence has been part
                                                 of a winner-take-all mentality that has
     In a 100-day period during April–July       dominated Rwanda’s governments during
1994, more than 800,000 people were mas-         the colonial and postcolonial periods. Both
sacred in a genocide historically un-            Hutu and Tutsi have used violence to ob-
matched in its intensity. The killing, or eth-   tain, and then maintain, absolute control
nic cleansing, eliminated close to three         over political and economic decisions.
fourths of the Tutsi population of Rwanda.
The international community (including                 This mentality has benefited a tiny
the United States) ignored, then acknowl-        elite, exclusively and handsomely, at the
edged, the genocide, but did little to pre-      expense of the poor majority. During 1990–
vent it. The U.S. secretary of state apolo-      93, the proportion of Rwanda’s population
gized for this failure to act in December        living in poverty increased from 40 per-
1997, as did the president in March 1998.        cent to 70 percent. During 1994–98, eco-

Country Context                                                                         11
nomic activity declined sharply. As a re-             community was feeding 1.3 million
sult, 75 percent of rural households (or 90           people—one of seven Haitians—each
                                                      day at 3,100 distribution points through-
percent of all households) currently live             out the country. It was also providing
below the poverty line, compared with 53              most of the country’s health services
percent five years earlier. Structural adjust-        (USAID 1995).
ment programs had not pulled Rwanda
out of its economic crisis.
                                                        Private voluntary organizations
     But this is not surprising, since most       implemented the humanitarian response
elements of these programs were not               in Haiti. The Adventist Development and
implemented. That prompted the World              Relief Agency operated 1,100 feeding cen-
Bank and the International Monetary Fund          ters in poor urban neighborhoods in Port-
to halt disbursements. According to the           au-Prince and in northern and central
World Bank, “rising poverty undoubtedly           Haiti. CARE worked in the northwest and
played some role in exacerbating social           Artibonite regions through 1,200 school
tensions leading up to the genocide”              feeding centers as well as hospitals, clin-
(World Bank 1998, i). But as Uvin states,         ics, and other distribution centers. Catho-
“structural adjustment did not cause these        lic Relief Services operated 800 feeding
[economic] problems; rather, it was irrel-        centers in the Port-au-Prince area and in
evant to their resolution” (Uvin 1998, 59).       the south and southwest. International
                                                  Lifeline implemented food aid programs
                                                  for two years, and the UN World Food
USAID’s Humanitarian                              Program, working through local and Eu-
Response                                          ropean nongovernmental organizations,
                                                  provided food to vulnerable children,
     Haiti. The need for humanitarian as-         pregnant and lactating mothers, and the
sistance was sharpened by the economic            destitute.
embargo imposed by the international
community in response to the 1991 coup                 USAID’s Office of Foreign Disaster As-
and the military’s subsequent political re-       sistance awarded seven grants to private
pression. USAID responded with an ex-             voluntary organizations totaling $5.4 mil-
panded program that included food aid,            lion. Catholic Relief Services, the princi-
potable water, and health and sanitation          pal grantee, distributed essential drugs,
assistance.                                       medical supplies, and agricultural inputs
                                                  (tools, seed, and fertilizer) and contributed
      USAID increased Haiti’s PL 480, Title II    to UNICEF’s oral rehydration therapy and
      program by 60 percent, from $15.4 mil-      measles immunization programs. The Of-
      lion in 1993 to $24.6 million in 1994. In
                                                  fice of Foreign Disaster Assistance also
      1995 it was increased by another 37
      percent to $33.6 million ( SAID/Haiti
                                                  supported efforts to purify drinking wa-
      1992–96). At its peak, the international    ter throughout Haiti, to purchase equip-

 12                  Complex Humanitarian Emergencies and USAID’s Humanitarian Response
ment for the Port-au-Prince municipal wa-        demobilization of the two armies; and
ter system, and to buy fuel needed to trans-     mine clearance. The number of internally
port emergency assistance to beneficiaries.      displaced emergency food aid beneficia-
The Pan American Development Founda-             ries was reduced from 1.5 million in 1993
tion played a pivotal role in the humani-        to only 600,000 in 1995. During 1996–97,
tarian response by implementing a $38            USAID assistance dropped to about $50 mil-
million jobs creation project.                   lion a year, and the mission resumed its
                                                 emphasis on development.
     Finally, the Office of Transition Initia-
tives provided $17.3 million in 1994–95 to            Rwanda. USAID provided food and
support Haiti’s transition to democratic         other types of emergency assistance to
governance. It funded the demobilization         Rwanda. According to USAID/Kigali, the
of the armed forces (not covered in the case     value of the assistance was almost $118
study) and over 1,900 microprojects de-          million in 1997 and over $56 million in
signed to bridge the gap between relief          1998. Most of it was food commodities.
and development.                                 Beans, cornmeal, and vegetable oil were
                                                 provided to genocide survivors, including
     Mozambique. In the mid-1980s, USAID         widows, orphans, unaccompanied minors,
assistance to Mozambique was under $50           and refugee-returnees. Most of the assis-
million a year, with emergency aid a small       tance was channeled through the World
proportion of total aid. During 1988–91,         Food Program, Catholic Relief Services,
as the civil war and humanitarian suffer-        the International Committee of the Red
ing escalated, USAID assistance doubled to       Cross, and World Vision Relief and Devel-
an average of $100 million a year. In 1989,      opment. The most vulnerable (children
an estimated one third of the population         under 5 and pregnant and lactating moth-
depended on food aid for 60 to 70 percent        ers) were reached through wet feeding
of their food needs. In 1992, in the final       programs in nutrition centers, inpatient
throes of the war, total U.S. aid doubled        feeding, and feeding programs in centers
again to $200 million annually. During           for unaccompanied children and orphans.
1993–95, after the 1992 peace accords, U.S.
assistance averaged $125 million a year, of
which emergency assistance was a large          The Office of Foreign Disaster Assis-
part.                                       tance simultaneously provided potable
                                            water, sanitation, and health services (as
     USAID’s relief-to-development pro- well as emergency food aid) along the
gram in Mozambique included several key route of returning refugees. Part of this
components: resettlement packages (food, assistance targeted orphans and unaccom-
seed, farming tools, household goods); re- panied minors. For example, a grant to the
building rural transport infrastructure; International Rescue Committee helped
support for elections and civic education; establish transit camps for such children.

Country Context                                                                        13
     OFDA also assisted over 50,000 vulner-   ported agricultural experts who identified
able farm families by providing seed,         appropriate seed stock, which then was
tools, and food rations for three months      multiplied.
through the World Food Program. The in-
tent was to jump-start agricultural produc-        Finally, OFDA provided $26 million to
tion. The Rwanda Emergency Seeds and          fund rapid-impact activities. These in-
Tools project, also OFDA funded, helped       cluded a shelter program to help meet the
90,000 families for one month following       needs of some of the 1.3 million returning
repatriation. Another agricultural project,   refugees and to preempt a potentially un-
implemented by Food for the Hungry, In-       stable security situation in the northwest.
ternational, distributed seed packages to     At the same time, the Office of Transition
25,000 vulnerable farmers. The project en-    Initiatives funded the Women in Transi-
couraged farmers to move onto rehabili-       tion program (reaching over 162,000
tated marshlands, trained farmers in new      women) and various activities to educate
practices, and rehabilitated rural infra-     local leadership and support local demo-
structure. The Seeds of Hope project sup-     cratic processes.

 14              Complex Humanitarian Emergencies and USAID’s Humanitarian Response
                             3         Humanitarian Effects

S                 and alleviating suffering
    AVING LIVES                                very much more than probably 60 per-
                                               cent of this being duly distributed, if not
     are key objectives of humanitarian as-
                                               to the intended beneficiaries, then at
sistance. When judged in terms of these        least to their representatives. (Apthorpe
criteria, most evaluations of humanitarian     1997, 97.)
assistance tell a “mission accomplished”
story. For example, Apthorpe’s review of       What about the complex humanitar-
six evaluations of humanitarian assistance ian emergencies in Haiti, Mozambique,
in Somalia, the Horn of Africa, Rwanda, and Rwanda?
Liberia–Sierra Leone, and Sudan con-
cludes as follows:                          Haiti
    . . . despite the horrendous and horren-
    dously difficult circumstances, what we        Emergency indicators. One indicator of
    read in these consultancies on the whole  the magnitude of Haiti’s complex emer-
    is that the humanitarian aid does actu-   gency is the national food supply. Haiti
    ally get through. Against all the odds,   historically has had a structural food defi-
    the job of getting it there is actually
                                              cit that makes the country dependent on
    done, if not always at the times sched-
    uled or as suitably composed as           imports. The deficit increased during the
    planned. (Apthorpe 1997, 101–2.)          crisis years, 1992–94. Domestic production
                                              was reported at 90 to 94 percent of nor-
     However, this message of overall suc- mal, while commercial imports decreased
cess is highly qualified. Apthorpe writes: by one third. Food aid increased by an
                                              average of 29 percent during 1993–95, but
     All accounts appear to find that, shall  this was insufficient to compensate for
     we say, making our own brave leap into   decreased domestic production and com-
     the quantitative blue yonder, normally
                                              mercial imports. As a result, Haiti’s food
     more than probably 90 percent of as-
     sistance has not failed to get through   deficit increased to an estimated 20 per-
     and be duly delivered with, say, usually cent of national food requirements, nearly
three times the deficit in a normal year        creased to 10.8 percent (1996) ( USAID/Haiti
(World Bank 1998a, WFP 1998).                   1992–96).

     Another indicator is food prices. Food           Humanitarian response. International
prices in Haiti rose sharply as commodi-        donors increased food aid deliveries to
ties became scarce. In Port-au-Prince, rice     Haiti by a third, primarily to address mal-
prices increased by 126 percent during          nutrition. The United States contributed an
1991–94; bean prices, by 167 percent; and       average of 68 percent of total food aid
corn prices, by 184 percent (USAID/Haiti        (World Bank 1998a). The United States also
1992–94). Although average food prices          initiated programs to provide short-term
decreased during 1994–96, they were still       employment, agricultural inputs, fuel, and
more than twice their 1991 levels.              medicine. Three major U.S. nongovern-
                                                mental organizations (CARE, Catholic Re-
      Malnutrition rates of children under      lief Services, and Adventist Development
5 is a third key indicator of the severity of   and Relief Agency, as previously noted)
a complex emergency. Most studies agree         implemented most of the emergency as-
that Haiti’s historically high malnutrition     sistance efforts funded by the United
rates increased in 1991–94 owing to the         States. According to the NGOs’ figures, ben-
combination of economic stress and a to-        eficiary levels nearly doubled during the
tal breakdown of the public health sector       emergency. In 1995, food aid was reach-
(IDB 1994, Ianotti 1997, World Bank 1998a).     ing 1.3 million direct beneficiaries, or 16
Data from OFDA’s monitoring reports also        percent of Haiti’s population (World Bank
show the trend of increased malnutrition        1998a).
but indicate the changes may not have
been significant. Nationally, nutritional          Efforts to target Haiti’s vulnerable
status (based on weight-for-age) of 50 per-   populations generally worked well, but
cent of Haitian children was normal in        the problems of looting and armed theft
1992. This figure increased to 52 percent     were always present. The Adventist De-
in 1993 but then declined to 49 percent in    velopment and Relief Agency, for example,
1994 and dropped still further to 47 per-     distributed dry rations for only six months
cent in 1995. It then rose to 51 percent in   in the low-income neighborhood of Cité
1996 (USAID/Haiti 1992–96).                   Jasmine in Port-au-Prince in 1992 because
                                              of violence. Catholic Relief Services also
     Of course, these national rates mask had difficulties in urban areas. Overall,
regional differences. For example, severe though, the estimated amount of leakage
malnutrition at the national level increased was 5 to 10 percent, regarded as normal.
from 3.3 percent (1992) to 3.9 percent (1994)
to 4.1 percent (1996). In the northwest,           The Jobs Creation project was imple-
though, it increased from 11.7 percent mented during 1993–96 primarily to off-
(1992) to 14.4 percent (1994), then de- set the embargo’s economic pressures. It

 16               Complex Humanitarian Emergencies and USAID’s Humanitarian Response
created almost half a million person-        $636 million during 1987–95. Of this, $529
months of short-term employment during       million was food aid, which accounted for
its 34 months of operation, of which 20      60 percent of total food aid provided dur-
percent was for women (Brown 1996).          ing this period.
Another USAID-funded activity supported
agricultural production and reduced               Indicators. The Mozambique evalua-
decapitalization of farm households. This    tion team found no valid quantitative data
project loaned funds to farmers for seed     to assess objectively the impact of U.S.
and fertilizer and sold them tools at half   emergency assistance. Although estimated
price. Approximately 13,000 farming          rates of malnutrition, mortality, and mor-
households and 47 farmers’ associations      bidity at the national level showed some
participated (Naval 1995). USAID’s Office    improvement, first in the late 1980s and
of Foreign Disaster Assistance funded the    then again in 1994 after the emergency, it
purchase of fuel needed to deliver emer-     was impossible to attribute these improve-
gency medical supplies, potable water,       ments to emergency assistance. NGOs re-
and food to more than 400 health centers     ported having little empirical basis for tar-
throughout the country.                      geting food aid because of the lack of sys-
                                             tematic information about food insecurity
     Most studies agree that the embargo     and nutritional status in rural areas. Bul-
seriously exacerbated Haiti’s historically   letins issued by Médecins Sans Frontières
high malnutrition rates. But they also con-  (Doctors Without Borders) contained
clude that emergency food relief alleviated  health and nutrition information, but they
that effect. And unlike the six evaluations  were produced only after 1992. Moreover,
reported by Apthorpe, there is quantita-     their data were based on small samples
tive evidence to support this conclusion.    and different methodologies and there-
The data from Haiti are approximate be-      fore, according to Médecins Sans
cause the sample populations and meth-       Frontières, “must be interpreted with cau-
ods of data collection were not standard-    tion.”
ized during the emergency. Nevertheless,
they provide a basis for making an in-           Notwithstanding the lack of accurate
formed judgment about the humanitarian quantitative data, there was consensus
effects of the emergency assistance. This among donors, relief workers, Mo-
was not the case in Mozambique.             zambican government officials, and
                                            Mozambicans who received food aid that
                                            the assistance aided people’s survival dur-
Mozambique                                  ing the emergency. All agreed that many
                                            more people would have suffered and
     The United States was Mozambique’s died without food aid, although it was
major donor during its complex humani- impossible to estimate the number of lives
tarian emergency, contributing a total of saved. Consistent with Apthorpe’s six-

Results: Humanitarian Effects                                                         17
evaluation review, “Mission accom- ents’ political affiliation and social status.
plished.”                                    Former internally displaced persons re-
                                             ported that everybody was hungry dur-
     Effectiveness. However, the impact of ing the war years, so those who received
emergency assistance varied greatly de- less food preyed on those who received
pending on (1) where the beneficiaries more, and when quantities were insuffi-
sought refuge (within Mozambique or in cient, people in some areas knifed open
nearby countries such as Malawi), (2) who sacks and fought for a share.
delivered the assistance (the relief unit of
the Mozambican government or NGOs),
                                                 Access to assistance by the internally
and (3) when people were uprooted and
                                             displaced also varied depending on
received assistance (before 1987, during
                                             whether they were in Frelimo- or Renamo-
1987–92, or during 1992–95).
                                             controlled areas. U.S. policy was to pro-
                                             vide emergency assistance only to govern-
     Mozambicans who fled to nearby ment- (Frelimo-) controlled areas—except
countries had a hazardous journey but for limited quantities provided to Renamo
generally received adequate food and territory through the International Com-
medical care once they reached the refu- mittee of the Red Cross. Regardless of its
gee camps. They were also relatively se- source, food aid attracted both Frelimo and
cure from Renamo or Frelimo harassment. Renamo soldiers. People from several vil-
By contrast, those who were internally dis- lages said they lost their food aid to
placed within Mozambique received less Frelimo by day and to Renamo by night.
adequate relief food that was supplied ir-
regularly. And they were often threatened        Finally, the effectiveness of emergency
and harmed by Renamo or Frelimo sol- assistance varied over time. Before 1987,
diers.                                       the war was disruptive, but few people
                                             had to flee their homes. Relief efforts were
     The internally displaced who re- relatively small and localized. The war
ceived aid directly from NGOs or the World intensified during 1987–92. People fled to
Food Program reportedly received more the relative safety of neighboring countries
adequate and regular supplies than those or the Beira corridor. (The corridor is an
aided by the Mozambican government. east–west swath across the country’s waist;
According to former internally displaced it was guarded by Zimbabwean troops
beneficiaries, the government did not pro- and thus served as a safe haven for
vide enough emergency food aid, and usu- Mozambican civilians.) This was an espe-
ally there was a two- to four-month time cially difficult period for internally dis-
lapse between distributions. They also re- placed persons. After the 1992 peace ac-
ported that the government’s distributions cords were signed, people began return-
were unfair, often influenced by the recipi- ing home. Resettlement packages (food,

