62032



WORLD DEVELOPMENT REPORT 2011
BACKGROUND PAPER




FINANCING PEACE:
INTERNATIONAL AND NATIONAL RESOURCES FOR
POSTCONFLICT COUNTRIES AND FRAGILE STATES
James K. Boyce and Shepard Forman


October 6, 2010




The findings, interpretations, and conclusions expressed in this paper are entirely those of the
authors. They do not necessarily represent the views of the World Development Report 2011
team, the World Bank and its affiliated organizations, or those of the Executive Directors of the
World Bank or the governments they represent.



 James K. Boyce is Professor of Economics at the University of Massachusetts and Director of the Program
on Development, Peacebuilding, and the Environment at the Political Economy Research Institute (PERI).
Shepard Forman is Director Emeritus and Senior Fellow of the Center on International Cooperation at
New York University.
                                          1. Introduction

After more than a decade of experience and research on financing arrangements in
postconflict countries and fragile states, a consensus has emerged on at least one matter:
the core objective is to build effective and legitimate governance structures that secure
public confidence through provision of personal security, equal justice and the rule of
law, economic well-being, and essential social services including education and health.
These governance structures are necessary to ensure that countries do not turn, or turn
back, to violence as a means of negotiating state-societal relations.

Building institutional capacity and providing critical public goods and services in these
settings requires financial resources. In many cases, these resources initially must come
in substantial measure from international sources. In the long run, however, external
resources must be replaced by domestic resources Therefore, a key task is to build the
national fiscal capacities and tax base that are necessary for a durable peace.

This understanding represents a shift from earlier years when peace-making,
peacekeeping and peacebuilding were treated as discrete activities; political, security, and
development objectives were pursued as if each was independent of the other; and
conflict prevention was, to put matters bluntly, nobody’s business.

Much remains to be done to translate this new consensus into effective policies, but its
basic principles are now well-established in the normative frameworks of the United
Nations,1 the World Bank,2 and bilateral donors as distilled in the OECD Principles for
Good International Engagement in Fragile States and Situations.3

While an integrated approach to early recovery and political stabilization is becoming the
norm, replacing the older notion of a sequential movement – starting with security, and
moving through humanitarian assistance, infrastructure recovery, and justice and
reconciliation, to development – critical sectors continue to be underfunded. Too little is
invested in effective capacity building in recipient countries, and for that matter in
building relevant capacities within bilateral and multilateral donor agencies themselves.
Too little is invested in building national fiscal capacities; among other things, too few
resources are channeled through national budgets and too little attention is paid to
building civil society capacities to monitor government finances in the interest of
ensuring transparency and accountability. Too little is invested in employment generation
and sustainable livelihoods, notably in rural areas.

On top of these substantive weaknesses are institutional ones. The current architecture of
international finance for aid to postconflict and fragile states militates against the timely,
predictable and effective delivery of resources. It lacks tracking, monitoring and
evaluation mechanisms that would be necessary for accountability among donors,

1
  UN S/2009/304 Report of the Secretary-General on peacebuilding in the immediate aftermath of conflict.
2
  Zoellick (2009).
3
  OECD (2007).

                                                   1
recipient governments, and their constituents. It is not well-suited to adapt to changing
circumstances and opportunities. And it lacks the capacity to formulate coherent policies
to address the regional dimensions of conflicts.

This paper discusses a number of the weaknesses in current financing arrangements for
postconflict countries and fragile states, with a focus on official development assistance
(ODA). We argue that tensions persist between business-as-usual development policies
on the one hand and policies responsive to the demands of peacebuilding on the other.
The preferential allocation of aid to “good performers,” in the name of maximizing its
payoff in terms of economic growth, militates against aid to fragile and conflict-affected
states. If the aim of aid is redefined to include durable peace, the conventional
performance criteria for aid allocation lose much of their force. Compelling arguments
can be made for assistance to “poor performers” if this can help to prevent conflict.

Yet the difficulties that initially prompted donors to become more selective in aid
allocation remain all too real. Experience has shown that aid can exacerbate problems
rather than solving them. The donor move to selectivity came in response to
accumulating evidence that in some contexts aid has perverse effects on economic
performance. The same dilemma arises when aid is assessed in terms of its impact on
peace and conflict: sometimes aid helps to prevent conflict and build peace, but
sometimes it has the opposite effect. In responding to this dilemma, simply retreating
from intractable problems in difficult settings does not offer a solution. The challenge is
to devise policies that make the problems more tractable.


                                          2. Literature Review

The OECD’s Principles for Good International Engagement in Fragile States &
Situations define states as fragile when “state structures lack political will and/or capacity
to provide the basic functions needed for poverty reduction, development and to
safeguard the security and human rights of their populations” (OECD 2007, p. 2). This
formulation suggests that states may have the capacity to provide these basic functions
but lack political will to do so; that they may have political will but lack capacity; or that
they may lack both.4 A subsequent OECD/DAC discussion paper adds a third dimension




4
  The dual criteria of will and capacity are adopted by other authors, as well. For example, in proposing an
“index for state weakness,” Rice and Patrick (2008, p. 3) define “weak states” as lacking “the essential
capacity and/or will to fulfill four sets of critical government responsibilities: fostering an environment
conducive to sustainable and equitable economic growth; establishing and maintaining legitimate,
transparent, and accountable political institutions, securing their populations from violent conflict and
controlling their territory, and meeting the basic human needs of their population.” Similarly, in a working
paper for the UK Department for International Development, Moreno Torres and Anderson (2004, p. 12)
similarly refer to both political will and capacity in defining “difficult environments” as settings where “the
state is unable or unwilling to harness domestic and international resources effectively for poverty
reduction.”


                                                       2
to the definition: legitimacy, or the ability of the state to meet its population’s
expectations (OECD 2008, p. 16).5

Postconflict states are often fragile, and fragile states are often at risk of violent conflict.
But the risk is especially high in postconflict settings. Describing this “conflict trap,”
Collier et al. (2003, p. 107) concluded that the risk of conflict is five times greater in
postconflict countries than in the “typical marginalized country that has not had a war for
10 years.” Peace agreements can turn out to be little more than ceasefires.

Public finance in the state-society contract

The term “statebuilding” is used in the literature to describe efforts to move from fragility
to a situation where states have the political will and capacity to provide the basic
functions to advance the economic well-being and safeguard the physical security of their
populations, thereby gaining legitimacy and reducing the risk of violent conflict.
“Peacebuilding” – the construction of a durable peace based in many cases in formal
peace accords – is statebuilding in settings where the task is complicated and jeopardized
by the legacy of conflict (Call and Wyeth 2008).

The twin goals of building resilient states and durable peace are increasingly understood
to hinge on the relationship between states and their citizens. A robust relationship is
sometimes termed the “state-society contract.” This understanding represents a break
from state-centric formulations that focus on capacity alone, as reflected, for example, in
the following statement: “The question of whether security will be provided in a way that
meets the needs of citizens, or will function primarily as an instrument of oppression, will
not be dictated by capacity, but shaped – indeed, often usefully constrained – by the basic
political process of state-society contract formation and reformation” (OECD 2008, p. 8).

Public finance – domestic revenue mobilization, budget allocation, and expenditure
management – is the circulatory system of the state-society contract. Recent literature on
peacebuilding and statebuilding has begun to make the public-finance connection (see,
for example, Boyce and O’Donnell 2007; Brautigam, Fjeldstad and Moore 2008; FIAS
2009), but its intellectual pedigree is long and deep. The OECD’s fragility triad –
capacity, will, and legitimacy – mirrors very closely “three reasons to give taxation
particular attention” enumerated by the historical sociologist Charles Tilly (2009, p. xiii):

         First, over the long run it constitutes the largest intervention of governments in
         their subjects’ private life, so much so that the history of state expansion becomes

5
  The criterion of legitimacy has also been invoked by other authors. For example, Rotberg (2004, p. 1)
writes: “Nation-states fail when they are consumed by internal violence and cease delivering positive
political goods to their inhabitants. Their governments lose credibility, and the continuing nature of the
particular nation-state itself becomes questionable and illegitimate in the hearts and minds of its citizens.”
In a similar vein, Ghani and Lockhart (2008, p. 126) write that “citizens provide the ultimate source of
legitimacy for a social order,” and Picciotto (2008, p. 9) characterizes fragility as a situation in which “the
society is fractured, the economy is mismanaged and social service delivery is so weak that the social
contract between the state and the people has been undermined or has broken down altogether.”