 18              Complex Humanitarian Emergencies and USAID’s Humanitarian Response
tools, and services) were provided by        in camps in Ruhengeri. It found what Save
NGOs through 1995. This final stage of the   the Children Foundation/UK described as
emergency was the smoothest.                 “alarmingly high rates of malnutrition, in
                                             particular severe malnutrition among chil-
                                             dren 6–59 months.” Specifically, it found
Rwanda                                       8.0 percent of children with acute malnu-
                                             trition, 4.7 percent with edema, 40.6 per-
     In late 1996 and early 1997, some 1.3   cent underweight, and 59.9 percent with
million refugees were repatriated to         chronic malnutrition. These 1999 figures
Rwanda from the border camps, either         were substantially higher than those re-
voluntarily or by force. Massive starvation  flected in a 1996 National Nutrition Sur-
and human suffering would have occurred      vey. Thus, the situation in Ruhengeri, at
without substantial infusions of predomi-    least, was not getting any better—despite
nately U.S. emergency food aid. As in        emergency food assistance.
Mozambique, though, this is difficult to
quantify because data were not systemati-         Targeting. Targeting assistance to the
cally collected.                             intended beneficiaries in pre-1996 Rwanda
                                             was mixed. In the Bukavu area in south-
     Monitoring. Following repatriation of ern Zaire, government soldiers formed
refugees, the World Food Program carried separate camps from the very beginning.
out a six-month general-distribution food By contrast, in the Goma area the army,
program. This was intended to last only militia, and civilian refugees were all
until the harvest in June 1997 but was re- mixed together (forming “refugee–warrior
instated in November 1997 in five prefec- camps”), and the military and former gov-
tures. In those areas local food prices had ernment leaders controlled relief distribu-
increased by a factor of three, signaling a tion. In Tanzania, the military was not as
significant food shortage. Starvation and visible among the refugees, but the mili-
death were on a sharp incline, and emer- tia and former officials were. There was
gency food aid continued to be provided little security in these camps, and food and
until the next harvest in June 1998.         other relief supplies were diverted from
                                             the intended beneficiaries.
     Food insecurity was especially acute
in Ruhengeri Prefecture owing to politi-          It was painfully obvious that the per-
cal instability as well as poor harvests and petrators of human rights abuses and
high prices. As many as 573,000 people of genocide were fed and assisted in the
an estimated population of 869,000 were camps. Médecins Sans Frontières believed
displaced, living in camps and awaiting the only alternative was to leave the camps
resettlement. The Ministry of Health did and suspend most services. By contrast,
a nutritional survey in January 1999 based most other NGOs and the UN High Com-
on a sample of 900 children under 5 living missioner for Refugees decided to stay.

Results: Humanitarian Effects                                                       19
They recognized the humanitarian im-           the largest initiatives ever undertaken by
perative to protect and assist the vast        the international community. Although
population of refugees, even if that meant     Bosnia–Herzegovina was not included as
assisting people guilty of crimes against      one of the country case studies for this as-
humanity (Joint Evaluation 1996, study 2,      sessment, the UN High Commissioner for
58–9).                                         Refugees and the World Food Program
                                               had completed a joint evaluation of the
     Poor monitoring (which did not be- assistance program in 1998 covering the
gin in earnest until 1998) also contributed entire period from 1992 to June 1997
to ineffective targeting. U.S. government (UNHCR /WFP 1998, 1). The evaluation ex-
officials reported considerable double- amined the effectiveness of targeting in re-
counting of refugees by former military sponse to beneficiary needs, the impact of
and government leadership, particularly the emergency operation on the war itself,
in the cross-border camps. As a result, and the relevance of food aid in a period
many experts believe that more food aid of reconstruction—the same issues covered
was supplied both inside and outside in the Haiti, Mozambique, and Rwanda
Rwanda during this period than was nec- evaluations.
essary—and that more food aid was mis-
appropriated in Rwanda than is usual in             According to the joint evaluation, an
emergency situations.                          average of 2.6 million people were reached
                                               annually during 1992–96: 1.2 million in-
     Interahamwe (Hutu militia respon- ternally displaced persons and 1.4 million
sible for the genocide) and former soldiers “war affected” (people who had no means
of the Rwandan army diverted food from of support, although they were neither
women and children for their own pur- refugees nor displaced). By September
poses. There was also evidence that camp 1997, 1.14 million tons of food had been
rosters were sometimes not updated to provided at a cost of $710 million. An esti-
remove the names of those deceased, mated 80 percent of the population of
which resulted in the accusation that the Bosnia–Herzegovina had been beneficia-
UN High Commissioner for Refugees ries of food aid supplied by the World
“feeds dead people.” Although this is an- Food Program and the UN High Commis-
ecdotal, there is little doubt that there were sioner for Refugees at one time or another
abuses in the feeding program.                 (UNHCR /WFP 1998, 2).

Bosnia–Herzegovina                           The evaluation concluded that there
                                         was no widespread hunger or malnutri-
     The humanitarian assistance opera- tion in Bosnia–Herzegovina. However,
tion in the former Yugoslavia was one of security-related problems hindered access

 20               Complex Humanitarian Emergencies and USAID’s Humanitarian Response
and distribution to isolated communities              A variation of this mechanism was to
and cities under siege. Air transport was       use the proceeds from the sale of the com-
used when access by land was denied, as         modities to establish wage-generation pro-
in Gorazde, Sarajevo, Srebenica, and Zepa.      grams rather than buy back the commodi-
The Sarajevo airlift was the longest run-       ties. This gave people cash, creating a mar-
ning humanitarian air bridge in history,        ket that the merchants then supplied. But
lasting from 3 July 1992 until 9 January        commercial channels are not always the
1996. These operations were generally suc-      solution. In Mozambique, most private
cessful. In fact, it was only in Bihac dur-     transporters refused to transport food aid
ing late 1994 and throughout 1995 that air-     because of bandits and land mines and
drops did not succeed in averting hunger        also because many that did had had their
(UNHCR /WFP 1998, 6).                           trucks stolen.

                                                     Military involvement in complex
Targeting                                       emergencies—both as protector and pro-
                                                vider of commodities—has been a mixed
     Efforts to target humanitarian aid to      blessing. The tremendous costs of military
intended beneficiaries often run into ma-       operations are generally disproportionate
jor problems. Several ways to improve tar-      to the value of the commodities protected.
geting and the overall effectiveness of hu-     Military humanitarianism can also get
manitarian assistance programs involve          wrapped up with geopolitics and foreign
alternative distribution channels, com-         policy objectives. In Liberia, for example,
modity selection, and planning.                 the Nigerian-dominated regional military
                                                force sent to Liberia for peacekeeping pur-
     Commercial channels may offer an al-       poses found itself in conflict with the larg-
ternative to other types of distribution. For   est rebel force in the country (Prendergast
example, the World Food Program con-            and Scott 1996).
tracted with Somali merchants in
Mombasa, Kenya, to transport commodi-                Carefully selecting commodities for
ties to targeted sites in Somalia. This in-     emergency assistance can reduce looting
volved selling the commodities to the mer-      and improve targeting. For example, rice
chants and then buying them back with a         and other high-value commodities are
10 percent profit margin. WFP paid a 10         typically much more attractive to looters
percent markup to avoid having 60 per-          than sorghum, maize, or blended foods.
cent of the food looted. This was because       But this varies by region: substantial quan-
bags printed GIFT OF THE USA were more          tities of maize were looted in Somalia.
likely to be looted than commodities            Similarly, looters are rarely interested in
stored and transported by businessmen.          cooked food distributed in numerous

Results: Humanitarian Effects                                                            21
kitchens that are widely dispersed. In-           In planning emergency assistance
stead, they are generally interested in com- programs, experts need to be sensible and
modities for which there is market de- beware of over-complexity. It is also
mand and which they can turn into cash. important to address capacity-building
                                              questions early on. In anticipation of re-
     Diversifying entry points for emer- habilitating a collapsed health service, for
gency supplies can help guard against example, capacity building might take the
empowering a particular authority. In form of training medical personnel. Ac-
Liberia, for example, all commodities cording to Anderson (1996), capacity
came into Monrovia rather than across the building should be a central part of any
border upcountry. By contrast, substantial emergency response.
food assistance was delivered to affected
sites in Ethiopia (Eritrea and Tigray) via        Finally, Prendergast and Scott (1996)
the Sudanese border.                          point out it is important to plan up front
                                              for monitoring and evaluation: “A com-
     There is an important distinction be- mitment to adequate, independent, and
tween distributing emergency commodi- continuous monitoring and evaluation of
ties to affected areas and to affected programs may reduce aid’s contribution
populations within areas. Food aid was dis- to conflict.” In Rwanda, as a result of moni-
tributed to Rwandan refugees in camps toring, diversion of food was reduced from
through prefectures, communes, and fi- 120 tons per month to 5 tons per month
nally cells; but it was not targeted to indi- between July 1993 and January 1994. “It’s
vidual families. As a result, the emergency monotonous, boring, but critical in cutting
aid perpetuated the authority of the mili- down mismanagement” (Prendergast and
tary and political leadership that had Scott 1996).
planned the genocide.

      Humanitarian aid is more easily          Efficiency
diverted when population figures are in-
flated. For example, to achieve purely po-          In addition to targeting, another mea-
litical and economic objectives, warring       sure of the effectiveness of humanitarian
factions in Liberia and Rwanda overesti-       assistance is efficiency. Borton and Macrea
mated the need for food. But underestimat-     (1997) synthesized the results of a broadly
ing need may lead to violent competition       representative sample of 28 evaluations,
for food. In assessing need, it is important   mostly of complex emergencies, under-
to understand people’s coping strategies       taken since 1991 by bilateral donors, UN
and their desire to preserve their liveli-     agencies, and the European Community
hoods. This may be as important as en-         Humanitarian Office (1–2, 12). They exam-
suring short-term hunger alleviation           ined humanitarian assistance in terms of
(Borton and Macrea 1997, 27).                  cost-effectiveness, which, unlike cost–ben-

 22               Complex Humanitarian Emergencies and USAID’s Humanitarian Response
efit analysis, does not involve the valua-    in Guinea and Côte d’Ivoire) improved
tion of lives in economic terms. What they    cost-effectiveness (Apthorpe 1996). As
found is instructive: very few of these       noted above, using lower value commodi-
studies even considered the issue of cost-    ties also reduced the likelihood of diver-
effectiveness—partly for methodological       sion as well as the disincentive effect on
reasons and partly because of reservations    local production (Borton and Macrea
about whether it should be a criterion for    1997, 24).
providing humanitarian aid. According to
Kleist (1994, 301–02), the reality is that
monetary costs are less important as an       Conclusion
evaluative criterion than the number of
lives saved and the security of the person-         The international community is gen-
nel delivering the humanitarian assistance.   erally unable to assess with any degree of
                                              certainty the number of lives saved
     Transportation (rail, road, air) is a    through humanitarian assistance provided
major cost of emergency operations that       in a complex emergency. The case of Haiti,
varies enormously. For example, commer-       where data were available, is the excep-
cial air transport within the Great Lakes     tion; the case of Mozambique, which
region was approximately 4 to 5 times         lacked data, is the norm. Only 1 of the 28
more expensive than road transport, and       evaluations reviewed by Borton and
10 to 20 times more expensive than rail       Macrea (1997) attempted to estimate the
transport (Joint Evaluation 1996, study 3).   number of lives actually saved by inter-
Therefore, on efficiency grounds it is bet-   national assistance interventions. This was
ter to transport food to conflict areas by    an analysis of the 1990–94 response to the
road or rail rather than by plane. On the     crisis in Somalia (Hansch 1994). It found
other hand, using ground rather than air      that 330,000 Somalis were at imminent risk
transportation could involve negotiating      of death in 1992 and 1993. An estimated
with rebel groups, thereby granting them      110,000 of these were sustained (that is,
a degree of legitimacy they otherwise         their deaths were averted) by health, food,
would not enjoy (Hallam 1998, 21).            and other interventions. At least 70 per-
                                              cent (154,000) of the famine-related deaths
      Cost also varied according to distri-   that did occur in 1992 probably could have
bution channel and type of commodity.         been prevented had primary health strat-
Military channels were estimated to be        egies been implemented earlier and more
four to eight times more expensive than       widely (Hansch 1994, cited in Borton and
civilian channels (Borton and Macrea 1997,    Macrea 1997, 2, 25).
2, 23). Moreover, replacing high-value rice
with alternative less expensive cereals in    However, another analysis of the ef-
coastal West Africa (bulgur wheat in Si- fectiveness of humanitarian assistance in
erra Leone and Liberia, and maize meal Somalia reports results with far less quan-

Results: Humanitarian Effects                                                        23
titative precision. It concludes that “sig-         nizations to meet urgent humanitarian
nificant numbers of lives were saved, se-           needs.
vere malnutrition and vulnerability to in-
fectious disease declined, the suffering of              This raises a question about the de-
the displaced was eased, and refugee                gree of planning and data collection that
movement was slowed—though these effects            is both feasible and desirable when an
are all difficult to quantify” (Kleist 1994, 305;   immediate humanitarian response is
emphasis added). According to Kleist, the           needed to save lives. USAID’s Bureau for
inability to measure results with any de-           Humanitarian Response has determined
gree of certainty reflects the fact that sav-       that one of its strategic objectives is to en-
ing lives is of the utmost priority. Requir-        sure that “critical food needs of targeted
ing answers to detailed questions about             groups are met.” Two indicators are speci-
humanitarian assistance provided during             fied to determine if the objective has been
a complex emergency could cause delay               achieved: (1) the percentage of target
and cost human lives. Therefore, little up-         populations reached by food aid programs
front planning or data collection is done.          and (2) the impact of the assistance on the
In Somalia, for example, many proposals             nutritional status of beneficiaries (USAID
from NGOs lacked such basic information             1998b, 4). However, data must be collected
as who the target group was and where it            for these two indicators, and data collec-
was located (Kleist 1994, 294–95). Notwith-         tion in the throes of a complex humanitar-
standing inadequate information, donors             ian emergency takes time when time is of
typically allocate resources to relief orga-        the essence.