                                                       3
          a history of violent struggle over taxes, and the history of state consolidation
          becomes a history of tax evasion by those who have the guile and power to
          frustrate the fisc. Second, follow the money: the circulation of resources from
          subjects to government-initiated activities provides a sort of CT scan for a
          regime’s entire operation. Third, it dramatizes the problem of consent.

International financing can impact all three dimensions of fragility – capacity, will, and
legitimacy – for better or for worse. The OECD’s Principles for Good International
Engagement in Fragile States & Situations recognizes, in its preamble, that “international
actors can affect outcomes in fragile states in both positive and negative ways” (OECD
2007, p. 1). In other words, it is not only the quantity of international assistance to fragile
states and postconflict countries that matters, but also the qualities that determine the sign
and magnitude of its impact on peace and state resilience.

International financing

Very substantial international financial resources – and human resources – are now
devoted to postconflict peacebuilding operations. To cite one indicator, UN troop levels
rose by 600% from 2002 to 2008 (OECD 2008, p. 28). Direct international provision of
security in postconflict states now engages more than 170,000 troops, a phenomenon that
has been described as “a massive shift in global governance arrangements” (OECD 2008,
p. 28). Official development assistance, the focus of this paper, is an important part of
this international engagement.

Some literature has pointed to a pronounced disparity, however, between the generous
international resources directed to postconflict states and the relatively few resources
directed to other fragile states that are merely at risk of violent conflict. Analyzing aid
flows to “difficult partnership countries,” an earlier euphemism for fragile states, Levin
and Dollar (2005) estimated that these countries received 58% less bilateral aid and 34%
less multilateral aid than their non-fragile counterparts, when controlling for differences
in population, per capita income, and CPIA (World Bank Country Policy and Institutional
Assessment) scores; without controlling for CPIA scores the difference would be greater.
Yet within this group, postconflict countries were an exception, receiving more aid per
capita than the average for low-income countries as a whole, a finding the authors rather
disarmingly characterize as supporting the “hypothesis that donors increasingly pay
attention to post-conflict opportunities” (ibid., pp. 14-15).

This disparity between the international financing devoted to postconflict and other
fragile states is poignantly captured in the words of an ambassador from a fragile but not
postconflict state: “Why do we have to wait until we have a war to get help with the
transformation of our justice system or our military?”6

These generalizations conceal significant differences, however, within the postconflict
and non-postconflict subsets of fragile states. The OECD Principles conclude with the

6
    Quoted in OECD (2008, p. 27).


                                                4
admonition to donors to “avoid pockets of exclusion,” citing the problem of “aid
orphans,” states (as well as sectors and groups within societies) where aid volumes
remain low even in the absence of significant political barriers to engagement (OECD
2007, p. 3). Noting the disparities between “darlings” and “orphans,” Levin and Dollar
(2005, pp. 18, 25) observe that aid orphans are concentrated in Africa, particularly
francophone Africa, and that their per capita aid flows are not only lower but also more
volatile. These disparities cannot be explained by differences in needs; instead the
explanation must be sought in the various motives of aid donors – which include
geopolitical considerations and commercial interests in addition to humanitarian,
developmental, and peacebuilding/statebuilding goals – and to the “strategic dilemmas”
that arise when these objectives are mutually contradictory (Paris and Sisk 2007; OECD
2010a).

Apart from these inter-country allocation issues, concerns have been raised in the
literature about the timing of aid disbursements in postconflict countries (Forman and
Patrick 2002). Collier and Hoeffler (2004) observed that aid levels tend to be high in the
first postconflict years but soon decline, and suggest that this pattern does not match
absorptive capacity – the country’s ability to translate aid into economic growth – which
may move in the opposite direction. This argument is echoed in the World Bank
publication Breaking the Conflict Trap (Collier et al. 2003, pp. 157-159). Kang and
Meernik (2004) find a similar pattern but suggest that this is responsive to immediate
postconflict needs. Suhrke and Buckmaster (2005), in contrast, offer evidence that this
pattern is by no means universal, and point to the importance of distinguishing between
commitments (when aid is pledged) and disbursements (when the money is actually
spent). In addition, Suhrke, Villanger and Woodward (2005) dispute the Collier-Hoeffler
conclusion that aid has significantly stronger impacts on growth in the period four to
seven years after peace, on the grounds that this finding is sensitive to changes in sample
selection and econometric specification.

The sectoral allocation of aid within postconflict countries and fragile states has also been
subject to occasional scrutiny and questions. In an analysis of aid to postwar El Salvador,
Boyce (1996, pp. 130-135) reported discrepancies between peace implementation
priorities, as defined by the government and United Nations, and donor priorities: high-
priority peace programs – including the National Civilian Police, the land transfer
program for ex-combatants, and judicial and democratic institutions – remained under-
funded, while donors directed more resources to lower-priority needs such as physical
infrastructure. A recent OECD study of statebuilding assistance in six case-study
countries observes that agriculture has received a small percentage of bilateral assistance,
decreasing over time, despite the crucial role of this sector in livelihoods and economic
growth (OECD 2010a, pp. 109-110).

Resolving allocation issues and achieving policy coherence in international financing is
made difficult by lack of coordination among dozens of donor agencies, despite
widespread recognition of its importance (Forman, Patrick and Salomons 2000). The
1997 OECD/DAC Guidelines on Conflict, Peace and Development Co-operation called
upon donors to “attempt to formulate and agree on a common integrated strategic

                                             5
framework” that would “provide a guide for prioritising resource allocations across
sectors and geographical areas, determining the division of labour amongst actors, and
defining common approaches towards key policy issues” (OECD 1997, pp. 20-22). To
say that this is easier said than done would be an understatement. The latest effort to
tackle this problem is the “integrated peacebuilding strategy” process initiated by the UN
Peacebuilding Support Office with support from the World Bank.

In the meantime, multi-donor trust funds (MDTFs) have emerged as one way to address
the coordination problem (Schiavo-Campo 2003; Leader and Colenso 2005), but they
represent only a small share of total aid flows. Some individual donor governments have
experimented with “whole-of-government” approaches in an effort to resolve
coordination problems among their own agencies. Apart from the attendant risk that the
bureaucratic balance-of-power within donor governments will elevate short-term security
priorities above long-term statebuilding, these efforts carry the risks that they will further
impede coordination among donors and efforts of recipient governments to forge their
own whole-of-government strategies (OECD 2008, pp. 45-46).

National finance

The literature on the impact of international financing on domestic revenues in
developing countries in general, and in fragile states in particular, has two main prongs.
One examines whether aid “crowds out” or “crowds in” domestic revenues. Findings
have been mixed (see Section 6 below), suggesting that outcomes are conditional on
context and perhaps policy choices.

The second prong asks whether the tax policy prescriptions of the donors – a centerpiece
of which has been a shift from tariffs to VAT – have had desirable effects. Baunsgaard
and Keen (2005) find that trade tax reductions have had a significant negative effect on
total revenue in low-income countries, with growth in other revenues making up only
30% of the resulting losses. In fragile states, where the tax/GDP ratio is often lower than
in other low-income countries, and expenditure needs (particularly in postconflict
countries) are often higher, the mismatch between revenue needs and tax policy
prescriptions has been especially striking (for discussion, see Boyce 2002, Di John 2006,
Boyce and O’Donnell 2007, and Fjeldstad and Moore 2008).