 24                 Complex Humanitarian Emergencies and USAID’s Humanitarian Response
                                   Political Effects and
                          4        Effects on Hostilities

H     UMANITARIANS TRADITIONALLY                  Each of the six evaluations reviewed
       have tried to remain impartial and    by Apthorpe asked whether humanitarian
                                             aid had had the perverse effect of prolong-
       thus apolitical. There are at least two
                                             ing the war or contributing to the war
good reasons for this. The first is pragmatic:
providing relief is often facilitated when   economy; that is, whether such aid in ef-
indigenous political actors perceive hu-     fect feeds conflict as well as its victims.
manitarian agents to be without political,   Each concluded that although food and
religious, cultural, or other agendas. Im-   other humanitarian assistance are not
partiality helps humanitarians gain access   meant to feed conflict, they often do
to victims. The second reason to remain      (Apthorpe 1997, 95). In short, humanitar-
impartial is principled: each society has theian assistance can have adverse political
exclusive prerogative and responsibility to  effects. Was this the case in Haiti,
shape its own destiny. Outsiders should      Mozambique, or Rwanda?
not interfere except in nondisruptive ways
to save lives. By helping everyone and re- Haiti
fusing to take sides, humanitarians place
themselves above the fray (Pasic and Weiss         The large quantities of aid-financed
1997, 198–99).                                 food injected into resource-starved Haiti
                                               were conspicuous and highly valued—ei-
     Nevertheless, the increasingly obvi- ther for direct consumption or as a politi-
ous reality is that humanitarian relief and cal tool for those who controlled their dis-
its consequences are inevitably political, tribution. Control over access to food aid
often in ways that are not self-evident. That became a new source of tension and
is because aid does not just keep people power, and violent elements—local gangs
alive in a political vacuum but also affects or groups connected to political factions—
the local power structure and changes the hijacked food supplies. Fighting some-
environment in which it is given. Aid is times erupted among beneficiaries when
rarely neutral.                                food was dropped off in urban neighbor-
hoods for distribution by volunteers. Some      its victims, ironically could have caused
municipal mayors used their access to food      more humanitarian distress than it allevi-
aid to favor supporters of one political fac-   ated. The point, of course, is speculative,
tion or another or to promote their per-        and one cannot conclude with any degree
sonal aggrandizement.                           of certainty that humanitarian assistance
                                                prolonged the conflict in Haiti.
     To their credit, the NGOs limited leak-
age and diversion (estimated at less than       Mozambique
10 percent) through regular monitoring,
convoy protection, and timely adjustment             The Reagan and Bush administrations
of their stocking and distribution methods.     debated whether to support the Frelimo
They also stopped distributions in some         government or the Renamo rebels when
neighborhoods in Port-au-Prince, such as        Mozambique’s civil war broke out in the
Cité Jasmin. Violence, political exploita-      mid-1980s. Many American conservatives
tion, and local tensions consequently were      viewed the war as an ideological battle
reduced to manageable if not minor pro-         over communism and believed that the
portions.                                       United States should therefore support
                                                Renamo. This view was buttressed by the
     Did the emergency assistance help the      fact that the Frelimo government had in-
de facto regime in Haiti withstand diplo-       stalled a socialist system and was receiv-
matic pressures and the effects of the eco-     ing support from its socialist allies. Oth-
nomic embargo—before the international          ers, however, believed the United States
community finally resorted to military          should assist the Frelimo government in
force? Many Haitians believe this is the        recognition of its support of the antiapart-
case. In their view, exempting humanitar-       heid movement in South Africa. In the fi-
ian aid from the embargo worked at cross-       nal analysis, the United States provided
purposes with the policy of economic            limited humanitarian assistance to
isolation. By reducing food distress the        Mozambique’s socialist government.
emergency assistance dampened public
pressure that might otherwise have risen       By the late 1980s, the Frelimo govern-
                                          ment had abandoned most of its socialist
to uncontrollable levels against the regime.
Thus, food aid may have permitted a de-   ideology and initiated a program of mar-
                                          ket-based economic reforms supported by
lay in the intervention by external forces,
intervention that finally proved unavoid- USAID and the World Bank. In response,
                                          the United States greatly expanded its hu-
able in order to eject the Haitian military
and return President Aristide to power.   manitarian assistance, much of it in the
                                          form of food aid. This U.S. assistance was
     This view implies that humanitarian provided on a government-to-government
aid, by extending the duration and extent basis, which meant it went only to inter-
of the emergency and the consequences for nally displaced persons in Frelimo-con-

 26               Complex Humanitarian Emergencies and USAID’s Humanitarian Response
trolled territories. The only U.S. assistance     transplantation of a well-organized politi-
provided to Renamo populations was sup-           cal, social, and security structure. The
plied indirectly through the International        Hutu extremists (ex-Rwandan army regu-
Committee of the Red Cross.                       lars, former government officials, and al-
                                                  lied militia) planned to use the refugee
     What was the impact of emergency             camps as a staging area for their eventual
food aid on the length of the civil war in        return to political power through
Mozambique? Expatriates and Mo-                   Rwanda’s northwest. They assumed, cor-
zambicans alike consistently reported that        rectly, that the international humanitarian
foreign political and military support for        relief agencies (and the national govern-
Frelimo and Renamo—rather than hu-                ment and regional authorities of Zaire)
manitarian assistance—was the primary             would not separate them from bonafide
resource that fueled Mozambique’s 16-             refugees. This meant they could consoli-
year civil war. U.S. humanitarian assis-          date both military and political control
tance had relatively little influence on the      over most of the camp population. In De-
course of the war when compared with the          cember 1994 a new government of
military assistance provided by the Soviet        Rwanda in exile was declared, and incur-
Union (to Frelimo) and by South Africa (to        sions from the camps into Rwanda began.
                                                 Mixing bonafide refugees with those
     Nevertheless, both Frelimo and who were probably guilty of genocide and
Renamo soldiers tried to steal food aid by other high crimes was seen by the UN
intimidating PVO workers and hijacking High Commissioner for Refugees, the In-
trucks. Food aid distributions in rural vil- ternational Committee of the Red Cross,
lages were a magnet for looting by both and the international humanitarian com-
militaries. Thus, food aid helped support munity as a conundrum, difficult to re-
the military forces to some extent, but the solve in the context of maintaining neu-
effect was relatively small. As in Haiti, trality and without military intervention.
emergency food aid in Mozambique was As a result, UNHCR did nothing. It was im-
sometimes politicized. Politicians at both portant to maintain stability in an inher-
the national and local levels reportedly ently unstable situation, and separating
used their influence over food aid distri- the refugees from the Hutu extremists was
butions to favor particular factions and considered risky. Moreover, UNHCR did
reinforce their political power.             not believe the new government of
                                             Rwanda would welcome the refugees back
Rwanda                                       home, certainly not those involved in the
                                             genocide. The Hutu refugees, themselves,
     The massive influx of Rwandans into understandably feared retribution (“re-
refugee camps in former Zaire was not verse genocide”) if they returned, whether
only a movement of people but also a or not they were guilty of crimes and

Results: Political Effects and Effects on Hostilities                                    27
atrocities committed in Rwanda. These destabilize and overthrow the present gov-
fears and concerns helped create a dead- ernment of Rwanda. Their objective is to
lock that lasted over two years.              complete the unfinished work of genocide,
                                              using a campaign of propaganda and ter-
      The inability of the international com- ror to destroy the political and economic
munity to resolve the deadlock gave the structures of the northwest (and beyond)
Hutu extremists in the refugee camps a and to gain support of the local Hutu
false sense of enhanced legitimacy. They population. In 1997, 30,000 to 40,000 sol-
used this opportunity to regroup, rearm, diers began arriving in the northwest,
and revitalize themselves with food ra- while several thousand remained in Zaire
tions intended for refugees—all in the rela- to maintain the camps as a base of opera-
tive safety of the camps. As indicated ear- tions. In 1998, several commune offices in
lier, once the camps became militarized, the northwest were looted and burned,
targeting became problematic. Beneficiary and the officials were murdered or terror-
figures were significantly inflated. Food ized. By March 1999, many services had
aid was diverted to and consumed by the ground to a halt. Water sources had been
ex–Rwandan army regulars and inter- destroyed and health problems multiplied.
ahamwe militia resident in the camps. The government of Rwanda requested
That had the unintended effect of prolong- massive food assistance from the World
ing the conflict. According to the U.S. am- Food Program. In response, USAID is pro-
bassador to the United Nations, Bill viding both development assistance and
Richardson, “the failure of the interna- humanitarian assistance to the northwest.
tional community to respond adequately
to both genocide and the subsequent mix-
ing of genocidal killers with the legitimate Bosnia–Herzegovina
refugee population in the former eastern
Zaire only served to prolong the crisis”           The conflict in former Yugoslavia also
(October 1996). It was not until 1996 that raises concerns about emergency aid en-
USAID became sufficiently concerned and abling protagonists to prolong the conflict.
ceased providing food aid to the World As early as 1994, some evidence, though
Food Program for use in the camps. The inconclusive, suggested that aid supported
idea of providing humanitarian aid to the troops, thereby releasing the authorities
planners and implementers of genocide from responsibilities they might have had
was seen as inconsistent with the stated to civilians (Prendergast and Scott 1996,
objectives of humanitarian aid.               11, citing Minear 1994a). This earlier evi-
                                              dence was examined by the recent UN
      The crisis is not yet over. Rwandan High Commissioner for Refugees/World
ex-political and ex-military leadership is Food Program joint evaluation of the
using former Zaire as a staging ground to Bosnia experience (UNHCR /WFP 1998).

 28              Complex Humanitarian Emergencies and USAID’s Humanitarian Response
     This joint evaluation offers two pos-        cording to UNHCR /WFP (1998, 7), there is
sible arguments supporting the thesis that        no convincing evidence that the increased
humanitarian aid to Bosnia helped pro-            suffering that would likely have occurred
long the conflict (UNHCR/WFP 1998, 3).            in the absence of humanitarian support
Both arguments continue to be a source of         would have been justified by an increased
controversy. The first suggests that by giv-      prospect for a swifter and satisfactory out-
ing generous support in the form of food          come of the conflict.
aid, donors were able to defend them-
selves against the charge of inaction. But
doing so in effect postponed the military              The provision of aid did have some
intervention that ultimately was needed           unavoidable negative political effects.
to end the conflict. This reasoning paral-        First, since the authorities on the ground
lels that of the Haiti evaluation, and as         controlled distribution of the assistance,
noted there, is purely speculative.               bargaining with them and agreeing to use
                                                  the channels they controlled inevitably
     The second argument suggests that            reinforced their authority. Anderson (1996,
humanitarian assistance prolonged the             3–4, 16–17) suggests that this is not unusual
war in Bosnia because it was diverted to          and that aid agencies often must negoti-
the combatants and thus supported their           ate with army leaders to gain access to
military efforts. Even if humanitarian sup-       civilian populations or to hire armed
plies were not diverted to combatants, the        guards to protect the goods they bring.
aid still would have allowed resources oth-       Second, in some cases military authorities
erwise needed to sustain the noncomba-            levied food taxes to allow convoys to pass
tant population to be used instead to sup-        and seized food when they were not
port the war effort. (Of course, combatant        paid, sometimes at gunpoint ( UNHCR /WFP
organizations in some complex emergen-            1998, 7).
cies have shown little concern over the con-
dition of “their” noncombatants.)
                                                  Policy Implications
     It is possible that without food aid the
consequent civilian suffering might have               This assessment of the political effects
hastened the cessation of hostilities in          of humanitarian assistance raises key ques-
Bosnia. However, while the conflict might         tions: Should aid be given if some of it is
have been shorter, the suffering would            being diverted to armed participants in the
likely have been greater. Moreover,               conflict? Should aid be distributed through
UNHCR / WFP suggests that the outcome             local structures if these are considered
would likely have been unsatisfactory: a          predatory or biased? USAID clearly would
world without Bosnia, with the country            not operate a development assistance pro-
instead divided among its neighbors. Ac-          gram if security deteriorated to the degree

Results: Political Effects and Effects on Hostilities                                      29
it did in Somalia, for example. Yet the        because it enables them to feed the popu-
United States can nearly always be             lations they seek to control. The same is
counted on to provide emergency assis-         true of the government and the agencies
tance, even under such volatile conditions.    they mandate to distribute relief. As a re-
                                               sult, organizations (often NGOs) that at-
     Prendergast and Scott (1996) are sen-     tempt to provide humanitarian aid neu-
sitive to the possibility that humanitarian    trally may support forces that carry out
aid designed to relieve suffering and pro-     violence against civilian populations, as in
mote peace often, inadvertently, fuels, sus-   Mozambique. But the alternative may be
tains, or exacerbates such conflicts by mak-   no better: to subject civilian populations
ing more resources available to warring        to the double punishment of violence and
parties. They like others recognize that       hunger.
humanitarian aid may be given without a
political agenda, but it rarely escapes hav-       2. Manipulating population movements.
ing political consequences. Moreover, it     Warring factions have used civilians as
can be deliberately manipulated to serve     shields or “vehicles” to obtain food and
as an instrument of war by providing a       other types of humanitarian aid. They po-
means for sustaining the conflict. This can  sition civilians near airstrips to enhance the
occur by (1) manipulating access to the aid, ability of their troops or militia to remain
(2) manipulating population movements,       in areas they otherwise would abandon for
and (3) diverting or looting the aid. Ex-    lack of supplies or difficulty in defending.
amples:                                      For example, aid supplied to the refugee
                                             camps in Zaire helped maintain the former
     1. Manipulating access. Warring parties Rwandan government’s control over a
often manipulate humanitarian aid to en- population that otherwise might have dis-
hance their power over civilian popula- persed or returned home.
tions or weaken their opponents by deny-
ing them food. Throughout the 1980s the            3. Diverting or looting aid. Warring fac-
Ethiopian and Sudanese governments lim- tions tax, steal, or divert humanitarian as-
ited the amount of aid going to rebel-held sistance—especially food and drugs, given
areas. Warring factions also have manipu- their easy monetization—for their own
lated access to aid in Bosnia, Liberia, consumption, for barter, or for sale. This
Mozambique, Rwanda, and Somalia.             is a principal means to buy arms. Liberian
                                             rebels looted relief supplies and stole re-
     Relief agencies generally abide by the sources, especially vehicles and fuel to use
principles of neutrality and impartiality. for hit-and-run campaigns. In Somalia,
They don’t take sides, and instead give food distribution operations were looted
both sides equal access to relief aid. But so often that as little as 12 percent of inter-
providing aid in rebel areas necessarily national food aid destined for refugees in
helps legitimize the rebel organizations, 1986 reached the intended recipients ( GAO

 30               Complex Humanitarian Emergencies and USAID’s Humanitarian Response
1986). Much of the remainder was taken            terms: “The people who don’t care and are
by the Somali army and associated mili-           not affected are the rulers.”
tia. Agencies had to negotiate for their own
security because there was no national se-        Food and medical aid are particularly
curity force, and this led to diversion of   valuable to combatants. But aid agencies
goods, especially food (Kleist 1994, 298).   rarely withhold such assistance despite
Similar occurrences took place during the    human rights abuses and looting, and
famine in northern Ethiopia in the 1980s     most will not completely withdraw from
and the wars in Mozambique and south-        an area unless the emergency is completely
ern Sudan (de Waal 1993).                    over or the security situation is untenable.
                                             In Liberia during 1990–93, aid agencies
     Given this negative and widespread were subjected to harassment and robbery
experience, under what conditions, if any, by warring factions. Diversion rates were
should humanitarian aid be halted? close to 50 percent, according to some re-
Prendergast and Scott suggest three fac- ports, and debate raged within the aid
tors to consider in deciding whether to community about how much aid was too
cease providing humanitarian assistance: much. All agencies eventually withdrew
lack of progress in peace negotiations; sup- operations until security improved. How-
port of undesirable political factions and ever, it remains unclear whether the with-
human rights abusers; and danger to aid drawal contributed to the 1995 peace
personnel.                                   agreement. And even if it did, was it at
                                             the cost of more human suffering? There
     During the civil war in Ethiopia in the is simply no way to measure the effect of
late 1980s, Lutheran World Relief urged stopping the aid.
agencies seriously to consider withhold-
ing aid if peace efforts failed (Prendergast      De Waal points out that war has be-
and Scott 1996, 44). Should we pay end- come synonymous with famine in much
less millions for humanitarian aid, they of Africa. In fact, war is often designed to
asked, when this may only exacerbate and create famine. War, according to de Waal,
prolong the conflict? Perhaps we should has received far less attention than it de-
disavow the principle that food should not serves—compared with other contributory
be used as a political weapon and instead causes of famine such as drought, environ-
use it to force peace negotiations. How- mental degradation, and inappropriate de-
ever, others have argued that political de- velopment strategies. This brings into
cisions should be separate from the basic question the tendency for donor govern-
human right to humanitarian aid, and that ments to fund humanitarian assistance but
relief should not be used as a political not to address the underlying causes of
weapon. The rationale underlying the lat- war (Hallam 1998, 5). Humanitarian agen-
ter point of view is summarized in cies are sometimes needed less than po-
Prendergast and Scott (1996, 44) in these litical or military actors. In Rwanda in

Results: Political Effects and Effects on Hostilities                                  31
1994, for example, a well-armed UN                 provided in the context of conflict (Hallam
peacekeeping force may have been able to           1998, 12–13). The political effects of
prevent or mitigate the genocide. The              humanitarian assistance in Haiti, Mo-
multitude of NGOs responding to humani-            zambique, and Rwanda were mixed.
tarian needs once the genocide was over            According to most accounts, the assistance
was much less effective. The issue of con-         clearly prolonged the conflict in Rwanda.
flict prevention, as distinct from cure, is        In Mozambique, external military assis-
briefly introduced in annex C.                     tance, rather than humanitarian assistance,
                                                   fueled the civil war for more than a de-
      Though war may be a principal cause          cade. In Haiti, evidence of the political ef-
of famine (as de Waal suggests), Sen (1993)        fects of the assistance is inconclusive. In
provides a compelling argument that de-            Bosnia, airlifting food to Sarajevo probably
mocracy and a free press are great forces          prolonged the war, but this does not mean
in preventing famine. Sen points out that          that the increased suffering that would
a government cannot ignore famine con-             have occurred by withholding food aid
ditions if (1) it has to face reelection, (2) it   would have been justified by the possibil-
cannot censor the terrible facts of starva-        ity of a shorter war.
tion, disease, and death that accompany
famines, and (3) it has to face criticism
from opposition parties and newspapers.           Perhaps what is most important to
“It is not surprising that even though fam-  keep in mind is the underlying principle
ines have happened in colonial economies     that, at the very minimum, aid that is in-
and in modern authoritarian states, never    tended to help victims in war settings
has a famine occurred in a democratic        should not cause additional harm (Ander-
country with a relatively free press” (Sen   son 1996, 6). The challenge, therefore, is to
1993, 88).                                   specify, on a case-by-case basis, clear ob-
                                             jectives and to monitor closely the extent
Conclusion                                   to which the humanitarian assistance is
                                             achieving those objectives. Allowing flex-
      While no aid is apolitical, humani- ible implementation and encouraging ef-
tarian assistance provided during complex fective communication and coordination
emergencies can result in substantial and among other involved authorities (diplo-
unpredictable political effects, since it is matic, military) is equally important.