In countries with substantial revenues from extractive resources, such as petroleum and
minerals, a related set of issues arises: how to ensure that these revenues do not
undermine the state-society relationship by freeing the state from the need to enter into
the fiscal compact that ties taxation to its delivery of public goods and services. By
weakening capacities in “rentier states,” resource rents can contribute to state fragility
(Karl 1997, Fearon and Laitin 2003). The examples of Botswana and Malaysia, countries
where resource rents have supported state capacities rather than undermining them,
suggest that this outcome is not inevitable (FIAS 2009). As Di John (2008, p. 22)
remarks, “factor endowments do not determine politics.”




                                              6
On the expenditure side of national fiscal capacity, the literature again cautions that
international assistance can have crowding-out effects. Aid to fragile states and
postconflict countries generally bypasses the government budget to a greater extent than
aid to “good performers” (Leader and Colenso 2005, pp. 11-15; OECD 2009b, Table
A.9). Apart from the opportunity cost of foregone development of state capacities to
allocate resources and manage expenditures, the resulting “dual public sector” – one
national, the other international – may diminish state capacities below what they would
have been in the absence of aid, as human resources are siphoned off by international
agencies and government administrative time is diverted into coping with the attendant
coordination problems (Ghani et al. 2007; Birdsall 2007; Boyce 2008; OECD 2010a).
“The largest adverse impact of the aid system,” in the view of Ghani and Lockhart (2008,
p. 100), “has been the undermining of a country’s budget as the central instrument of
policy.”

Recommended policies to address this problem are context-specific. Where fragility takes
the form of adequate political will but inadequate capacity, increasing the share of aid
that is channeled through the budget, with technical assistance and “dual control” systems
to minimize leakage and corruption, is seen as desirable (see, for example, Leader and
Colenso 2005, pp. 22-23; Boyce and O’Donnell 2007, pp. 287-289; OECD 2008, pp. 40-
41). Where the state lacks the political will to advance the economic well-being and
protect the physical security of the population, efforts to work “around the state” have
more appeal, although as Leader and Colenso (2005, p. 51) observe, “governments in
fragile states are rarely homogeneous, and reformers and pro-poor constituencies will
exist both within and without the state,” creating possible openings for mixed strategies.

Galtung and Tisné (2009) emphasize the potential role of “social-accountability
mechanisms” whereby communities and citizen-based groups provide “monitoring from
below,” noting that this has been successful in a number of developing countries albeit
seldom in countries recovering from war. An example is the Poverty Action Fund
established in Uganda in 1998, in which 5% of the fund is devoted to monitoring by civil
society organizations (Krafchik 2001; Ndikumana and Nannyonjo 2007). Such initiatives
could be facilitated by greater transparency on the part of aid donors themselves, an
objective of the recently launched international “Publish What You Fund” campaign
(http://www.publishwhatyoufund.org/).7

Insofar as international financing stimulates local economic activity, this may indirectly
advance the fiscal dimension of statebuilding by expanding the country’s tax base. Local
procurement of goods and services, as opposed to external procurement, serves this goal.
This opportunity is often neglected by donors, however, as illustrated in a study on the
economic impact of peacekeeping operations in eight countries by Carnahan, Durch and
Gilmore (2006). Durch (2009) reports that local content on average represented only 6%
of peacekeeping mission procurement in the years 2004-2008; he concludes that for
peacekeepers, their contribution to the host economy “is largely incidental to doing their

7
 For a discussion of tensions between “bottom-up” and “top-down” approaches to the measurement of
government performance, see Radin (2007).


                                                  7
main job, yet what is incidental to them may be critical to those whose goods and
services may be utilized by these sometimes-large civil-military operations” (ibid., p. 15).

Finally, the ability of the state to fulfill the service delivery side of the state-society fiscal
contract is affected by the extent to which tax revenue must be devoted instead to
external debt service, a legacy of past international financing (Alvarez-Plata and Brück
2007). These debts – sometimes repackaged as new “emergency credits”8 – drain scarce
resources away from the needs of the population. If the original loans provided
commensurate benefits to the country’s people, this could be argued to justify the cost.
But there is econometric as well as case-study evidence that a substantial fraction of past
borrowing financed the accumulation of private external assets through capital flight.9


                               3. Trends in International Financing

This section reviews trends in international assistance to postconflict and fragile
countries.10 First we discuss the magnitude and timing of postconflict aid; then we
discuss the allocation of aid between postconflict countries and fragile yet non-
postconflict countries; finally we discuss sectoral priorities in the allocation of aid.

Postconflict aid: Magnitudes and timing

Postconflict states account for a substantial percentage of official development assistance
(ODA) worldwide. Trends in their share of total aid disbursements since the end of the
cold war in 1989 are shown in Figure 1, based on two alternative definitions of what
constitutes a postconflict country: (i) “formal peace,” defined as the period up to 10 years
after the signing of a peace accord, as long as conflict does not resume; and (ii) “negative
peace,” defined as the period up to 10 years after battle-deaths fall below the level of
1,000/year, regardless of whether or not this was accompanied by a peace accord, again
as long as conflict does not resume.11 Although the percentages have declined in the last
five years, postconflict countries by the formal-peace definition continue to receive
almost 10% of total ODA; by the negative peace definition their share exceeds 15%.

[INSERT FIGURE 1 HERE]

8
 For example, the IMF’s emergency credit window for postconflict countries has provided loans to repay
bridge loans that in turn were used to clear previous IMF arrears (Boyce 2002, p. 67).
9
  For a case study of the DRC, see Ndikumana and Boyce (1998); for an econometric analysis of the
relationship between external borrowing and capital flight in sub-Saharan Africa, see Ndikumana and
Boyce (2010).
10
  In this paper our focus is official development assistance (ODA). For discussion of other important
elements of the external resource envelope in post-conflict settings, including peacekeeping expenditures
and humanitarian assistance, see OECD (2010b).
11
 For this purpose, resumption of conflict is defined as at least 1,000 battle-related deaths/year, as
measured by the “intensity of civil war” variable from the PRIO/Uppsala database.


                                                      8
There are large differences among recipient countries in the amount of aid received.
Figure 2 illustrates this by comparing annual aid per capita (in constant 2007 dollars)
during the first five postconflict years across six countries. Bosnia and Herzegovina tops
the list by a wide margin, receiving almost $350/person/year. The other five countries –
all in sub-Saharan Africa or Central America – received less than $100/person/year, with
aid to Guatemala being the lowest at barely $30 per capita. These disparities almost
certainly reflect the geopolitical priorities of the donor countries more than differences in
peacebuilding needs and human needs across recipient countries.

[INSERT FIGURE 2 HERE]

As noted above, some authors have argued that the timing of postconflict aid is poorly
aligned with the “absorptive capacity” of recipients, at least if by the latter we mean
opportunities to maximize positive impacts on economic growth: it is argued that aid
peaks in the immediate postconflict years and subsequently declines, whereas growth
impacts are greatest in the middle years of the postconflict decade. The trends depicted in
Figure 3 for the same six countries suggest that this picture is not easily generalizable:
significant “front-loading” of aid occurred only in Bosnia and Herzegovina, whereas in
the other five countries aid was disbursed at a relatively constant level over the
postconflict decade. Moreover, objectives other than economic growth underpin the
rationale for immediate postconflict assistance.