 32                Complex Humanitarian Emergencies and USAID’s Humanitarian Response
                            5        Economic Effects

T    HE RELATIONSHIP between short-term
      relief and long-term development has
been viewed as a continuum in which re-
                                                       None of the six evaluations reviewed
                                                 by Apthorpe (1997, 92) concludes that re-
                                                 lief is poorly done if it is not specifically
lief operations, in response to a humani-        forwardly linked to development. Two of
tarian crisis, are followed by rehabilitation    the six explicitly reject the linear linkage
and then development activities (USAID           as too simplistic and do not see it as the
1998a, 18–19). More recent literature ques-      best guide to what is needed in relief aid.
tions the utility of this concept of a relief-   A third notes that the continuum concept
to-development continuum. According to           derives from natural disasters and there-
the Department of State’s Bureau for Popu-       fore has only limited application to com-
lation, Refugees, and Migration, relief and      plex emergencies, which are often politi-
development assistance have significantly        cal in nature. Apthorpe concludes that
different aims, and implementers of each         much of the literature on the relief–devel-
type of assistance should address the ap-        opment continuum has little operational
propriate aims. Messer (1998, 15) notes that     value.
“although relief officials try to make relief
function as development assistance, the               Nevertheless, emergency programs
‘relief-to-development continuum’ they           can have an important effect on shaping
talk about appears to be more wishful            the pattern and direction of subsequent
thinking than fact. The bulk of emergency        economic development. Societies recover-
food assistance is devoted to meeting ba-        ing from disastrous conflict are in the pro-
sic human welfare needs.” Miller (1997, 15)      cess of remaking themselves, and the eco-
believes too much emphasis and attention         nomic opportunities that emerge from this
have been given to the concept of relief-        process can be influenced by how emer-
to-development, given the relatively brief       gency assistance is designed and deliv-
period when relief assistance overlaps           ered. What was the experience in Haiti,
with longer term development assistance.         Mozambique, and Rwanda?
Haiti                                          ture could have been built, that would
                                               have required purchasing materials (rather
     In Haiti, USAID emergency assistance      than hiring labor), and the primary objec-
supported economic development in two          tive of generating employment may have
main ways: employment generation and           been compromised.
agricultural production. In addition,
USAID’s Office of Transition Initiatives pro-      Agricultural production. Another emer-
vided small grants to various organiza- gency activity, funded by USAID’s Office
tions to fund numerous microprojects.         of Foreign Disaster Assistance and admin-
                                              istered by Catholic Relief Services, was de-
     Employment generation. The 1991–94 signed to support agricultural production
crisis worsened rural poverty in a coun- and reduce decapitalization of Haitian
try where poverty levels already were farm households. The project loaned agri-
among the highest in the world. When the cultural inputs (seed and fertilizer) to
assembly plants in Port-au-Prince closed farmers who were otherwise being forced
because of the U.S.–led economic em- to sell their productive assets to buy food;
bargo, 400,000 urban poor returned to the it also sold them tools at half price. Some
countryside. That increased pressure on 13,000 farming households (less than 10
rural households’ scarce resources. USAID’s percent of all farmers) and 47 farmer as-
Jobs Creation project was implemented sociations participated (Naval 1995). By
during 1993–96 primarily to generate em- supplying inputs needed for food produc-
ployment needed to maintain household tion, the program provided emergency
incomes and thereby offset the embargo’s assistance in a way that helped maintain
economic pressures. Its secondary objec- beneficiaries’ incomes and livelihoods and
tive was to rehabilitate productive infra- at the same time reduced their dependence
structure.                                    on short-term relief.

     The project created half a million per-         Microprojects. The Office of Transition
son-months of short-term employment            Initiatives typically funds programs de-
during 34 months. More than 120 indi-          signed to bridge the gap between short-
vidual projects were carried out, resulting    term relief (often managed by the Office
in the repair of 1,000 miles of roads, 2,000   of Foreign Disaster Assistance) and long-
miles of irrigation canals, and 4,500 miles    term development (typically managed by
of soil conservation barriers (PADF). The      resident USAID missions). These programs
project achieved its main objective—em-        often inject cash into an economy to gen-
ployment creation. However, maintenance        erate employment or provide commodi-
and long-term sustainability of the infra-     ties to meet peoples’ most pressing needs
structure was not a project objective, and     quickly. They are meant to create precon-
over time the infrastructure has deterio-      ditions for development and at the same
rated. Although more durable infrastruc-       time facilitate the phaseout of emergency

 34               Complex Humanitarian Emergencies and USAID’s Humanitarian Response
assistance. This is important, because suc-     and economic chaos in which the goal of
cessful rehabilitation is more difficult the    long-term sustainable development was
longer relief is provided (Kleist 1994,         eclipsed by the immediate short-term need
300, 307).                                      for relief. Haitians themselves were con-
                                                cerned primarily with physical security
     The Office of Transition Initiatives       and survival rather than development.
funded an $11 million program in Haiti          Therefore, emergency assistance programs
that supported more than 1,900 micro-           generally were designed with only inci-
projects over a 27-month period during          dental links to economic development, and
1994–96. (This means that, on average,          they had minimal developmental impact.
more than two microprojects were initi-         Although it is desirable to incorporate
ated each day during this period.) They         long-term development objectives when
ranged from rehabilitation and construc-        designing short-term emergency re-
tion of community schools, roads, markets,      sponses, the Haiti experience highlights
canals, and bridges to the organization and     the difficulty of doing both well.
implementation of literacy, public health,
sanitation, reforestation, and civic educa-
tion activities. The projects were explicitly   Mozambique
designed to deliver assistance rapidly,
have high visibility, provide tangible ben-      Apart from the Beira and Tete corri-
efits, and support the legitimacy of local  dors, which were secured by the Zimba-
grass-roots organizations. According to a   bwean Army, there was no functioning
midterm evaluation, “there is absolutely    infrastructure (roads, bridges, or rail lines)
no doubt that this program has had un-      or rural markets in Mozambique in 1990.
precedented success in mobilizing highly    They had been destroyed by years of civil
valued resources to tens of thousands of    war. During the war USAID provided
needy beneficiaries all over Haiti” (Chan-  emergency assistance to save lives and
dler 1996). The evaluation also notes, how- alleviate suffering. When the war ended,
ever, that the program emphasized instal-   assistance programs turned toward reha-
lation more than maintenance.               bilitation and reconstruction. They had
                                            two objectives: to restart the rural economy
      In sum, employment generation ac- and to restart subsistence agriculture.
tivities provided short-term benefits but Roads were demined, rebuilt, and re-
not permanent, off-farm sources of in- opened. That helped restore the private
come. Rehabilitated infrastructure contrib- transport sector and facilitated free
uted to increased economic activity in the movement of goods (especially food) from
short term, but links to long-term eco- surplus to deficit areas, which spurred
nomic development were tenuous at best. market development. At the same time,
This is understandable. Relief agencies assistance was provided to displaced
were working in an environment of social farmers to help them resume agricultural

Results: Economic Effects                                                             35
production and reduce their dependence           ing poorly, and trade was limited. Civil
on food aid.                                     war and socialist economic policies had
                                                 taken their toll. USAID and the World Bank
      Seeds and tools. Farmers needed seeds      supported efforts to reduce state control
and tools to recapitalize their farms. The       of markets and prices and to promote
total cost was small in absolute terms, less     privatization of state-owned enterprises.
than $50 per household. But it was large         These measures helped establish the foun-
relative to per capita income. PVOs located      dation for rapid growth of small markets
seeds and tools (generally not available in      and increased activity of private traders.
local markets), purchased them, and made         They also encouraged refugees and inter-
them available to farmers. Food for the          nally displaced persons to resettle. At the
Hungry International also provided agri-         same time, USAID funded a commodity
cultural inputs, conducted field trials to       import program that supplied imports
identify higher yielding varieties with          needed to support economic liberalization.
shorter growing seasons, and introduced          These initiatives together with the pro-
improved farming practices. Both pro-            gram to demine and rehabilitate roads
grams helped restart subsistence produc-         helped open up trade in rural areas.
     Food for work. USAID gradually
stopped providing relief food and began               Approximately one million Hutu
supporting food-for-work projects. Doing         refugees returned to Rwanda within a one-
so helped break the dependency mental-           month period in late 1996. The humani-
ity. The program supported labor-inten-          tarian community immediately was faced
sive rural road construction and rehabili-       with the difficult task of helping this mass
tation, construction of schools and health       of humanity put their lives back together.
clinics, and rehabilitation of small-scale ir-   The task of linking relief to development
rigation works. The quality of construction      assumed new dimensions. USAID ad-
was generally satisfactory, but mainte-          dressed not only development needs in ag-
nance was questionable. As economic re-          riculture, health, education, and commerce
covery continued, the food-for-work              but also the preconditions for develop-
projects evolved into cash-for-work              ment: political stability, physical security,
projects. That helped create sustainable         justice, and legitimacy of the new govern-
market mechanisms for supplying food             ment.
and other consumer goods as the cash
economy developed.                           Seeds and tools. The Office of Foreign
                                        Disaster Assistance supported the distri-
   Economic liberalization. In 1990 Mo- bution of seeds and tools in almost every
zambique’s rural markets were function- region of Rwanda in 1995 and 1996. The

 36               Complex Humanitarian Emergencies and USAID’s Humanitarian Response
largely untargeted distribution was report-           Capacity building. Most complex hu-
edly successful. In 1997 the size of the pro-   manitarian emergencies have occurred in
gram was reduced, and by 1999 it was ter-       countries where the government was very
minated, except in the two northwest pre-       weak (as in Somalia). This meant nongov-
fectures where continued high levels of         ernmental organizations could operate in
insecurity had led to widespread looting        an environment relatively free from gov-
of farm supplies. Although livestock tra-       ernment intervention. Rwanda was an ex-
ditionally had been used to maintain soil       ception. The government preferred min-
fertility and provide an important source       istry-administered programs rather than
of nutrition to farm families, relief organi-   NGO-administered programs, and it found
zations (including OFDA) were slow to in-       it difficult to incorporate NGOs into its pro-
clude livestock in emergency packages.          grams. Communication between the two
The conventional wisdom was that ani-           deteriorated, and in late 1995, 16 of about
mals represented a level of assistance far      60 humanitarian NGOs were expelled or
beyond “emergency” requirements. How-           asked to suspend operations.
ever, most families are now acquiring ani-
mals, either through loan programs or                USAID   was an early and strong sup-
with their own savings.                         porter of the government’s effort to claim
                                                control of relief and development pro-
     Seeds of Hope. Agricultural experts        grams. Believing that a stable, fair, com-
from the international agricultural re-         petent government was key to Rwanda’s
search centers recognized that productive       successful agricultural and economic de-
cropping in the various microclimates of        velopment, USAID helped strengthen the
Rwanda required adapted seed. They              government’s capacity. Addressing justice
identified appropriate seed and rootstock       in response to the genocide had high pri-
from their own seed banks as well as in         ority. USAID funded the Rwandan-initiated
Rwanda where adapted seed still existed.        International Genocide Conference in
Seed stocks were multiplied and made            1995; trained court clerks in the Ministry
available in 1995. Because the experts had      of Justice; supported a media campaign on
alerted nongovernmental organizations           the genocide trial process; and developed
about the importance of planting adapted        a central database for genocide prosecu-
seed, more local seed was used than oth-        tors. USAID also supported decentraliza-
erwise would have been the case. Seed           tion of the Ministry of Health and helped
multiplication is now being expanded            the ministry establish an emergency re-
under Seeds of Hope II. At the same time,       sponse unit. Finally, USAID provided ba-
food-for-work programs were imple-              sic equipment to 10 ministries including
mented to reclaim wet lowland farming           justice, health, interior, and the president’s
areas (marais) and to improve terracing and     office. Most NGOs now work more closely
land productivity.                              with government officials and seek oppor-

Results: Economic Effects                                                                 37
tunities to help build government capac-      In Haiti, emergency food assistance (dry
ity.                                          rations) was still being distributed in the
                                              northwest as recently as July 1998—years
      Democratic initiatives. The Office of after the end of the crisis. (However, it was
Transition Initiatives has funded two ac- scheduled to terminate in September
tivities designed to support decentraliza- 1998.)
tion and educate local leadership. One is
the Women in Transition program, which             In 1998, four years after the emer-
has reached over 160,000 women. It en- gency in Mozambique ended, most of the
courages commercial interaction among village groups interviewed by the evalua-
different ethnic groups. As part of a broad- tion team in the Beira corridor were still
based program to support women in asking NGOs for free seed, tools, food, and
postconflict situations, it also assists even tractors. Although the NGOs had in-
women’s groups in the northwest, where formed recipients that free food would end
there has been an increase in the number by a specific date, many did not expect that
of women farmers. The other democratic to happen. They remained in the refugee
initiative is administered by Africare in the camps until free food was actually termi-
Ministry of Interior. It supports election nated.
education by building local decision-mak-
ing processes and grass-roots organiza-            Donor pressure increased after the
tions. The program operates in 15 com- Dayton Peace Accords in 1995 to reduce
munes in four prefectures.                    emergency assistance to Bosnia. The logic
                                              was that peace would bring stability and
      All the activities described above economic recovery, making large reduc-
were designed to meet immediate needs tions in food aid possible. There were also
while preparing for follow-on programs. the usual arguments about avoiding food
In contrast to long-term development-ori- aid dependency and disincentives to ag-
ented programs whose success depends on ricultural production. In response to do-
their being sustainable, this was not a cri- nor demands, the joint UN High Commis-
terion for success in Rwanda. As the evalu- sioner for Refugees/World Food Program
ation states, “relief and transition pro- mission recommended reducing the num-
grams . . . do not have to be sustainable.” ber of direct beneficiaries from 1.6 million
                                              to 600,000 (from over 50 percent of the
                                              population to about 20 percent as esti-
When Is                                       mated by WFP in 1996) ( UNHCR /WFP 1998,
The Emergency Over?                           8). That would help ensure that food aid
                                              was not seen as an alternative to a social
      At some point the emergency ends welfare system. However, the joint evalu-
and development resumes. But the demar- ation also pointed out the importance of
cation between the two is not always clear. not scaling down too rapidly because an

 38              Complex Humanitarian Emergencies and USAID’s Humanitarian Response
effective social welfare system was not yet aid provided in a complex emergency is
in place.                                    driven not only by an assessment of needs
                                             but also by the availability of donor re-
     In 1996–97, as many as 118,000 farm sources. The implication is that complex
families in Liberia received food rations emergencies may end too quickly (with
as well as seed and tools under a program humanitarian needs still unmet) or not
funded by the Office of Foreign Disaster quickly enough—depending on what the
Assistance, the European Union, and the various political interests stand to gain or
Food and Agriculture Organization to lose.
support the transition from war to recov-
ery. This ensured that rice seed was              Hill (1997) reports that many conflicts
planted rather than consumed and also are unreconciled and have not ended. This
that farmers had the energy to work. As a is true, for example, in Afghanistan, Iraq–
result, rice production increased from ap- Kurdistan, Lebanon, and Somalia. When
proximately 30 percent of prewar levels in a conflict does subside, it is for a reason.
1996 to 60 percent of prewar levels in 1997 Sometimes it is because one side wins. In
(USAID/Liberia 1998, 13). At the beginning the post–Cold War era, though, this has
of 1997, 350,000 beneficiaries were receiv- rarely been the case. More often it is be-
ing emergency food aid in IDP camps; by cause the war is no longer profitable. Of
the end of the year, 150,000 had been per- course, economic issues are not the only
manently resettled in rural areas. The do- motives perpetuating complex emergen-
nors developed a plan to end general food cies. Religious or ethnic hegemony have
aid distribution in IDP camps after Febru- often driven conflict (in Bosnia and Sri
ary 1998 and to rechannel these resources Lanka, for example). But whatever its
to targeted activities in rural areas. These causes, war is expensive and must be fi-
included rural resettlement of refugees nanced. And for those who invest in the
and internally displaced persons, agricul- war, it must be seen as profitable, at least
tural recovery, school feeding, and food for eventually; otherwise, they would cease
work.                                        their support.