[INSERT FIGURE 3 HERE]

Speed of disbursement is widely regarded as a critical factor in postconflict and fragile
state situations where actual and perceived progress is essential to confidence-building.
Yet the international financing toolbox remains sluggish, sequential and unintegrated.
The case of southern Sudan is illustrative. Sudan’s Comprehensive Peace Accord (CPA)
was signed in January 2005. A multi-donor trust fund in support of building domestic
institutions in south Sudan was established in August 2005, but by January 2006 – a full
year after the signing of the CPA – no significant spending from the MDTF had occurred.
Limited spending began only in March 2006. Disbursements ran into a Catch-22
stumbling block: World Bank administrators of the trust fund could not disburse without
south Sudanese officials themselves setting priorities and approving expenditures, but
“the government officials in question had next to no human resources, and the purpose of
the MTDF was precisely to help build that capacity” (Chandran et al. 2008, p. 20).

Comparison between aid to postconflict and other fragile states

Compared to the international resources provided to postconflict countries – which are
substantial, albeit uneven across countries and in some cases over time – the resources
provided to “fragile” states at risk of conflict (but not postconflict) are far more modest.
Indeed, states that are not “postconflict” but are nevertheless classified as “fragile” by the
World Bank receive less aid per capita than non-fragile states. Figure 4 depicts average
aid per capita in 2007 for three subsets of low-income sub-Saharan African countries:

                                              9
non-fragile, fragile (but not post conflict), and postconflict (by the formal peace
definition). Fragile countries receive only 65% as much aid per capita as non-fragile
countries and only 41% as much as postconflict countries.

[INSERT FIGURE 4 HERE]

Figure 5 shows per capita aid disbursements in the six low-income West African
countries classified as “postconflict” (Guinea-Bissau, Sierra Leone, and Liberia) and
“fragile” but not postconflict (Togo, Guinea, and the Gambia) in 2007. Controlling for
region, the disparities between the two are even more striking.

[INSERT FIGURE 5 HERE]

Figure 6 depicts per capita aid trends over the period 1993-2007, again for low-income
sub-Saharan African countries. Here we can observe not only that postconflict countries
consistently received more aid than fragile countries, but also that the disparity between
the latter and other (non-fragile and non-postconflict) countries has widened over time.
This trend coincides with a shift in donor policies towards greater “selectivity” (also
known as “ex post conditionality”) in inter-country aid allocation, and away from
traditional (“ex ante”) conditionality, in an effort to foster “country ownership” and
“development partnerships” and to increase the impact of aid on economic growth.12

[INSERT FIGURE 6 HERE]

Two observations can be made based on these comparisons. First, selectivity in aid
allocation across countries, so as to favor “good performers” with “good policies,” may
contribute systematically to lower aid to fragile countries than to non-fragile countries.
Donors face a dilemma: preferentially allocating aid to those countries where it is likely
to have the strongest pro-growth impacts (at least, in the short run) means providing less
to those countries classified as “fragile states.” Second, if and when the latter “at-risk”
countries descend into violent conflict, much greater resources ultimately are provided as
postconflict aid. Together, these observations suggest that it would be more cost-effective
to allocate more resources to fragile states before the onset of violent conflict.

This conclusion rests, however, on the less-than-certain premise that additional
international assistance would reduce rather than increase the likelihood of conflict; in
other words, the assumption that aid to fragile states would support conflict prevention. It
would be naïve to assume that this happy result necessarily holds. On the contrary,
inappropriate aid can foster polarization and generate conflict.13 As an OECD report on
“Resource Flows to Fragile and Conflict-Affected States” observes, both natural
resources and ODA are “particularly easy targets” for corrupt (loot-seeking) behavior
(OECD 2009a, p. 96, emphasis added). Corruption, in turn, may feed, and feed upon

12
     For discussion, see World Bank (1998), Koeberle (2003), and Hout (2007).
13
     For discussion, see Boyce (2002), Flores and Nooruddin (2009), and Hartzell and Hoddie (2010).


                                                     10
weak governance and insecurity, combining in what Zoellick (2008) describes as a
“downward spiral.”

Yet the same caveat applies to postconflict countries: here, too, aid may have perverse
effects, exacerbating the likelihood of renewed conflict and becoming a target for
corruption. Indeed there is no reason to assume that these risks are greater in fragile states
without a recent history of violent conflict than in postconflict countries. Hence the
relatively low per capita aid provided to fragile states can be viewed as a symptom of
resource misallocation by the international community. This said, the case for more aid to
“at-risk” countries would be greatly strengthened by the development and use of
analytical tools for ex ante conflict impact assessment and ex post evaluation of aid’s
effectiveness as an instrument for conflict prevention.14

Apart from the level of aid per capita, the volatility of aid flows may be important since
greater volatility can exacerbate uncertainties and make budget and planning decisions
more difficult.15 Table 1 presents data on aid volatility, as measured by the coefficient of
variation of annual ODA disbursements, in sub-Saharan African countries in the period
1989-2007. On average, aid flows to postconflict countries and fragile states are more
volatile than flows to countries that are neither fragile nor postconflict. There are
substantial differences among countries within each group, however, making
generalization difficult. For example, if Nigeria and Republic of Congo were omitted
from the “fragile, not postconflict” group, the average volatility among the remaining
countries would be about the same as that in the “not fragile, not postconflict” group.
Moreover, in some cases the greater volatility in the postconflict group can be explained,
at least in part, by disruptions attributable to conflict itself. These observations suggest
that greater aid volatility may be a country-specific issue rather than a general correlate of
state fragility.

[INSERT TABLE 1 HERE]

Sectoral allocation

The sectoral allocation of international assistance may be just as important as its overall
magnitude and timing. The sectoral distribution of aid to fragile and postconflict
countries differs from that of aid to other countries (see Figure 7). A lower fraction of aid
is spent on infrastructure, although this remains the single largest category of assistance,
accounting for more than 30% of total aid in both fragile and postconflict countries.
Higher fractions are spent on governance aid and humanitarian relief. These differences
are not surprising, and are consistent with reallocation in response to needs specific to
these subsets of aid-receiving countries.

[INSERT FIGURE 7 HERE]

14
     For discussion, see Hoeffler and Billerbeck (2010).
15
     For discussion, see Bulir and Hamann (2003), Levin and Dollar (2005), and Kharas (2008).


                                                       11
This is not to say, however, that sectoral allocations are optimal. While inter-country
differences in needs and aid allocations (as well as lack of finely disaggregated data)
make generalizations difficult, there is evidence of a widespread tendency toward
underinvestment in the agricultural sector. Table 2 presents data on the agricultural share
of ODA in low-income sub-Saharan African countries. Overall, agriculture receives a
small share of aid relative to its share of the labor force: it receives more than 10% of aid
in only six of the 26 countries, while accounting on average for roughly three-quarters of
the labor force. Of course, some fraction of other sorts of aid (such as transportation,
health and education) goes to rural areas, too, mitigating the apparent bias against rural
areas if not offsetting it entirely.

What is striking in these data, however, is that the share of aid directed to agriculture in
fragile and postconflict countries tends to be lower than in non-fragile countries. Insofar
as this is correlated with a lower rural share of non-agricultural aid (for roads, schools,
health clinics, etc.), this suggests that the rural population receives a lower percentage of
aid in these countries despite the importance of inclusive economic development for
conflict prevention and postconflict peacebuilding. Even if we acknowledge the existence
of many other pressing needs, it is hard to avoid concluding that something is amiss in
Burundi and Guinea-Bissau, for example, where agriculture’s share of the labor force is
more than 90% and yet agriculture receives barely 1% of ODA.16

[INSERT TABLE 2 HERE]

To provide some idea of the extent to which aid directly supports statebuilding, Table 3
presents data on the share of ODA allocated to governance and budget support in sub-
Saharan African countries. “Governance” here is broadly measured as the sum of two
categories in the OECD Creditor Reporting System database: “government and civil
society” and “other social infrastructure and services.”17 Budget support refers to money
that is routed through government budgets, which can strengthen state capacities to
allocate resources and manage expenditure. As might be expected, a somewhat larger
fraction of aid is allocated to governance in fragile and postconflict countries (12.2% and
16.3%, respectively) than in non-fragile countries (10.3%). The reverse is true in the case
of budget support, however: the average share is only 1.9% in fragile countries and 6.1%
in postconflict countries, compared to 7.9% in non-fragile countries. Put differently, 98%
of the aid to fragile countries in sub-Saharan Africa bypasses the government budget.