      In Somalia, as elsewhere, humanitar-           Economies in post–Cold War con-
ian assistance and related relief operations   flicts, as with economies in all wars, re-
generated substantial local employment         volve around scarcity. With the disruption
and purchasing power. Over 50,000 Soma-        of outside trade, loss of incomes, and sev-
lis found cash or food-for-work employ-        ered or restricted corridors for delivery of
ment. But this all ended when the relief       goods, food and other essentials become
ended, and the relief ended with termina-      very expensive. Profits are enormous for
tion of the donor-funded contracts with        those who have access to scarce resources
NGOs that provided the relief (Kleist 1994,    and can deliver them to areas where they
295). This suggests that the amount of food    are needed. The combatants and their al-

Results: Economic Effects                                                              39
lies are in the best position to manipulate     also important. External assistance poten-
and profit from this trade.                     tially has an important role to play in all
                                                of these areas.
     Relief organizations are not. On the
contrary, relief organizations attempting to
operate in conflict situations provide a rich   Conclusion
source for exploitation. There are several
reasons. They have to operate with large           The notion that relief assistance can
amounts of cash, which can be stolen. War-    be made more “developmental” or that it
ring factions can exact exorbitant fees from  can be linked to development activities is
relief organizations in return for provid-    highly problematic in the context of ongo-
ing them protection and ensuring their        ing armed conflicts owing to the frequent
access to insecure areas (as in Somalia).     lack of local social and economic structures
And they can charge rents for warehouses      that might legitimately be strengthened
that are higher than market value. Some       (Borton and Macrea 1997, 8). Unlike the
of the most obvious spoils of war derive      case of natural disasters, with complex
from the blatant looting of infrastructure.   emergencies there is no institutional
In Bosnia, for example, an entire             framework to provide security and justice,
Volkswagen production plant near              both of which are necessary preconditions
Sarajevo was dismantled and sold. In So-      for successful development activities. On
malia, almost every phone line, electric      the contrary, complex emergencies are of-
cable, and water line was taken or ripped     ten characterized by (1) a total breakdown
up, put on a ship, and sold in some port      of state institutions (Somalia, Liberia, Af-
on the Indian Ocean (Hill 1997, 5). Thus,     ghanistan); (2) large areas of territory held
peace can threaten a very profitable situa-   for prolonged periods of time by rebel
tion for the combatants.                      movements (Eritrea, Tigray, south Sudan);
                                              or (3) a situation in which the occupying
     Can donors help end wars by dem- regime had not received international rec-
onstrating that more can be gained ognition or was subject to international
through peace? According to Hill, the ba- sanctions (Sudan, Rwanda, Cambodia)
sic elements of reconstruction include (1) (Borton and Macrea 1997, 30).
identifying and creating markets for
manufactured and agricultural goods; (2)           In the absence of physical security and
reestablishing the rule of law, especially political stability, support of long-term
in economic issues; (3) increasing oppor- economic development can be risky. In
tunities for entrepreneurs; (4) supporting Sudan, for example, water and health in-
self-sustaining lending institutions; and (5) frastructure that had been rehabilitated
reestablishing acceptable levels of water, was later destroyed by military action. In
power, heat, sanitation, and other basic Sri Lanka, the assets rebuilt by a World
services. Education, often overlooked, is Bank–led emergency rehabilitation and

 40               Complex Humanitarian Emergencies and USAID’s Humanitarian Response
reconstruction program in 1987–90 at a cost    opportunities that may arise. For example,
of $125 million were destroyed when hos-       many NGOs providing emergency assis-
tilities resumed (Borton and Macrea 1997,      tance also routinely provide funds for ba-
32). These examples reflect the unpredict-     sic agricultural inputs, food-for-work pro-
ability of warfare. They also caution          grams, and housing construction. But ex-
against prematurely supporting long-term       perience suggests that “large rehabilitation
investments in economic development            financing may be more appropriately pro-
before political stability exists.             vided after resolution of the political frame-
                                               work, rather than during the process of
     This does not suggest that relief agen-   political transition itself” (Borton and
cies blindly ignore development-oriented       Macrea 1997, 30–31).

Results: Economic Effects                                                                41
                                     Lessons Learned, and
                            6        Recommendations

S   OME OBSERVERS CONSIDER        each com-
     plex humanitarian emergency unique.
The implication is that past experience is
                                                 though inadequate monitoring makes it diffi-
                                                 cult to quantify results.

not applicable to future crises. By contrast,          One-half million to 1.3 million Hai-
evaluation analysts tend to look for com-         tians (as many as one in seven) received
mon themes—even when assessing the ef-            food aid during 1991–96. In Mozambique
fectiveness of interventions in response to       an estimated one third of the population
complex emergencies. Their underlying             of 16 million depended on food aid for 60
premise is that past experience is, in fact,      to 70 percent of their food needs in 1989.
relevant for future crises. On the basis of       In late 1996 and early 1997, 1.3 million
the three country case studies (Haiti,            refugees were repatriated to Rwanda from
Mozambique, and Rwanda) and evalua-               neighboring countries and received food
tions of other complex emergencies, at            aid. Without massive infusions of pre-
least six common themes and four recom-           dominately U.S. emergency assistance,
mendations emerge. Implications specific          more Haitians would have fled Haiti seek-
to the Kosovo crisis are summarized in            ing refuge in the United States. Massive
annex D.                                          starvation and human suffering would
                                                  have occurred in Mozambique and
                                                  Rwanda. Emergency assistance clearly
Conclusions and                                   helped save lives and alleviate suffering.
Lessons Learned                                   However, except in Haiti, data collection
                                                  and monitoring were not done (or were
     1. Saving lives. Emergency assistance done poorly), so it is difficult to quantify
programs funded by USAID and implemented results.
by American nongovernmental organizations
appear to deliver sufficient assistance to ensure      2. Relief distribution. Effective distri-
the survival of a country’s vulnerable poor, bution of emergency assistance requires orga-
nization and control to limit theft, minimize lacked the technical expertise to plan, or-
abuse, guard against political manipulation, ganize, and manage the distribution of
and protect beneficiaries.                    massive supplies of relief aid. Leakage was
                                              typically 30 percent, and at one point 50
      Distributing relief supplies was a percent was lost, stolen, or diverted. In
problem to some extent in all three coun- response, donors, NGOs, and the private
tries. The large quantity of food aid, in sector took over much of the distribution,
particular, became a source of violent com- and losses dropped to under 5 percent. In
petition—not only for its value as food for camps in Tanzania and Zaire, more food
consumption but also as a source of po- aid was supplied than was necessary, and
litical power for those controlling access. more than usual was misappropriated.
In Haiti, fighting among beneficiaries Some NGOs suspended their operations
sometimes erupted when food was distrib- because they knew they were assisting
uted. Distribution points used to stockpile people guilty of crimes against humanity.
food supplies were looted and supplies
were hijacked. Local authorities some-             3. Political and social unrest. Emer-
times used food to favor certain political gency assistance can help maintain social calm
factions or for their personal aggrandize- and mitigate political instability. Conversely,
ment. In Mozambique as well there were it can exacerbate political tensions. Rarely is it
reports of corruption, theft, and political politically neutral.
or personal favoritism in food aid distri-
bution. Target populations did not always          The international community pro-
receive timely and sufficient food aid. In vided massive quantities of emergency
Rwanda the military and former govern- assistance to Haiti, Mozambique, and
ment leaders controlled much of the relief Rwanda. The political effects of the assis-
distribution. Thus they were able to divert tance varied. In Haiti, food aid reduced
food from the intended beneficiaries for the probability of food riots during a pe-
their own purposes.                           riod of political and economic stress and
                                              may have had a dampening effect on po-
      NGOs were mainly in charge of relief litical tensions; but it also may have re-
distribution in Haiti. They addressed these sulted in a political status quo that enabled
problems by stocking and distributing the de facto military regime to stay in
food aid in neutral settings (schools, fac- power longer. In Mozambique, external
tory yards), using ration cards to track the military assistance provided by the Soviet
receipt of food aid, and having NGO per- Union and by South Africa fueled the war.
sonnel and occasionally police present to Food aid, by comparison, had relatively
monitor distribution. These measures lim- little effect on the country’s political dy-
ited diversion to less than 10 percent and namics, although food diverted to soldiers
helped reduce violence. In Mozambique may have contributed to the war effort. In
the government emergency relief agency Rwanda, where genocidal killers were

 44              Complex Humanitarian Emergencies and USAID’s Humanitarian Response
mixed with legitimate refugees in camps, obliged to sell their agricultural and house-
humanitarian assistance served to prolong hold assets to survive. Numerous farmers
the emergency.                                  in Mozambique and Rwanda also lost their
                                                productive resource base when they fled
      4. Demobilization. Demobilization, their villages. These people became depen-
disarmament, and reintegration of armed forces dent on emergency relief. USAID and the
is vital in ending a complex emergency and NGOs responded—not only with food as-
beginning a period of recovery.                 sistance but also with agricultural inputs
                                                (seeds and tools) and household goods.
      Demobilization of Haiti’s armed That assistance encouraged refugees and
forces removed one source of violence in internally displaced persons to return to
the country. However, many of the demo- their villages. It enabled them to resume
bilized soldiers retained their arms, and food production and decreased their de-
because most were unemployed owing to pendence on food aid. NGOs in all three
the weak economy, they are believed to countries also implemented food-for-work
have caused at least part of the post-1994 programs that created short-term jobs and
rise in theft and street violence. In helped rehabilitate productive infrastruc-
Mozambique, demobilization of Renamo ture (roads, irrigation) needed for eco-
and Frelimo armed forces and their rein- nomic development. Often, though, the
tegration into civilian life was essential for infrastructure was not maintained.
the transition from relief to recovery. As
in Haiti, though, many weapons were not              6. Donor coordination. A clearly des-
turned in and that contributed to a rise in ignated, agreed-upon central authority can
crime. In Rwanda, soldiers and militia make the delivery of humanitarian assistance
loyal to the former government remain more effective.
armed. They are still trying to destabilize
the present government of Rwanda by us-              In Haiti the United Nations officially
ing a campaign of propaganda and terror designated the Pan American Health Or-
to destroy the political and social struc- ganization as the coordination point for
tures of the country, beginning in the overall health planning and services dur-
northwest.                                      ing the U.S.–led embargo. That enabled
                                                numerous NGOs to deliver medical sup-
      5. Relief to development. Emergency plies and food to vulnerable populations
assistance that enables people to protect their more effectively. In Mozambique, by con-
livelihoods (as well as meet immediate needs) trast, donor efforts at times overlapped or
helps reduce dependency and contributes to worked at cross-purposes. One donor was
long-term economic development.                 giving free seed while another was sell-
                                                ing it; one donor was shifting to develop-
      In Haiti many urban factory workers ment assistance while another was still
lost their jobs, and some farmers were providing grant relief. That confused ben-

Conclusions, Lessons Learned, and Recommendations                                      45
eficiaries and undermined efforts to re-         able political consequences. Control over
duce dependency. Similarly, lack of donor        final distribution often has reinforced the
coordination was a serious problem in            power of local authorities or political fac-
Rwanda.                                          tions. It has strengthened their relative
                                                 position during or after the conflict and
                                                 facilitated their self-aggrandizing, often
Recommendations                                  exploitive, behavior toward the intended
                                                 noncombatant beneficiaries. Decisions to
    Four key recommendations emerge              continue, withdraw, or modify aid distri-
from these six conclusions and lessons           bution should be made as a matter of de-
learned. They are mainly management              liberate policy on a regular basis by each
oriented.                                        individual donor.

      1. Monitoring and evaluation. Estab-           3. Reducing dependency. Give refu-
lish a central monitoring and data-collection gees incentives to return home and impose dis-
unit to serve all donors during the early weeks incentives on those remaining outside their
of a complex emergency.                         country of origin.

     Baseline data for socioeconomic indi-              Generally, the longer encampment or
cators (e.g., malnutrition rates, food prices,     temporary foreign residence lasts, the less
population displacement) can help man-             willing refugees are to return home. A
agers identify appropriate kinds of emer-          combination of “push” factors (such as ter-
gency relief, target its distribution, and         minating free food distribution) and
subsequently measure and evaluate its ef-          “pull” factors (such as including seeds and
fectiveness. Close monitoring enhances             tools in resettlement packages) is likely to
donor coordination and is essential for            accelerate the repatriation process. But for
assessing aid needs, avoiding work at              reasons of political and bureaucratic self-
cross-purposes, identifying recipient              interest, local governments may not re-
groups no longer needing emergency aid,            move from the rolls those no longer need-
shifting from relief to reconstruction and         ing relief. Therefore, donors must moni-
development, and designing and adjust-             tor each situation closely, recognizing that
ing economic policies.                             both relief and development assistance
                                                   may be needed if some areas remain in
      2. Adverse political consequences. Be emergency status while others stabilize
alert to potential undesirable political or social more quickly. After populations have been
effects that relief aid may cause.                 repatriated and are settled, the agricultural
                                                   base begins to be reestablished, depen-
      Emergency food distribution, in par- dency on free food distribution drops, and
ticular, can have unintended and undesir- long-run food security is enhanced.

 46                Complex Humanitarian Emergencies and USAID’s Humanitarian Response
                                           Such skills are likely to be in short supply,
     4. Capacity building. Train technocrats
                                           especially if preconflict professionals and
to manage the postconflict economic transition,
                                           the intelligentsia were targeted for delib-
and train others in skills for which there is
employment demand.                         erate elimination or have permanently left
                                           the country. Recovery also needs to be
    Complex emergencies seriously dovetailed with postconflict economic re-
weaken the capacity of governments to alities. Job training is fruitless if unemploy-
provide basic public services. Economic ment in the depressed economy remains
recovery requires a cadre of high-level high. Training is especially critical for de-
technocrats with both management and mobilized soldiers who, because they of-
conceptual skills, especially in macroeco- ten remain unemployed, tend to turn to
nomic and sectoral policy formulation. destabilizing criminal activity.

Conclusions, Lessons Learned, and Recommendations                                   47
                     Emergencies and
             Annex A Donor Assistance

T     HE TABLE BELOW   indicates the num-
     ber of people affected by humanitar-
ian emergencies in recent years. The four
                                                  levels of official development assistance
                                                  and humanitarian assistance provided by
                                                  the donor community and the United
figures in this annex show changing               States over the past 30 years.

 Table A1. Ongoing Humanitarian Emergencies and Number of
          People Affected (Millions), 1996, 1997, 1998
                                     January 1996          January 1997             April 1998
 Afghanistan                             4.0                    3.5                   4.1
 Angola                                  2.5                    2.5                   2.5
 Azerbaijan                              0.95                   0.78                  0.77
 Bosnia and Herzegovina                  3.7                    3.1                   1.5
 Burundi                                 0.8                    1.0                   0.75
 Colombia                                                                             1.0
 Croatia                                  0.5                    0.48                 0.45
 Eritrea                                  1.0                    0.7                  0.30
 Ethiopia                                 3.5                    2.5                  0.85
 Georgia                                  1.0                    0.3                  0.3
 Haiti                                    1.1                   >0.5                  0.5
 Iraq                                     2.65                   2.1                  1.7
 Liberia                                  1.5                    2.0                  1.1
 North Korea                                                    >5.0                 >7.4
 Russia (Chechnya)                        0.3                    0.35
 Rwanda                                   1.0                    0.63                 0.3
 Sierra Leone                             1.8                    1.5                 >1.0
 Somalia                                  1.0                    1.0                  1.0
 Sri Lanka                                0.85                   0.85                 0.7
 Sudan                                    4.0                    4.4                  4.4
 Tajikistan                               1.0                    0.63                 0.9
 Uganda                                                                               0.4
   Total                                 33.15                   33.82               31.92
 Source: U.S. Mission to the United Nations, April 1997 and September 1998. Figures are based on
 data provided by the U.S. Committee for Refugees.