[INSERT TABLE 3 HERE]

16
  The OECD (2010a, pp. 109-110) similarly suggests that in recent years donors have accorded insufficient
priority to the agricultural sector in fragile and postconflict countries.
17
  In the OECD classification scheme, the latter category includes not only aid for purposes closely related
to statebuilding, such as “statistical capacity building” and “employment policy and administration
management,” but also aid for less closely related purposes, such as low-cost housing, narcotics control,
and “social mitigation of HIV/AIDS.” It would be useful if more precise classifications of governance-
related assistance could be introduced in the OECD database.


                                                    12
The share of aid allocated for support of the rule of law – an element of “good
governance” that is often a critical weakness in fragile states and postconflict countries –
generally is even smaller than the share for agriculture. Table 4 presents data on ODA
disbursements in sub-Saharan Africa in the two sub-sectors in the OECD Creditor
Reporting System database that correspond most closely to rule-of-law: “legal and
judicial development” and “human rights.” Although there is some evidence of
responsiveness to the need for greater attention to this sector in fragile and postconflict
countries – the share of ODA in these groups averaged 1.0% and 1.5%, respectively, as
opposed to 0.7% in non-fragile countries – the most striking feature of these data is the
small fraction of aid devoted to this sector.18

[INSERT TABLE 4 HERE]

Money is not the only metric for assessing donor priorities or their goodness-of-fit with
recipient needs, however, nor is it necessarily the most important. A recent review of
donor-supported justice-strengthening programs concluded that “absolute expenditures
are often not a good measure of significance,” because “some activities simply cost more
than others, and some of the most important ones may cost donors very little as they
ultimately hinge on cooperating countries’ willingness to change policies” (Hammergren
2008, p. 13). Since inadequate donor support for justice and the rule of law is often cited
as a problem, it is worth quoting the review at some length:

         The problem arises when donors finance things (buildings and equipment in
         Guatemala or the rest of Central America for that matter) the government could
         easily fund, engage in piloting activities unlikely to be replicated (Cambodia’s
         model court), duplicate each others’ efforts, or fund activities likely to disappear
         once they leave the scene. This is where the imbalances occur – less between what
         is promised and what is financed, but as part of a more rational consideration of
         what makes the most sense…. If an activity does not support something that would
         not otherwise be done; if it does not leverage more change; and if it is redundant
         or not sustainable, perhaps it should not be funded. (Ibid., p. 14).

In other words, there is scope for improving the inter-sectoral allocation of aid in terms
not only of “how much” but also “what” – that is, in terms of how the allocated funds are
spent. In particular, we argue below that too little is invested in building national fiscal
capacities to sustain the ability of the state to mobilize revenue and deliver public goods
and services, and that too little attention is paid to building civil society capacities to
monitor government finances in the interest of ensuring transparency and accountability.


18
  There is no obvious benchmark for ascertaining the “right” amount of aid that ought to be allocated to
support for the rule of law (or, for that matter, to other sectors). The percentages of ODA reported in Table
3 are similar to the share of judicial and legal system expenditures in total government spending within the
donor countries themselves: in 2006-07 these amounted for example to 0.7% in the U.K.; 0.9% in France;
and 1.0% in the U.S. (calculated from data from the European Commission for the Efficiency of Justice and
the U.S. Census Bureau).


                                                     13
Such lacunae in the qualities of aid can be as important – indeed, sometimes more
important – than shortfalls in the quantity of aid.


                             4. Modalities of International Financing

There are multiple channels and modalities by which bilateral and multilateral financing
is provided to postconflict countries and fragile states. These include grants and loans
directly provided by donors to recipient governments, global thematic trust funds,
country-specific trust funds, and debt relief.

Disbursements of ODA by individual donors in 2007, and the percentages of their
resources that went to postconflict countries (here defined by the “negative peace”
criterion) and to fragile states are reported in Table 5.19 These data indicate that top
donors to postconflict countries were the United States, the European Community, IDA,
and the United Kingdom, together accounting for more than half of the total. Among
these donors, the UK devoted the largest fraction (37.6%) of its disbursements to the
negative-peace countries. Overall, postconflict countries received 17% of ODA and
fragile states received 25%.

[INSERT TABLE 5 HERE]

The multiplicity of financing mechanisms has benefits and costs. On the plus side, they
permit substantial resource flows, open channels for non-traditional funders, and may
facilitate assistance in program areas that are critical to postconflict recovery and
stabilization but are not included among traditional ODA activities. On the minus side,
the fragmentation of financing among diverse agencies leads to serious planning,
coordination, and evaluation problems that make it difficult or impossible to ensure that
financial support gets to where it is most needed in a timely, predictable and effective
manner. The systematic tracking of financial flows is itself a major challenge yet to be
met, as illustrated by gaps in coverage in the OECD-DAC database. The many
distribution channels open multiple pathways for corruption and manipulation of aid
flows. For recipient governments with limited administrative capacities, the task of
interacting with multiple donors, each with distinctive procedural and reporting
requirements, can pose a formidable burden.

In this section we briefly discuss five international financing modalities that, one way or
another, seek to address the aid fragmentation problem: IDA, the main window through
which the World Bank channels resources to postconflict and fragile states; multi-donor
trust funds; “whole-of-government” approaches to bilateral aid, that seek to foster
coherence among different agencies within the same donor government; regional
approaches to funding that seek to foster coherence across aid to different countries
within the same conflict-affected region; and the UN Peacebuilding Fund, established in

19
  “Fragile states” as defined in the World Bank’s fragile states list, on which a given country’s
classification can vary over time.


                                                     14
2006 in response to perceived funding gaps particularly in the immediate aftermath of
violent conflict.

IDA

The World Bank’s main window for funding to postconflict and fragile states is IDA. The
2007 IDA-15 review of “Operational Approaches and Financing in Fragile States”
reports that IDA allocated SDR 11 billion to fragile states (about 19% of its total
resources) over the past decade, although in per capita terms IDA flows to non-fragile
states (excluding capped blend countries, that qualify for a mix of IBRD and IDA
lending) were 50% higher than to fragile states. The review recommends measures to
close gaps in IDA’s exceptional financing arrangements, where “exceptional” means
arrangements that depart from its usual Performance-Based Allocation (PBA) that links
aid to progress in improving performance indicators. This points to the inherent tension
between “ex post conditionality” (or “selectivity”), which seeks to allocate aid so as to
reward good performers, and the needs of postconflict and fragile states where
performance according to the usual indicators is often weak.

In addition to IDA, the World Bank administers a State and Peacebuilding Fund (SPF)
that provides support for countries in arrears (to which further lending is barred under
Bank rules) and for small but urgent activities not easily funded by IDA. The SPF was
created in 2008, replacing two earlier trust funds – the Post-Conflict Fund and the Low-
Income Countries Under Stress Fund – on the grounds that conflict and fragility are
closely related. Initial resources came from $100 million in World Bank administrative
funds plus contributions from the governments of Norway and the Netherlands.

Multi-Donor Trust Funds

A diverse set of multi-donor trust funds (MDTFs) have been used to channel resources to
postconflict countries and fragile states in recent years. As of 2007, the total funding
through some eighteen operative MDTFs amounted to $1.2 billion, still a small fraction
of the international financing for postconflict and fragile states. Over the preceding five
years, MDTFs aimed at fragile states disbursed $2.6 billion, with the Afghanistan
Reconstruction Trust Fund (ARTF) accounting for roughly half of this total (IDA 2007,
p. 37). The resource-pooling achieved by MDTFs is intended to encourage joint
assessment, planning and program implementation; provide an umbrella for donor
participation in response to national development plans and priorities; and reduce
transaction costs, especially for recipient governments with limited administrative
capacity (UNDP Review of Trust Funds, 2007).

At present, both the World Bank and UNDP-UNDG operate MDTFs. The World Bank’s
trust funds can provide direct budget support to recipient governments, including
recurrent costs, whereas UN trust funds must be disbursed through a UN cooperating
agency, each of which operates under its own rules and procedures.20

20
     See OECD (2008), pp. 55-58, for a detailed review of operational challenges to the MDTFs.


                                                     15
While potentially competitive and/or duplicative, trust funds can complement each other,
for example by providing direct budget support and sector-specific support. In
Afghanistan, under the so-called “two window” approach, the Bank-administered ARTF
has provided budget support, covering anything but humanitarian assistance and law
enforcement, while the UNDP-administered Law and Order Trust Fund for Afghanistan
(LOTFA) channeled resources to police salaries and other recurrent costs.21 The UN PBF,
the EU African Peace Facility (APF), and the EC Instrument for Stability are also able to
fund non-ODA eligible activities. The Bank and the UNDG have engaged in efforts to
standardize their procedures in order to facilitate their accessibility and use (Scanteam
2007; Development Initiatives 2006). The merits and complications of the two-window
model are now being debated in what could either be defense of turf or a credible
assessment of the value of complementary mechanisms.

As the use of MDTFs becomes more popular, it would be useful to have a comparative
analysis of aggregate and sector financing over time by global, country, and country-
thematic MDTFs to see where adjustments and improvements may be needed.

Whole-of-government approaches

Whole-of-government approaches have been initiated in several OECD donor countries
in an effort to formulate an integrative approach to security, governance, and
development by bringing together defense, foreign, and development ministries. While
these efforts hold potential for policy coherence and harmonization of procedures and
regulations, they are still in the testing phase and suffer the usual problems of timely,
predictable and effective funding. Since they tend not to pool their funds with other
efforts, they also do not resolve the coordination problems that arise when multiple
bilateral and multilateral donors operate in recipient countries. Moreover, the whole-of-
government approaches have tended to focus on postconflict states, where defense
ministries often take a lead role, rather than fragile states where development ministries
would likely take the lead. Differing doctrinal stances and competition for resources have
posed continuing problems in these integrative approaches.22

Regional approaches

While considerable attention has been paid to the regional dimension of conflicts and the
importance of engaging neighboring states in processes of conflict mitigation and
peacekeeping, there have been few concomitant developments in the financial
architecture or structure of aid flows, and even fewer efforts to evaluate them. One such
initiative was the Multi-Country Demobilization and Reintegration Program (MRDP) for
the Great Lakes region of Africa, and this experience should be evaluated for its lessons
21
  In addition, the UNDP-administered Afghan Interim Authority Fund provided initialsupport for the civil
service and Loya Jirga.
22
  For discussion, see (Patrick and Browne 2009; CIC, “Recovering from War” 2008; OECD Resource
Flows, 2008, pp. 52-59).


                                                   16
on the benefits and difficulties of a regional approach. The UK government’s Africa Pool
and Global Conflict Prevention Pool contain regional initiatives, but these operate at
small scale relative to direct country aid programs. Other initiatives aimed at resolving or
staving off regional crises, that merit further examination, include: the OECD-sponsored
Club du Sahel, which provided support and distributional planning for eight countries in
the region subject to drought and famine; the recommended (but not implemented)
proactive approach to regional conflict prevention in the Ferghana Valley in Central
Asia23; the India, Nepal, Bangladesh watershed program aimed as preventing water
resource conflicts in the region; the similarly intentioned Nile River Basin Initiative; and
the Mano River Basin initiative, which focused on early warning and training in the
forest area shared by Guinea, Liberia, and Sierra Leone.24

The potential for regional development banks, including the ADB, AfDB, and IDB, to
play a role in conflict prevention and postconflict reconstruction via cross-border
initiatives should be examined, as should the potential roles of the African sub-regional
organizations that were established initially to promote regional economic integration but
increasingly have become involved in conflict prevention, conflict mitigation and
postconflict reconstruction. ECOWAS Bank for Investment and Development (EBID)
and the Development Bank of Southern Africa (DBSA), affiliated with SADC, may also
have some potential in this regard. While not a funding entity itself, the UN’s regional
office for West Africa (UNOWA) could extend its cross-border functions to advise on
subregional funding opportunities to advance security and development.

The UN Peacebuilding Fund

The one financing modality established specifically to address early recovery in
postconflict situations is of recent origin: the UN Peacebuilding Fund (PBF), created in
2006. It holds promise, but many observers agree that it has failed to utilize the more than
$300 million raised as of February 2010 in the most effective manner. While exceeding
its initial funding goal by some $50 million, with contributions from 45 donor countries,
more than two-thirds of the PBF’s finances come from its five lead donors, making it
highly dependent on the preferences of donors who are able to “forum shop” among a
variety of funding channels. As of March 2009, the PBF had disbursed $122 million to
twelve recipient countries, with 70% of its resources going to the UN Peacebuilding
Commission’s four “adoptee” countries: Burundi, Central African Republic, Guinea-
Bissau, and Sierra Leone. PBF is required to disburse its funds through UN agencies, and
73% of its total allocations have been passed through UNDP. Despite its early recovery
mandate, PBF resources have often been provided in later stages of war-to-peace
transitions.


23
  Center for Preventive Action, “Calming the Ferghana Valley: Development and Dialogue in the Heart of
Central Asia.” New York: Council on Foreign Relations and Century Foundation, 1999.
24
  O’Brien, Graybow and Iacopetta (2000) compiled an inventory of 59 regional intergovernmental
organizations. The fragmentary nature of the available data is reflected in the fact that this study provides
total budgets for only 28 of them. We know of no more recent effort to collect this information.


                                                      17
Established with the goals of financing urgent but underfunded peacebuilding activities
and mobilizing on-going resource flows, the PBF has run into the same constraints that
characterize much of the international aid architecture: lack of strategic planning and
prioritization; complex and time-consuming decision-making and disbursement
procedures; role confusion and coordination failures; unpredictability and uncertain
sustainability; limited staff capacities; and, stemming from all of these, questionable
effectiveness in meeting peacebuilding objectives. On top of these generic problems, the
PBF suffers from having a complicated management structure: it is part of a tripartite
arrangement with the UN’s Peacebuilding Commission and Peacebuilding Support
Office, and it is subject to third-party management through UNDP’s Multi-Donor Trust
Fund office. Ball and van Biejnum (2009) offer a set of recommendations, in part
developed by internal review, to re-align the Fund’s early and medium-term funding
windows, set more explicit peacebuilding priorities, and streamline responsibilities and
procedures to help the PBF fulfill its promise.


                      5. Trends in National Government Finance

This section reviews trends in national government finance in postconflict and fragile
countries. First we discuss domestic revenue; then we discuss the composition of
government expenditure. In section 6, we discuss relationships between international
financing and national government finance.

Domestic revenue

A critical feature of legitimate and effective states is the social contract between the
government and the citizenry. The state provides services including physical security,
economic infrastructure, health care, and education. The public, in turn, pays taxes based
on the perception that the state will fund these services. In postconflict countries and
fragile states, this nexus is often weak, and in some parts of a country it may be non-
existent.