 Note: The table includes only major emergencies, those in which at least 300,000 people required
 international humanitarian assistance to avoid severe malnutrition or death.
                                      Figure A1
Official Development Assistance and Emergency/Distress Relief, All Donors, 1969–98
                                (billions $US 1998)
             80                                                                                       5
                                                      ODA                                             4
             50                                                                                       3


             30                                                                                       2

                 0                                                                                    0
                     1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997

                     Source: OECD/DAC

                                       Figure A2
     Emergency and Distress Relief as a Percent of Official Development Assistance,
                                 All Donors, 1969–98

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

                         Source: OECD/DAC

A2                              Complex Humanitarian Emergencies and USAID’s Humanitarian Response
                                          Figure A3
       U.S. Official Development Assistance and Emergency/Distress Relief, 1971–98
                                    (billions $US 1998)
              16                                                                           2.0
                                            U.S. ODA
              12                                                                           1.5


               8                                                                           1.0
                                                         U.S. E/DR
               4                                                                           0.5
               0                                                                           0.0
                   1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997

                   Source: OECD/DAC

                                                Figure A4
              Trends in U.S. Official Development Assistance and Emergency/Distress Relief,
                                      U.S. E/DR % of all E/DR

                                                         U.S. ODA % of all ODA



                               U.S. E/DR % of U.S. ODA
                   1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997

                   Source: OECD/DAC

Annex A: Humanitarian Emergencies and Donor Assistance                                                  A3
                    Evaluation Objectives
                    And Constraints and
                    Implications for
            Annex B Donor Coordination
Evaluation Objectives                                1. Did U.S. emergency assistance save
                                                lives and alleviate suffering during the

T    HE OVERALL OBJECTIVE       of the CDIE     complex emergency?
      evaluation was to assess the effective-
      ness of U.S. emergency assistance in           2. Did U.S. emergency assistance af-
response to complex humanitarian emer-          fect social and political hostilities (or ten-
gencies.* It covered all types of relief as-    sions) associated with the complex emer-
sistance including (1) food, (2) water and      gency?
sanitation supplies, (3) medical services
and health care, and (4) clothing, shelter,          3. Did U.S. emergency assistance con-
and resettlement assistance. The evalua-        tribute to economic development?
tion in each country examined the politi-
cal and historical events that caused the       The first question derives directly
emergency as well as the effectiveness of  from legislation that gives USAID primary
the U.S. humanitarian response. Conclu-    responsibility within the U.S. government
sions and lessons learned are cast in termsfor responding to overseas disasters. Ques-
of management recommendations to help      tions 2 and 3 are not directly related to the
guide USAID’s future policy, program, and  overarching mandate to “save lives and
budget decisions in humanitarian assis-    alleviate suffering.” The second question
tance.                                     concerns the potential political impact of
                                           emergency assistance and whether such
                                           assistance has had the unintended effect
Evaluation Methodology                     of prolonging complex emergencies. The
                                           third question addresses the potential eco-
      The evaluation looked at three main nomic impact of the assistance and the so-
topics framed in the form of these ques- called relief-to-development continuum.
                                                Background work for the evaluation
*The “Emergency Assistance” concept paper
                                           was carried out in Washington, and field-
(11 December 1997) provides the underlying
rationale for the assessment.              work was conducted in Haiti, Mo-
zambique, and Rwanda. During the first             This mainly qualitative methodologi-
phase, desk studies for each of the three     cal approach does not produce statistically
countries were completed, a workshop          valid proof of impact. Rather, it allows an
was convened to review the desk studies,      interpretation of the links between a USAID
and two topical guides were developed to      intervention and various effects that plau-
help structure key informant interviews       sibly can be associated with that interven-
and field observations. One topical guide     tion. Given the characteristics of humani-
was for implementers and experts and the      tarian assistance, a more scientific ap-
other was for beneficiaries.                  proach is rarely feasible (Hallam 1998, 28).
                                              One can judge the validity of the interpre-
      The second phase involved fieldwork     tations on the basis of several criteria: the
in each of the three countries. Evaluation    logic and consistency of the arguments
teams sought answers to the three ques-       substantiating them, the strength and qual-
tions above through careful analysis of       ity of the evidence, triangulation, and the
secondary sources not available in the        reputations of those involved. This is
United States (reports, evaluations, agri-    sometimes called the common-sense
cultural production surveys, health statis-   school of evaluation: impact is deduced
tics), interviews with key informants and     from a combination of information from
focus groups, and site visits. The results    key informants and from the evaluators’
of the three case studies were synthesized    own sense of how the world works.
during the third phase.

     During the second phase, evaluators      Evaluation Constraints
talked with a broad range of national and
expatriate experts (field technicians, pro-        Donor agencies that provide relief in
gram directors, administrators) who had       the context of conflict or civil war must
managed or implemented emergency as-          consider factors they normally would not
sistance programs. These people typically     need to. These factors include sovereignty,
included donor agency staff, NGO partners,    international law, the appropriate balance
host government officials (both national      of aid between opposing sides, and per-
and provincial), UN agencies, and some        haps national foreign policy interests. The
academics and journalists. Beneficiary-       political and legal questions associated
level information was collected in both       with humanitarian assistance in conflict
urban and rural areas. Site visits produced   areas make the evaluation of relief pro-
valuable insights and helped corroborate      grams far more sensitive than that of de-
information from other sources and to         velopment programs (Borton 1994, 11–12).
ground-truth the teams’ interpretations.
The process of interviewing and listening          Moreover, difficulties typically asso-
produced answers and also generated sec-      ciated with evaluating development assis-
ondary questions.                             tance programs seem to be magnified

B2               Complex Humanitarian Emergencies and USAID’s Humanitarian Response
when evaluating relief programs. These                 2. Humanitarian organizations are
include (1) lack of adequate baseline data;       action oriented. As a result, evaluations
(2) difficulty in identifying control groups;     generally are not read and lessons are not
(3) the dynamic context of relief programs,       learned (11).* Although the action orien-
which makes it difficult to isolate the ef-       tation of NGOs is a positive attribute, it
fect of the relief intervention; (4) the lim-     should be informed by past experience as
ited utility of cost–benefit analysis; (5) the    well as current political, military, and so-
large number of agencies involved; and (6)        cial realities. Unfortunately, NGOs some-
the high political and media profile of re-       times ignore this experience rather than
lief programs (Borton 1994, 12).                  use it as a basis for more strategic inter-
     There is yet another difficulty with
evaluating relief programs: the prevalent              3. Humanitarian organizations are
attitude among many relief agencies that          often defensive about criticism, even con-
assessments of their programs are unnec-          structive criticism. As such, they are un-
essary. To paraphrase their view: our mo-         likely to use it to improve their effective-
tives were well intentioned, we did our           ness (12). Nevertheless, constructive criti-
best under difficult circumstances, why           cism can be valuable, which highlights the
should we now subject ourselves to a criti-       need for independent research and evalu-
cal examination? (Borton 1994, 1). Relief         ation.
agencies also may believe that “all disas-
ters are different, so what is the point in            4. Humanitarian organizations often
trying to learn the lessons of our response       lack accountability. They are not held re-
to this particular disaster?” (Borton 1994).      sponsible for their actions: everybody—
                                                  and thus nobody—is responsible. This lack
     According to Larry Minear (1998),            of accountability will be difficult to rem-
four characteristics of humanitarian orga-
nizations make them resistant to change:          *The multidonor Joint Evaluation of Emergency As-
                                                  sistance to Rwanda demonstrates the importance of
     1. As just noted, humanitarian orga-         evaluations but also demonstrates their limited
                                                  ability, in and of themselves, to produce institu-
nizations tend to approach every crisis as
                                                  tional change (6). This comprehensive evaluation
unique. “As long as every crisis is per-          reviewed aid programs that cost $1.4 billion from
ceived as wholly without precedent or             April through December 1994. It was done by 37
parallel, there will be little scope for insti-   institutions (governmental, intergovernmental,
tutional learning (10).” Correcting this ten-     and nongovernmental), enlisted 52 consultants,
                                                  produced a five-volume report, and cost $2 mil-
dency will require greater institutional
                                                  lion. The broadest of its 64 recommendations were
memory, greater attention to comparative          ignored; those stressing “coordination by com-
analysis, and more support for in-house           mand” were rebuffed; and only the least radical
evaluation.                                       options were acted on to some degree.

Annex B: Evaluation Objectives and Constraints and Implications for Donor Coordination B3
edy because of the accepted approach of United Nations is not always best
coordination by consensus (or default) equipped to coordinate an international
rather than coordination by command (12– response to a complex emergency.
                                                   The only country with a demon-
                                             strated ability to energize the United Na-
Implications for                             tions and the Security Council in a crisis is
Donor Coordination                           the United States. But in the case of
                                             Rwanda, even the United States, haunted
      This raises the issue of donor coordi- by the memories of Somalia, was deter-
nation. Coordination is a concept ap- mined not to get involved in another Afri-
proved by all but defined by few can conflict. Not crossing the “Mogadishu
(Prendergast and Scott 1996). A major line” became the guiding principle. Wash-
emergency assistance operation can in- ington was also preoccupied with crises
volve numerous bilateral and multilateral elsewhere, especially in Bosnia and Haiti,
donors, hundreds of NGOs, a range of UN and the potential financial burden of
agencies, military contingents, and na- Rwanda was a major concern. The United
tional governments. An effective division States at the time was assessed 31 percent
of labor among these and other actors is of the costs of all UN peacekeeping op-
needed to maximize the comparative ad- erations. (Joint Evaluation 1996, study 2,
vantage and impact of each.                  11). In short, even the United States is not
                                             always well positioned to coordinate an
      Donor coordination is often perceived appropriate humanitarian response to all
as a role for the United Nations. That body complex emergencies.
expanded its peacekeeping operations at
the end of the Cold War when complex               Lack of donor coordination in provid-
emergencies proliferated. But peacekeep- ing humanitarian assistance is well docu-
ing operations required endorsement by mented. This is true within a single orga-
the Security Council. This meant the Per- nization as well as among several organi-
manent Five could control UN peacekeep- zations. For example, within the UN sys-
ing and enforcement operations given tem, UNICEF has responsibility for women
their veto power and control over finances. and children; but UNHCR has responsibil-
In the case of Rwanda, the major powers ity for refugees, including women and
on the Security Council (except France) children. Similarly, WFP has responsibil-
made clear they were not interested in a ity for food assistance; but UNICEF and
small African country that was marginal WHO may be involved in actual program
to their economic or political concerns and operations. Lack of coordination can re-
peripheral to international strategic rival- sult in inefficiency. For example, in most
ries. This experience suggests that the situations each agency arranges local

B4               Complex Humanitarian Emergencies and USAID’s Humanitarian Response
transport for its own supplies, but this pro-    sponse to future complex emergencies
cess only bids up prices. Another issue          should necessarily be built around the
having cost-related implications is that hu-     UN—and suggests there are a variety of
manitarian organizations are heavily             reasons why it should not. He recom-
weighted toward mop-up operations; they          mends the International Committee of the
are reactive, not proactive. Too little atten-   Red Cross or a new organization outside
tion is given to investments in conflict pre-    the UN system (Apthorpe 1997, 96, 98).
vention (Minear 1994b, 4, 6); see annex C.
                                                      In disaster after disaster, local insti-
     The creation of the UN Department           tutions and people provide the first line
of Humanitarian Affairs (and subse-              of response. By contrast, the world’s hu-
quently, the Office for the Coordination of      manitarian system relies heavily on out-
Humanitarian Affairs) is generally seen as       side resources that marginalize local re-
a positive step. Nevertheless, the UN is         sources and expertise. The idea seems to
often criticized. For example, each of the       be fixed that in complex emergencies there
six evaluations reviewed by Apthorpe calls       is simply no alternative to using donor
for some type of reform in the United Na-        resources delivered by NGOs. This may be
tions. James Ingram (former head of WFP)         true. If so, donor resources should be pro-
is more draconian: he believes there is no       vided to support existing livelihoods or
reason that a coordinated international re-      coping strategies as much as possible.

Annex B: Evaluation Objectives and Constraints and Implications for Donor Coordination B5
                     Are Complex
             Annex C Predictable?

I   N REFERENCE       to the Rwanda crisis,       ian emergencies. It postulates that
    James Kunder notes that “an ounce of          ethnicity (based on differences of lan-
prevention is worth 25,000 tons of food           guage, race, tribe, religion, national origin,
aid” (Minear 1994, 9). The point is well          or some other cultural sense of identity) is
taken. It is far better to prevent complex        the primary factor underlying a complex
emergencies from occurring in the first           humanitarian emergency. By contrast, the
place than it is to respond to victims’ needs     economic model views complex humani-
afterwards. What factors seem to be re-           tarian emergencies and ethnic conflict in
sponsible for igniting complex emergen-           the context of economic development and
cies? Are complex emergencies predict-            structural change. In this model, economic
able? Regions that are particularly prone         factors are pivotal in shaping conflicts,
to seasonal natural disasters (for example,       though these conflicts may be triggered by
cyclones in India, hurricanes in Florida,         political or ethnic causes. That is, economic
drought in the Sahel) have developed              factors create the conditions for ethnic or
early-warning systems designed to predict         political explosions that in turn lead to
the next natural disaster and to mitigate         complex emergencies (Nafziger 1996,
its effects. Such is not the case with com-       3–4).
plex emergencies. However, studies have
been undertaken to help explain the causes                Paul Collier and Anke Hoeffler (1998)
of civil wars and complex emergencies,               are among those who explain the phenom-
and that may be the first step in predict-           enon in economic terms using economic
ing their occurrence. This annex summa-              analysis. They identify four variables that
rizes the results of several of these stud-          they hypothesize are associated with civil
ies.*                                                wars: per capita income, size of the natu-
                                                     ral resource base, population size, and
      The ethnic model is the prevailing ap- ethnolinguistic fractionalization. Using a
proach to explaining complex humanitar- sample of 98 countries (of which 27 had
                                                     civil wars during 1960–92), they analyze
*A substantial body of literature has developed on   the relative importance of these four vari-
this subject. This annex only scratches the surface. ables. Their results show that all four vari-
ables are significant determinants of both     in turn suggests there would be a greater
the probability of civil wars occurring and    likelihood of various ethnic groups want-
their duration. Specifically:                  ing to secede owing to cultural and lin-
                                               guistic disparities. (Note, however, that the
     1. Per capita income. The study shows effect of population size is ambiguous be-
that civil war is overwhelmingly a phe- cause potentially it could be inconsistent
nomenon of low-income countries. Other with point 4, next.)
things being equal, the probability of civil
war is substantially greater in countries           4. Ethnolinguistic fractionalization. This
with a very low per capita income than in characteristic is an index that ranges from
countries with a relatively high per capita 0 (complete homogeneity) to 100 (maxi-
income. Moreover, the predicted duration mum fractionalization). The index would
of civil war is much shorter in countries be 100 when each individual in a country
with a higher per capita income. This is was in a different ethnolinguistic group.
because a high-income population has Conversely, the index would be 0 in a so-
more to lose during a conflict, and the costs ciety with a single ethnolinguistic group.
of rebellion increase with its duration. Contrary to conventional wisdom (and to
Conversely, the opportunity cost of being point 3, above), the authors find that more
a rebel and prolonging a conflict is low for fractionalized societies are no more prone
a low-income population.                       to civil war than highly homogeneous
                                               ones. Rather, the danger of civil war arises
     2. Natural resource base. The effect of when a society is polarized into two
natural resource endowments is not as groups. Polarized societies have a much
straightforward. Initially, increased natu- higher probability of civil war than either
ral resources increase the risk and dura- homogeneous or highly fractionalized
tion of civil war. This is because the tax- ones. Thus, a country with two similar-size
able base of the economy constitutes an at- ethnolinguistic groups could reduce the
traction for rebels wishing to capture the risk of civil war either by partition or by
state. But at a high level, natural resources union with other countries.
start to reduce the risk of civil war. This is
due to the government’s greater financial           Not surprisingly, Collier and Hoeffler
capacity to defend itself through military conclude that the “ideal society” (one en-
expenditures.                                  dowed with the most favorable of each of
                                               these characteristics) has less risk of civil
     3. Population. Countries with larger war than the “catastrophic society” at the
populations have a higher risk of civil war other end of the spectrum (one with the
and wars that are likely to last longer. This least favorable of each characteristic). They
reflects the greater likelihood that coun- also conclude that poverty is the main
tries with larger populations will have a cause of civil war. Table C1 lists the sample
larger number of diverse groups, which of 98 countries in their analysis. Countries