Postconflict countries often have a very low ratio of domestic revenue to national income.
In many cases, they also have a relatively low national-income denominator, with the
result that their ability to fund public goods and services is exceptionally limited. As
shown in Figure 8, this pattern is found across regions, even in countries where a number
of years have elapsed since the onset of peace. In Latin America, the tax coefficient (ratio
of tax revenue to GDP) in Guatemala and El Salvador in 2007 was roughly two-thirds of
the regional average. In sub-Saharan Africa, the tax coefficients in Uganda and the DRC
were 61% and 38% of the regional average, respectively. In South Asia, those in
Afghanistan and Nepal were 43% and 75% of the regional norm. And in the East Asia &
Pacific region, the tax coefficients of Cambodia and Timor-Leste (apart from its offshore
oil economy) were approximately two-fifths of the regional average.

[INSERT FIGURE 8 HERE]



                                            18
The sluggish growth of domestic revenue in postconflict countries reveals a serious
lacuna in policies intended to promote statebuilding and peacebuilding. In many cases,
the fiscal dimension is under-emphasized, if not missing altogether, in the postconflict
policy priorities of national governments and international donors alike. Even in the case
of Guatemala, where the 1996 peace accords explicitly mandated an increase in tax
revenues from 8% of national income – the lowest level in the western hemisphere next to
Haiti – to 12% by the year 2000, progress has been very slow: in 2000, the country’s tax
coefficient barely reached 10%, and six years later it had yet to reach the 12% target (see
Figure 9).

[INSERT FIGURE 9 HERE]

One factor exacerbating the slow revenue growth in postconflict countries has been donor
support for tariff reductions in the name of trade liberalization and economic efficiency.
In Guatemala, for example, even as the IMF’s Managing Director publicly endorsed the
revenue-enhancement goal mandated by the peace accords, the Fund’s staff urged the
government to cut tariffs.25 As shown in Figure 9, tariff revenue in Guatemala fell both
relative to total tax revenue and absolutely. Yet tariffs are often a major component of tax
revenue in low-income countries, among other reasons because the administrative costs
of collecting customs duties are relatively low. Moreover, the efficiency argument for
trade liberalization – that domestic industries should not be protected from international
competitors who can produce at lower cost – is largely irrelevant in contexts where trade-
competing industries do not exist. Guatemala, for example, does not have an automobile
industry, nor is it likely to have one in the foreseeable future; hence there is no efficiency
argument against substantial tariffs on Guatemala’s automobile imports, which could
provide a significant source of tax revenue.
We return to this topic in section 6, where we consider ways that the international
community can actively support – rather than obstruct – revenue generation in low-
income countries in general, and in postconflict countries in particular.
Managing revenues from extractive resources

Extractive industries, such as petroleum and mining, can provide the state with
substantial revenues to fund public expenditure without taxing its citizens directly. There
is a risk, however, that this “easy money” will weaken the state-society fiscal contract by
reducing the government’s accountability to taxpayers, as well as fostering rent-seeking
behavior and exacerbating social tensions. Whether extractive resources prove to be a
blessing or a curse is not a foregone conclusion: the results are contingent on policies and
25
  Visiting Guatemala in May 1997, IMF Managing Director Michel Camdessus stated that the Fund’s only
condition for a stand-by agreement would be that the government complies with its peace-accord
commitments, including a 50% increase in the revenue-to-GDP ratio. Yet at the 1998 meeting of the
Consultative Group for Guatemala, the IMF representative urged the government to ‘resist pressures to
increase import duties or delay the scheduled reduction in customs tariffs,’ arguing that ‘these actions will
have adverse effects on output growth.’ For discussion, see Boyce (2002, pp. 40-47).




                                                     19
politics. The establishment of resource-specific revenue instruments, like the Petroleum
Fund of Timor-Leste (see sidebar), is one recent strategy to address these concerns.
Another, complementary, strategy would be to set up an extractive sector forum, bringing
together citizens, political leaders and companies to forge a “resource compact” that
includes an independent body to monitor the extractive sector (Le Billon 2008).

[INSERT SIDEBAR ON TIMOR-LESTE PETROLEUM FUND HERE]

Expenditure priorities

The other side of the state-society contract is the delivery of services by the state that the
public expects in return for paying taxes. Expenditure patterns in postconflict countries
often differ from regional norms, with relatively high levels of military expenditure and
relatively low levels of expenditure on social services, such as health and education. This
is illustrated in Figure 10, which compares the ratio of military expenditure to public
health expenditure in four postconflict states – Burundi, Angola, Cote d’Ivoire, and Sierra
Leone – to the average ratio in other countries in sub-Saharan Africa. In Sierra Leone,
military expenditure is nearly 40% higher than public health expenditure, and in Burundi
it is more than six times higher; in contrast, the regional norm is for public health
expenditures to be roughly 65% higher than military spending.

[INSERT FIGURE 10 HERE]

These differences are not surprising, but they suggest a need for careful attention to the
composition of government expenditure in light of the requirements of statebuilding and
peacebuilding. Postconflict countries often inherit relatively large military forces;
expenditures associated with postwar disarmament, demobilization, and reintegration
(DDR) may continue to lead to larger-than-normal military budgets and yet be fully
consistent with peace implementation goals. Moreover, in some cases ongoing security
challenges may provide a rationale for above-average military expenditures. But the
political influence of the military may also play a role, independent of security needs that
carry legitimacy in the eyes of the wider public. In the immediate postwar years in El
Salvador, for example, military spending remained three times higher as a percentage of
GDP than it had been before the war; meanwhile inadequate funding for new democratic
institutions, including the National Civilian Police, threatened to undermine the peace
implementation process.26 In such cases, the goals of building an effective and legitimate
state and securing a durable peace require reallocations of expenditure in accord with
these priorities.


                         6. International Assistance and National Finance

This section considers how the provision of external resources affects national finance in
postconflict countries and fragile states. First, we consider the relation between inflows of

26
     For discussion, see De Soto and Castillo (1994) and Boyce (1996, Chapters 5-7).


                                                      20
international assistance and the national tax effort. We then discuss several policies by
which the international community could strengthen the contribution of aid to building
sustainable fiscal systems. Lastly, we examine the impediments to building effective and
legitimate states posed by external debt legacies.

The issues discussed in this section are relevant to aid in general; they are not entirely
unique to postconflict countries and fragile states. But in these countries, where the fiscal
dimension of statebuilding is (or ought to be) a high priority, these issues stand out in
particularly sharp relief. Elsewhere the relationship between international assistance and
national finance may be dismissed as a matter of less-than-critical importance, but in
postconflict countries and fragile states it cannot.

Aid and tax revenue

In principle, international financing can help in building, or repairing, the state-society
contract that ties taxation to the provision of public goods and services. Aid can be used
to increase government expenditures beyond what domestic revenue would otherwise
allow, and if this builds public confidence in the willingness and ability of the state to
provide desired goods and services this can, in turn, help to build support for increased
taxation.

In practice, however, this effect is not a foregone conclusion. By providing a soft option
for recipient states to obtain revenue – compared to the harder route of collecting taxes –
aid may “crowd out” domestic revenue mobilization rather than promoting it. There is no
a priori reason to assume that one effect always will prevail over the other (see Figure
11). More importantly, there is scope for donor policy decisions to influence the direction
and magnitude of the net effect of aid on tax revenue, as discussed below.

[INSERT FIGURE 11 HERE]

Figure 12a shows the relationship between aid and tax revenue (as percentages of GDP)
in the low and lower-middle income countries of sub-Saharan Africa in 2006-08. Two
observations can be made on these data. First, there is some evidence of an inverse
relationship between the two variables – a constant-elasticity estimate of the relationship
indicates that a 10% increase in aid is associated with a 2.4% reduction in taxes27– but
this relationship is primarily attributable to two outliers, Lesotho (with relatively low aid
and high tax revenues) and Sierra Leone (with relatively high aid and low tax revenues).
Second, among the other countries there is no indication that higher aid is associated with
higher tax revenues.