C2                Complex Humanitarian Emergencies and USAID’s Humanitarian Response
    Table C1. Countries With and Without a Civil War, 1960–92
 With a Civil War                                 Without a Civil War
  Algeria                    Argentina             Hong Kong            Sierra Leone
  Burundi                    Australia             Iceland              Singapore
  Chad                       Austria               Ireland              South Africa
  Dominican Republic         Barbados              Israel               Spain
  El Salvador                Benin                 Italy                Sweden
  Ethiopia                   Bolivia               Ivory Coast          Switzerland
  Guatemala                  Brazil                Jamaica              Syria
  India                      Burkina Faso          Japan                Tanzania
  Indonesia                  Cameroon              Kenya                Thailand
  Iraq                       Canada                Korea                Togo
  Liberia                    Cent. African Rep.    Madagascar           Trinidad and Tobago
  Mauritania                 Chile                 Malawi               Tunisia
  Morocco                    Congo                 Malaysia             United Kingdom
  Mozambique                 Costa Rica            Mali                 United States
  Myanmar                    Denmark               Malta                Uruguay
  Nicaragua                  Ecuador               Mauritius            Venezuela
  Nigeria                    Egypt                 Mexico               Zambia
  Pakistan                   Finland               Nepal
  Peru                       France                Netherlands
  Philippines                Gabon                 New Zealand
  Somalia                    Gambia                Niger
  Sri Lanka                  Germany               Norway
  Sudan                      Ghana                 Panama
  Turkey                     Greece                Papua New Guinea
  Uganda                     Guyana                Paraguay
  Zaire                      Haiti                 Saudi Arabia
  Zimbabwe                   Honduras              Senegal

  Source: Collier and Hoeffler 1998, 573.

in column 1 experienced civil war during    cies have experienced several years (or
1960–92; the rest did not.                  even decades) of negative or stagnant eco-
                                            nomic growth. Below a given threshold, a
     Nafziger also identifies economic fac- protracted decline in incomes is likely to
tors as primarily responsible for complex trigger increasingly fierce competition for
emergencies (Nafziger 1996, v). He speci- scarce resources, jobs, and opportunities
fies four, some of which parallel those (5).
identified by Collier and Hoeffler:
                                                 2. Unequal growth. The situation is
     1. Prolonged stagnation. The majority likely to deteriorate more rapidly if income
of countries with humanitarian emergen- and asset distribution worsen. Skewed

Annex C: Are Complex Emergencies Predictable?                                             C3
economic growth increases the relative           flicts were often fought over issues related
deprivation of substantial sections of the       to agriculture, such as land ownership,
population, even if it does not cause abso-      environmental change, water scarcity, and
lute deprivation (6).                            food shortages. However, divisions over
                                                 these issues often fall along ethnic lines,
     3. Population pressure on resources.        obscuring the fundamental causes.
Rapid population growth coupled with en-
vironmental degradation and resource                    The researchers conclude that condi-
depletion can contribute to diminishing          tions in poor countries that undermine the
returns to agricultural land. Declining ag-      rural economy can generate political griev-
ricultural returns, often exacerbated by         ances that result in endemic armed con-
maldistribution of land and water, are a         flict. Agriculture is the dominant economic
source of conflict (8).                          sector in most poor countries. Poor coun-
                                                 tries that invest in their agricultural sec-
     4. Distributional shifts owing to adjust-   tors provide livelihoods to people and
ment programs. Large and abrupt shifts in        thereby lower the incidence of conflict.
the distribution of income and wealth dur-       People do not need to turn to violent move-
ing stabilization and liberalization pro-        ments as a means of survival. These find-
grams can affect the distribution of power       ings, like those of Collier and Hoeffler and
within a country (9).                            Nafziger, stress the importance of eco-
                                                 nomic factors in explaining the outbreak
      According to Nafziger, the way in          of complex emergencies—and in particu-
which elites react to these four factors in-     lar, the importance of agriculture.
fluences the probability of political con-
flict and humanitarian disasters occurring        Finally, Mary Anderson also identi-
(Nafziger 1996, 10).                         fies economic considerations among the
                                             key underlying factors that contribute to
      De Soysa and Gleditsch of the Inter- civil war. She suggests that some engage
national Peace Research Institute, Oslo, in civil war because they have little to lose;
analyzed 103 armed conflicts that occurred others, because they have something to
during 1989–97, since the end of the Cold gain. Those with few economic alterna-
War. They found that political instability tives (the poor) are often the rebels who
that led to violent conflict has sprung have little to lose. They are often supported
mainly from economic concerns, rather by those who have something to gain
than political or ideological differences. (arms merchants and other profiteers).
They also found that most contemporary Others who stand to gain include those
armed conflicts have occurred in impov- whose employment depends on continu-
erished countries where agriculture was ation of the war and the funding it gener-
the mainstay of the economy. Armed con- ates.

C4                Complex Humanitarian Emergencies and USAID’s Humanitarian Response
     Although many leaders claim to be          the major reason for civil war in 75 per-
engaged in a struggle against past injus-       cent of the cases.
tices, the evidence available suggests that
justice and fairness are neither the funda-           In sum, complex emergencies clearly
mental motives nor the likely outcomes of       are not predictable. Moreover, the incon-
their wars. Rather, these leaders seek          sistencies among the findings of the stud-
power. However, once a civil war starts,        ies just noted illustrate that there is no gen-
the war itself creates a spiral of atrocities   erally accepted theory of conflict. It seems
and reprisals and hatred that become the        equally clear, though, that economic fac-
legitimate root cause for its continuation      tors are of major importance in generat-
(Anderson 1996, 10–12). As reported in          ing these crises, and they may well have
one anthropological study, vengeance is         an important role in preventing them.

Annex C: Are Complex Emergencies Predictable?                                             C5
                    Implications for
            Annex D The Kosovo Crisis

I   N MAY 1999 the Kosovo crisis embraced
    much of the Balkans in one way or an-
other. Various proposals were being
                                                Department for International Develop-
                                                ment of the UK, the Directorate General
                                                of International Cooperation of Nether-
mooted for international and regional ac-       lands, the Danish International Develop-
tion to secure a stable peace when the con-     ment Agency, the Organization for Eco-
flict ended. The length of the conflict, the    nomic Cooperation and Development/
extent of casualties and population dis-        Development Assistance Committee, the
placement, the amount of physical de-           UN High Commissioner for Refugees/
struction, the postconflict sovereignty con-    World Food Program, the International
figuration, and the lingering political and     Bank for Reconstruction and Develop-
social effects on the states most affected      ment/Operations Evaluation Department,
were all unpredictable at that time.            the International Crisis Group, and this
      This annex identifies lessons from this
and other evaluations of complex humani-
tarian emergencies that might be useful in Crisis Management
coping with the Kosovo crisis. Applicabil-
ity of any individual lesson depends, of                1. Coordination. The Kosovo crisis de-
course, on how the crisis develops, how it mands an effective coordination mechanism.
is resolved, and what postconflict politi-
cal architecture emerges. The lessons and               Virtually all evaluations dealing with
observations have been culled from docu- complex humanitarian emergencies cite
ments prepared by the following: CDIE, the inadequate donor coordination as causing
                                                  serious problems—inefficiencies, waste,
Robert J. Muscat prepared the original version of lost leverage. Complex emergencies typi-
this annex (25 May 1999). In anticipation of the
                                                  cally face unusually complex coordination
end of the Kosovo crisis, Muscat synthesized les-
sons from this and other evaluations of complex problems: large numbers of multi- and
humanitarian emergencies for possible application bilateral donors and nongovernmental
in Kosovo.                                        organizations working in a context of con-
tested or collapsed governance, sometimes      transitioning from relief to reconstruction,
with external military organizations in-       and designing and adjusting economic
volved (Bosnia, Cambodia, Kosovo, Soma-        policies. The Rwanda experience demon-
lia). Coordination has ranged from volun-      strated that good information on the con-
tary, committeelike structures to more hi-     cerns and expectations of encamped refu-
erarchical, or command, systems involv-        gees, and on the power and leadership
ing lead agencies or even a UN or other        structures in the camps, is critical for main-
sanctioned authority with powers of di-        taining orderly relations and confidence
rection (Bosnia, Cambodia).                    among refugees, assistance authorities,
                                               and host governments.
     Postconflict Kosovo is likely to have
an interim international administration              3. Food distribution. Local food distri-
under which a centrally guided coordina-       bution organizations commonly need close
tion system would be appropriate. Of all       monitoring to avoid factional diversion or
recent complex humanitarian emergen-           politicization.
cies, the Bosnian and Cambodian experi-
ences have the most relevant lessons in this         Judging by the international
regard, especially concerning the security–    community’s generally successful hu-
political–economic interfaces and the          manitarian response record, malnutrition
exercise of authorized powers. (Extensive      or disease is unlikely to become a signifi-
independent and internal agency evalua-        cant complication of the current crisis, with
tions of the international administration      two important caveats. First, given inter-
experience of these two cases are avail-       nal displacement, exposure to ethnic
able.)                                         cleansing, and the bombing of supply in-
                                               frastructure, those remaining inaccessible
     2. Monitoring. A central monitoring       inside Kosovo could suffer severe priva-
and data collection unit should be set up to   tion, depending on how long the conflict
serve all donors.                              lasts. Second, Kosovo, Albania, and
                                               Macedonia each have deeply divided po-
     Complex emergency information-            litical factions and parties. In similar cases
sharing and monitoring systems have also       (Ethiopia, Sudan), factions that controlled
been evaluated negatively. A central unit      food aid distribution have withheld food
to collect and analyze socioeconomic data      from entitled beneficiaries loyal to rival
and program information, one that serves       factions. Thus far, the political dynamics
all donors (again, as in Bosnia or Cambo-      in Albania and Macedonia have reportedly
dia), is needed to ensure donor coordina-      facilitated rather than hindered aid distri-
tion. Close monitoring is essential for as-    bution to refugees. Given the fragile poli-
sessing needs, avoiding work at cross-pur-     tics in both countries and the stresses gen-
poses, identifying recipient groups no         erated by the crisis, the integrity of emer-
longer needing emergency aid,                  gency aid distribution should be closely

D2                Complex Humanitarian Emergencies and USAID’s Humanitarian Response
monitored. If postconflict arrangements       trust funds from donors interested in in-
retain Kosovo’s ethnic Serb minority, the     ducing and supporting a larger Bank ef-
problem could arise there, unless distri-     fort than might otherwise be the case, and
bution is well controlled by external agen-   (b) strong advocacy, by one or more board
cies.                                         members, for maximum Bank perfor-
     4. Refugee repatriation. Refugees may
require incentives to return home.
                                              Relief to Development:
      Generally, the longer encampment or     Transition and Links
temporary foreign residence lasts, the less
willing refugees are to return home               1. Funding discontinuities. USAID
(Mozambique). A combination of incen- should avoid interruptions in funding and
tives for returning to Kosovo and disin- operations during the transition phase.
centives for remaining outside may be re-
quired if NATO ’s expressed optimism              Relief funding affords considerable
about early return does not materialize.     flexibility compared with the procedural
                                             complexity and slower pace of long-term
     5. World Bank coordination of recon- development operations. Thus, USAID and
struction. Several mechanisms can be used to NGOs have experienced funding gaps and
strengthen World Bank performance.           program interruptions between the phase-
                                             out of relief assistance and the phasein of
      The international community has re- development assistance (Mozambique).
cently relied on the Bank to take the lead The two funding systems may need to
in reconstruction, monitoring, and other operate simultaneously in countries where
nonpolitical and nonmilitary functions. On some areas remain in emergency status
the basis of the Bank’s own evaluation, its while other areas have stabilized. The ex-
effectiveness in complex humanitarian istence of land mines in Kosovo could ne-
emergencies has varied, depending to a cessitate operating in both modes at once.
considerable extent on senior manage- Projects undertaken as rapid transition
ment’s level of interest and commitment responses should not ignore longer run
as reflected by (a) authorities and size of reconstruction and development objec-
staff of the in-country resident represen- tives. This may require close coordination
tative’s office, (b) speed of headquarters’ among USAID’s units responsible for dif-
processing and willingness to cut bureau- ferent funding sources and operations.
cratic corners, and (c) size of administra-
tive budget.                                      2. Resettlement planning. Realistic
                                             planning for resettlement in Kosovo should al-
     The Bank’s performance in some cases ready be under way.
has been enhanced by (a) receipt of (grant)

Annex D: Implications for the Kosovo Crisis                                             D3
     The needs of returning refugees dur-        agement requirements. Otherwise, reha-
ing the initial stages of resettlement may       bilitation may be impeded. Development
seem obvious. In practice, though, resettle-     was delayed and opportunities missed in
ment planning has often been deficient,          Cambodia because the interim UN admin-
based on unrealistic assumptions, inad-          istration interpreted its mandate—against
equate information, and poor analysis of         the advice of its economics unit—as limit-
the conditions prevailing in repatriation        ing its authority in economic development;
destinations (Cambodia, Haiti, Mo-               the intent was to leave such decisions to
zambique). If repatriation begins as early       the successor national authorities yet to be
as autumn 1999 (as NATO spokesmen have           established. The economic recovery of the
asserted), then planning, preparation, and       resettled Kosovar population should not
financial provision should already be un-        be delayed by making the interim inter-
der way, including mobilization for com-         national administration’s economic scope
munity and home demining.                        too restrictive.

     Since repatriation will occur at the              4. Economywide distortions. Moni-
onset of winter, the repatriation package        tor the effects of external interventions to de-
will need to include housing reconstruc-         tect inflationary and other economywide dis-
tion (fraught with problems of finance and       tortions.
implementation), food aid until the next
harvest, agricultural assets and inputs for            Local expenditures made by large
next season’s cultivation, and household         numbers of international military, civilian,
items including fuel, livestock restocking,      and NGO personnel can have major posi-
and perhaps cash (if press accounts are          tive and negative effects on fragile crisis
correct that departing Kosovars have been        and postconflict economies. Such expen-
stripped of their money and valuables). If       ditures can help stimulate the recovery of
early return proves infeasible, contingency      local production and service sectors, but
planning for winterizing the camps should        they can also have inflationary and other
also be undertaken immediately.                  adverse effects. For example, they can af-
                                                 fect local housing markets, wage levels of
     3. Agenda for negotiating settle-           scarce local professional and technical per-
ment. Economic dimensions should constitute      sonnel, and wages and perks (including
a key element of future negotiations to settle   supplements) of civil servants seconded as
the conflict.                                    aid-project staff. Well-positioned officials
                                                 and other elite may capture much of this
     Conflict-settlement negotiations and        expenditure, producing new, large income
arrangements should address not only             disparities and consequent resentment.
political and security dimensions but also       The obverse may also be a problem: eco-
economic implications, with due attention        nomic downturn and employment loss
given to the policy framework and man-           when stabilization allows major reduc-

D4                Complex Humanitarian Emergencies and USAID’s Humanitarian Response
tions in external military and other per-            e. NGOs dedicated to specific benefi-
sonnel. Monitoring such effects should be      ciaries and discrete activities (traumatized
initiated right from the start.                women, youth voluntarism, sports, fam-
                                               ily therapy, interethnic reconciliation, cul-
     5. NGO experience. Benefit from the ex-   tural revival) should not be pressed into
perience of nongovernmental organizations.     new activities in which they lack exper-
                                               tise (e.g., microenterprise) merely because
    NGOs play a major role in complex hu-      donor priorities have shifted. Instead, do-
manitarian emergencies, and lessons from       nors should encourage the creation of
their experience should be identified, es-     groups of NGOs (“strategic” rather than ad
pecially in Bosnia and Croatia. Among          hoc) with complementary capabilities and
those lessons:                                 objectives suitable for the array of prob-
                                               lems that need attention in specific com-
      a. NGOs emerging in response to the      munities.
unprecedented availability of funds for
civil society often have only shallow ex-         f. Financial support for local NGOs
perience; they should be encouraged with from the ethnic Albanian Diaspora can fa-
small grants for an initial testing period. cilitate the gradual phaseout of aid depen-
                                              dence, but political capture of such remit-
      b. Technical assistance is often essen- tances can create disillusionment and
tial; it can bring even small NGOs to the choke off these funds, as demonstrated in
point of effective management.                Croatia.

     c. Most Bosnian and Croatian NGOs            g. Local staffs of the International Res-
were founded and managed by women—           cue Committee and other strong NGOs in
teachers, mental health workers—re-          Bosnia and Croatia could be valuable re-
sponding to the emergency need to assist     sources for aid-funded projects utilizing
displaced families and then to promote       (or assisting) Kosovar (or other) local
resettlement, normalization, and commu-      NGOs. For example, the experience of agen-
nity reconciliation.                         cies helping reconstruct housing in Bosnia
                                             (such as Mercy Corps International) also
     d. Youth and women often appear may be valuable in Kosovo. Finally,
more ready than adult males to reconcile Kosovar NGOs that may be intact in refu-
across ethnic lines. If this is also true in gee status, outside Kosovo, should be sup-
Kosovo (assuming postconflict Kosovo is ported to ensure their survival pending
not monoethnic), the bias toward reach- return.
ing youth and women and neglecting
adult males needs correction for reconcili-       6. Safety nets. Anticipate a need for aid-
ation to work.                               ing widow-headed households.

Annex D: Implications for the Kosovo Crisis                                             D5
     When the fate of the large number of           Purchasing aid-funded goods and
missing Kosovar males becomes known,           services internally and from the affected
Kosovo may join the ranks of other coun-       neighboring countries (with an eye to the
tries that have emerged from complex           possible price effects, as noted above) can
emergencies with tragically distorted de-      contribute to regional employment, eco-
mography. A major loss of adult males          nomic recovery, and the restoration of
could create a severely disadvantaged          cross-border economic relations. This may
group of widow-headed households, es-          require flexibility in the application of pro-
pecially for Kosovo’s large rural popula-      curement regulations in the affected coun-
tion. The need for effective safety nets and   tries.
the need to restore such households to eco-
nomic viability should be anticipated.              9. High-level technocrat training.
Other disadvantaged groups needing tai-        Training technocrats to manage the
lored support may include the elderly,         postconflict economic transition should begin
widower heads of households, single            immediately.
mothers, and the disabled.
                                                   Reconstruction and development in
     7. Employment-oriented training. Kosovo (under non-Yugoslav administra-
Train people in skills for which there is em- tion) is likely to be associated with a new
ployment demand and link training with set of economic institutions and rules of
startup capital.                              the game that mirror Western European
                                              norms. As with many complex humani-
     Recovery activities need to match tarian emergencies, overall macroeco-
postconflict economic realities. Job train- nomic and sectoral policy formulation
ing for youth, women, and demobilized during recovery-cum-transition is likely to
soldiers (who have often turned to desta- require broad conceptual and manage-
bilizing criminal activity when they re- ment skills. Those skills are likely to be in
mained unemployed) is virtually fruitless short supply, especially if the preconflict
if unemployment in the depressed professional and intellectual cadre have
economy remains high. Although oppor- been targeted for deliberate elimination.
tunities for self-employment in microen-
terprise may also be limited, they are likely      Although Kosovo’s civil society had
to be enhanced if the training is linked to developed considerable institutional expe-
the provision of seed capital.                rience (mainly in social and cultural sec-
                                              tors) during the autonomy period of 1974–
     8. Commodity procurement. Maxi- 89, the fate of many civil society leaders is
mize regional procurement to help stimulate unknown. If possible these cadre should
economic recovery.                            be found, perhaps among the refugees,
                                              and readied for return to Kosovo to help

D6                Complex Humanitarian Emergencies and USAID’s Humanitarian Response
revive the social infrastructure and prepare           In an evaluation of its own
for economic recovery. A balance will be          postconflict experience, the World Bank
needed between foreign implementing               asserted that donors could have and
agencies and contractors on the one hand          should have exerted more forcefully the
and Kosovar organizations on the other.           leverage they had during the initial
Otherwise, the recovery may be jeopar-            postconflict period. This is typically a pe-
dized either by overburdening the                 riod of almost total dependence on exter-
Kosovars or by sinking them with over-            nal financing and security. The study re-
bearing disregard.                                fers to governance functions within the
                                                  normal scope of the Bank’s mandate (e.g.,
Political Effects                                 fiscal practices). The point applies as well,
                                                  though, to the notion of peace condition-
      1. Potential political consequences.        ality—that is, making provision (or with-
Monitor emergency aid distribution for pos-       holding) of aid flows dependant on local
sible unintended and undesirable political con-   authorities’ adhering to the political com-
sequences.                                        mitments in the peace accords (or compa-
                                                  rable instruments). Lack of donor consis-
     Emergency aid (especially food com-          tency in this regard can undermine such
modities) are a source of power. A flawed         potential (Cambodia), which reinforces the
distribution process can have unintended          need to design an effective coordination
and undesirable political consequences. In        structure. The Bosnian experience with
several cases (especially Rwanda), control        peace conditionality is perhaps the most
over final food distribution reinforced the       pertinent for postconflict Kosovo.
power of local authorities or factions,
strengthened their relative position, and              3. Regional destabilization. Non-
facilitated their self-aggrandizing, often        project aid for budget support can help front-
abusive, behavior toward the intended             line states maintain domestic stability.
noncombatant beneficiaries. Conse-
quences of this sort would not be surpris-            There is concern that the Kosovo cri-
ing under prevailing conditions in Kosovo        sis may destabilize Albania and Mace-
and neighboring countries. Reconstruction        donia. The presence of Kosovar refugees
aid in the Balkans will far exceed emer-         and the burdens they impose jeopardize
gency food aid in amount and potential           the ability of the two governments to con-
political consequences. It also will need        tinue to finance and sustain precrisis lev-
careful monitoring.                              els of service delivery to their citizens. Aid
                                                 in the form of nonproject budget support
     2. Peace conditionality. Use aid as le- could help them sustain domestic budget
verage to enforce adherence to the peace accords outlays. It could also help diminish the po-
and responsible governance.                      tential for ethnic polarization in

Annex D: Implications for the Kosovo Crisis                                                 D7
Macedonia (e.g., by enabling the govern-         Political problems have arisen in host
ment to maintain pensions and civil ser-    countries when encampment extends for
vice salaries) and the deepening of the so- some considerable period. This may well
called left–right polarization in Albanian  be the case with the Kosovar refugees
politics. Fast-disbursing aid flows may     owing to mines, housing destruction, and
                                            other problems. (One recent estimate as-
also be critical for sustaining the positions
of these governments regarding the con-     sumed 3–5 years, security considerations
flict itself, a political point beyond the  aside.) Evaluations have stressed that refu-
scope of this note. Understandings and      gees should not be supported at a stan-
commitments regarding allocation of the     dard that evokes resentment among the
local currency counterpart should be ex-    host population. Resentment can also stem
plicit and should be monitored to ensure    from refugees competing for scarce local
compliance. Disbursements in tranches       employment. After resettlement, resent-
should be considered to encourage com-      ment could arise if country allocations of
pliance.                                    reconstruction aid are perceived as unfair
                                            or unjust. Aid to refugees should therefore
     4. Resentment of refugees. To avoid conform with regional living standards.
resentment, keep refugee support standards
modest in relation to host-population stan-

D8                Complex Humanitarian Emergencies and USAID’s Humanitarian Response

Africa Rights. 1998. Rwanda: The Insurgency      Human Emergencies.” Journal of Hu-
   in the Northwest. London: Africa Rights.      manitarian Assistance (23 October).

Anderson, Mary B. 1996. Do No Harm: Sup- Brown, D.G.; P.W. Bartholomew; J–C.
   porting Local Capacities for Peace Through     Lucas; and M. Jolivert. 1996. “Final
   Aid. Cambridge: Collaborative for De-          Evaluation: USAID/GOH-funded JOBS
   velopment Action.                              Creation Program.” Prepared for Pan
                                                  American Development Foundation
Apthorpe, Raymond. 1997. “Some Relief             and USAID/Haiti.
   from Development: Humanitarian
   Emergency Aid in the Horn of Africa Buttari, Juan J. 1997. “Haiti: Intervention,
   (Including Sudan), Rwanda, and                 Reform, and Politics.” Draft report.
   Liberia.” The European Journal of Devel-
   opment Research, vol. 9, no. 2 (Decem- Chandler, Dennis; and others. 1996.
   ber): 83–106.                                  “Evaluation of the Haiti Communal
                                                  Governance Program.” Management
Borton, John. 1994. “Assessing the Impact         Systems International. Washington:
   and Effectiveness of Relief Assistance.”       USAID.
   Paper prepared for the 1995 World
   Disasters Report.                          Collier, Paul; and Anke Hoeffler. 1998. “On
                                                  Economic Causes of Civil War.” Oxford
Borton, John; and Joanna Macrea. 1997.            Economic Papers 50. Oxford University
   “Evaluation Synthesis of Emergency             Press.
   Aid.” London: Department for Inter-
   national Development.                      de Soysa, Indra; and Nils Petter Gleditsch.
                                                  “To Cultivate Peace: Agriculture in a
Brandt, Donald P. 1995. “War Work: Hu-            World of Conflict.” Oslo: International
   manitarian Assistance in Complex               Peace Research Institute.
de Waal, Alex. 1993. “War and Famine in        ban Disaster?: Dilemmas of Contem-
   Africa.” IDS Bulletin, vol. 24, no. 4.      porary Post-War Reconstruction. Cam-
                                               bridge, Mass.
Duffield, Mark. 1998. “Aid Policy and
  Post-Modern Conflict: A Critical Re- Ianotti, L. 1997. “Nutrition in Haiti: Con-
  view.” School of Public Policy. Bir-    straints and Solutions.” Washington:
  mingham: University of Birmingham.      Inter-American Development Bank.

Dupuy, Alex. 1997. Haiti in the New World Inter-American Development Bank. 1994.
  Order: The Limits of the Democratic Revo-   “Emergency Economic Recovery Pro-
  lution. Boulder, Colo.: Westview Press.     gram: Haiti.” Report of the Joint As-
                                              sessment Mission. Washington.
Government Accounting Office. 1985.
  “The United States’ Response to the Joint Evaluation of Emergency Assistance
  Ethiopian Food Crisis.” Washington.         to Rwanda. 1996. “Early Warning and
                                              Conflict Management.” Study 2. The
———. 1986. “Famine in Africa: Improv-         International Response to Conflict and
  ing Emergency Food Relief Programs.”        Genocide: Lessons from the Rwanda Ex-
  Washington.                                 perience.

Hallam, Alistair. 1998. “Evaluating Hu- ———. 1996. “Humanitarian Aid and Ef-
   manitarian Assistance Programmes.”        fects.” Study 3. The International Re-
   Prepared for the DAC Expert Group on      sponse to Conflict and Genocide: Lessons
   Aid Evaluation and reviewed by the        from the Rwanda Experience.
   DAC Task Force on Conflict, Peace, and
   Development Cooperation, February Kitay, Michael. N.d. “Land Tenure Issues
   2–3.                                      in Postconflict Countries.” Washing-
                                             ton: USAID.
Hansch, Steve; and others 1994. “Lives
   Lost, Lives Saved: Excess Mortality
   and the Impact of Health Interventions Kleist, Ted. See Netherlands Development
   in the Somalia Emergency.” Washing-       Cooperation. 1994.
   ton: Refugee Policy Group.
                                          Lautze, Sue; Bruce D. Jones; and Mark
Hill, Richard. 1997. “Post–Cold War Re-      Duffield. 1998. Strategic Humanitarian
   construction Programs: Observations       Coordination in the Great Lakes Region
   From the Field on the Sword and the       1996–97. Office for the Coordination of
   Ploughshare.” Paper presented at a        Humanitarian Affairs. New York:
   Conference on Urban Triumph or Ur-        United Nations.

Biblio2         Complex Humanitarian Emergencies and USAID’s Humanitarian Response
Lieberson, Joseph; and others. 1999. “Pro- ———. 1994b. “The International Relief
   viding Emergency Aid to Mozam-              System: A Critical Review.”
   bique.” Center for Development Infor-
   mation and Evaluation. Washington: ———. 1998. “Learning to Learn.” Discus-
   USAID.                                      sion Paper prepared for a seminar on
                                               Lessons Learned in Humanitarian Co-
Maguire, Robert; and others. 1996. Haiti       ordination. Office for the Coordination
   Held Hostage: International Responses to    of Humanitarian Affairs and the Min-
   the Quest for Nationhood, 1986 to 1996.     istry of Foreign Affairs of Sweden.
   Occasional Paper no. 23. Providence:
   Watson Institute.                        Nafziger, E. Wayne. 1996. “The Econom-
                                               ics of Complex Humanitarian Emer-
McClelland, Donald G.; and others. 1999.       gencies: Preliminary Approaches and
   “Providing Emergency Aid to Haiti.”         Findings.” UNU/WIDER.
   Center for Development Information
   and Evaluation. Washington: USAID.       Naval, G. 1995. “Evaluation of the Crisis
                                               Program of the Catholic Relief Services
McHugh, Heather S.; and Joshua M.              in Haiti.” Final report.
  Kaufman. N.d. “A Summary of Prin-
  cipal Lessons Learned from USAID Ex- Netherlands Development Cooperation.
  perience in Humanitarian Assistance.”   1994. Humanitarian Aid to Somalia.
  Washington: USAID.                      Evaluation Report directed by Ted
                                          Kleist. Operations Review Unit.
Messer, Ellen; Marc J. Cohen; and Jashinta
  D’Costa. 1998. “Food from Peace: Organization for Economic Cooperation
  Breaking the Links Between Conflict      and Development. 1999. “Guidance for
  and Hunger.” Washington: Interna-        Evaluating Humanitarian Assistance
  tional Food Policy Research Institute.   in Complex Emergencies.” Develop-
                                           ment Assistance Committee. Paris.
Miller, John M. 1997. “USAID/Washington
   and Field Collaboration in the Use of Organization for Economic Cooperation
   Central Resources.” Washington.          and Development. DAC/o Database
Minear, Larry. 1994a. “Humanitarian Ac-
   tion in the Former Yugoslavia: The Pan American Development Foundation.
   UN’s Role 1991–93.” Providence R.I.:  N.d. “Jobs Creation Project: Final Re-
   Watson Institute.                     port.” Washington.

Bibliography                                                                  Biblio3
Pasic, Amir; and Thomas G. Weiss. 1997. ———. 1995. “Survey Report on Losses of
   “Humanitarian Recognition in the           Public Law 480 Commodities in Haiti
   Former Yugoslavia: The Limits of Non-      During the Political Transition Period
   State Politics.” Security Studies 7, no. 1 September 15, 1994, to January 15,
   (autumn 1997): 194–227.                    1995.” Audit Report No. 1–521–95–008.

Prendergast, John; and Colin Scott. 1996. ———. 1998a. Agency Performance Report,
   “Aid with Integrity: Avoiding the Po-     1997. Center for Development Infor-
   tential of Humanitarian Aid to Sustain    mation and Evaluation. Washington.
   Conflict: A Strategy for USAID/BHR/
   OFDA in Complex Emergencies.” ———. 1998b. “Results Review and Re-
   Occasional Paper no. 2. Washington.       source Request: Fiscal Year 2000.” Bu-
                                             reau for Humanitarian Response. Pre-
Prunier, Gérard. 1995. The Rwanda Crisis:    liminary draft.
   History of a Genocide. New York: Colum-
   bia University Press.                   ———. 1999. Agency Performance Report,
                                             1998. Center for Development Infor-
Renison, Bill; and others. 2000. “Humani-    mation and Evaluation. Washington.
   tarian Assistance in Rwanda: Hard
   Choices, No Exits.” Draft Report. USAID/Haiti. 1992–96. “USAID Monitoring
   Washington: USAID.                        Report” 1–5. Port-au-Prince.

Sen, Amartya K. 1993. “The Political USAID/Liberia. 1998. “Fiscal Year 2000 Re-
   Economy of Hunger.” Overcoming Glo-         sults Review and Resource Request.”
   bal Hunger. Ismail Serageldin and           Bureau for Humanitarian Response.
   Pierre Landell–Mills, eds. Proceedings
   of a Conference on Actions to Reduce U.S. Mission to the United Nations. 1997.
   Hunger Worldwide. Washington.               “Global Humanitarian Emergencies,
UNHCR /WFP. 1998. “Joint Evaluation of
   Emergency Food Assistance to Return- ———. 1998. “Global Humanitarian
   ees, Refugees, Displaced Persons, and       Emergencies, 1998–99.”
   Other War-affected Populations in
   Bosnia and Herzegovina.” Rome.           Uvin, Peter. 1998. Aiding Violence: The De-
                                               velopment Enterprise in Rwanda. West
USAID. Congressional Presentation, FY 1990–    Hartford: Kumarian Press.
   FY 2000. Various years. Washington.

Biblio4          Complex Humanitarian Emergencies and USAID’s Humanitarian Response
World Bank. 1998a. “Haiti: The Challenges ———. 1998c. “Rwanda: Poverty Update.”
  of Poverty Reduction,” vols. I and II.    Washington.
  Technical papers. Washington.
                                          World Food Program. 1998. “The Food Aid
———. 1998b. “Haiti: The World Bank’s        Monitor.” 1997 Food Aid Flows. Spe-
  Experience With Postconflict Recon-       cial issue. Rome: International Food
  struction,” vol. v. Desk Reviews of       Aid Information System.
  Cambodia, Eritrea, Haiti, Lebanon,
  Rwanda, and Sri Lanka. Report no.
  17769. Washington.

Bibliography                                                              Biblio5