[INSERT FIGURE 12a HERE]



27
     The estimating equation: ln(taxes) = 3.003 - 0.239ln(aid).



                                                       21
Cross-sectional correlations cannot prove causation, of course. In the case of a negative
correlation, higher aid could lead to lower taxes, or lower taxes could lead to higher aid,
or both could be affected by other variables, or some combination of these could be at
work. But we can say that the absence of a positive correlation between aid and taxes
suggests that aid may not be doing much to improve tax revenues.

Econometric studies that have sought to identify causal relationships in the effect of aid
on tax revenues have not settled this issue, but many researchers have found negative or
insignificant impacts. In a review of evidence from sub-Saharan Africa, Moss, Pettersson
and van de Valle (2006) conclude that aid “may have undercut incentives for revenue
collection.” To illustrate, they note that a scattergram showing aid and tax shares in low
and lower-middle income countries in the 28-year period from 1972 to 1999 (reproduced
here as Figure 12b), the top right quadrant (high aid and high taxes) is empty. They also
conclude that aid has adversely affected the quality of governance: “states which can
raise a substantial portion of their revenues from the international community are less
accountable to their citizens and under less pressure to maintain popular legitimacy.”

[INSERT FIGURE 12b HERE]

An IMF study of 107 recipient countries found that the effects of grant aid differed from
those of loans. Grants had a negative impact on domestic revenues, an effect that was
strongest in countries plagued by “high levels of corruption” where “the decline in
revenues completely offsets the increase in grants” (Gupta et al. 2003). Concessional
loans were associated with higher domestic revenue mobilization, an effect the authors
suggest may be due to the need to raise revenue to repay the loans. From the standpoint
of assisting in building the state-society contract, however, increasing revenue for the
purpose of servicing debt to external creditors is a less than resounding success, for it
fails to uphold the state’s side of the state-society contract: the delivery of services to its
citizenry.

Can Aid Crowd in Fiscal Capacity?

There are several ways that aid donor policies could seek to “crowd in” greater fiscal
capacity – the capacity of states to raise revenue, allocate resources, and manage
expenditure – in recipient countries. These include:

        Technical assistance: Donors can – and, to a limited extent, do – provide
        assistance in the area of tax policy and administration. The amounts are generally
        small, however: the OECD (2009a, p. 95) reports that in 2006 only 0.087% of
        official development assistance was dedicated to “tax-related tasks,” and that only
        one-eighth of this ($11 million worldwide) went to fragile and conflict-affected
        countries, primarily Afghanistan, Rwanda, and Kenya.

        Budget support: Channeling aid through the national budget could not only build
        government capacity to allocate and manage expenditure but also strengthen
        public confidence in the ability of the government to deliver goods and services,

                                               22
           thus enhancing their willingness to pay taxes. Yet at present, only a small fraction
           of aid is allocated to budget support. Aid agencies have a fiduciary responsibility
           to see that their money is well spent, but this can be met by “dual control”
           systems that seek to ensure accountability and transparency. Such a system has
           been implemented by the Afghanistan Reconstruction Trust Fund (ARTF),
           discussed in section 4: recurrent budget expenditures are submitted to the ARTF
           for reimbursement, which is disbursed after review by an external monitoring
           agent whose function is akin to that of an internal auditor.28

           Revenue conditionality: Donors could (but rarely do) make access to aid
           conditional upon progress in domestic revenue mobilization, a policy that would
           be akin to the provision of matching grants by philanthropic foundations. DfID’s
           former chief economist has proposed that donors collectively limit aid to 50% of
           the tax revenue collected by the recipient government, excluding oil and mineral
           revenues (Wood 2008). Among the low and lower-middle income countries of
           sub-Saharan Africa, this ceiling would require aid reductions (and/or tax
           increases) in the countries in the red half of the scattergram in Figure 12a. A
           “matching funds” approach could be facilitated if inter-donor coordination were
           enhanced by channeling more medium-to-long term aid commitments through
           multi-donor trust funds (FIAS 2009, p. 27). Alternatively, revenue conditionality
           could be applied by individual donors to a subset of total aid.29 For example,
           budget support in Afghanistan through the ARTF now includes an incremental
           amount tied to revenue performance in addition to a gradually declining base
           level.

           Taxing aid: Currently, not only aid itself but also aid-generated income flows are
           largely tax-exempt in the recipient countries. This deprives governments of
           needed revenues, impedes learning-by-doing in tax administration, and, above all,
           sends a perverse message to the populace: powerful people do not trust the
           government with their money and do not pay taxes. Efforts to tax aid – including
           tariffs on aid-financed imports, taxes on the incomes of national and expatriate aid
           agency employees and private contractors, and taxes on rental incomes of the
           owners of housing and offices leased to aid agencies and their staff – could help
           to prime the pump of domestic tax collection. This would require substantial, and
           in our view quite sensible, changes in current aid protocols.30



28
     For more on the ARTF, see Ghani et al. (2007) and Peace Dividend Trust (2009).
29
   Revenue conditionality also has been proposed in a study by the IMF Fiscal Affairs Department (Gupta
et al. 2003, p. 18): “Traditionally, donors have imposed conditions on the expenditure side on how their
resources could be utilized, without taking into account the impact of assistance on revenues… A similar
type of requirement could be considered for the revenue side.”
30
  For further discussion, see Boyce and O’Donnell (2007); Carnahan (2007); and OECD (2010a, pp. 105-
107).


                                                     23
           Local procurement: It is estimated that only one of every ten dollars spent by the
           UN Transitional Administration in East Timor ‘actually reached the East
           Timorese’ (Chesterman 2004, p. 183).31 Local procurement can help to spur
           economic recovery and development (see sidebar on “spending the development
           dollar twice” in Afghanistan).32 In so doing, it can also help to expand the
           country’s tax base, particularly if procurement policies are tied where appropriate
           to tax compliance by local firms. Channeling a larger share of donor expenditure
           into local economies will require measures beyond the formal “untying” of aid
           envisioned in the 2005 Paris Declaration on Aid Effectiveness (Clay, Geddes and
           Natali 2009): it will require preferential policies to favor local procurement. Even
           when local procurement would entail higher transaction costs (or other costs),
           such policies cannot be dismissed as “inefficient” once we recognize that what is
           being procured is not only a given volume of goods and services, but also positive
           externalities created in their provision.

       [INSERT SIDEBAR ON LOCAL PROCUREMENT IN AFGHANISTAN HERE]

           Support for civil society capacity to demand transparency and accountability: Aid
           donors can seek to strengthen the society side of the state-society fiscal contract,
           too. In particular, efforts to promote transparency and accountability can work
           from the demand side as well as the supply side. The public expenditure
           monitoring by civil society organizations in Uganda’s Poverty Action Fund,
           mentioned earlier, is an example. Ugandan civil society groups have received
           support for this work from ODA donors, including DANIDA and DfID, as well as
           from private agencies such as OXFAM (de Renzio et al. 2006). Similar initiatives
           have been launched in other countries, including Croatia and South Africa
           (Robinson 2006). To date, however, support for such efforts has been the
           exception rather than the rule.

Postconflict debt burdens

Many postconflict countries face high external debt burdens. These debts are often a
legacy, at least in part, of borrowing that fed into social tensions that precipitated the
conflict and of borrowing that was used to finance military expenditure during the
conflict. Figure 13 illustrates this pattern in sub-Saharan African countries: in Liberia,
Guinea-Bissau, Burundi and the DRC, the ratio of external debt to GDP in 2007 exceeded
1.0, compared to an average of 0.4 in other countries in the region.

[INSERT FIGURE 13 HERE]


31
  Chesterman (2004, pp. 183-4) gives this telling example: “At one point, $27 million was spent annually
on bottled water for the international staff – approximately half the budget of the embryonic Timorese
government, and money that might have paid for water purification plants to serve both international staff
and locals well beyond the life of the mission.”
32
     Peace Dividend Trust (2009). See also Carnahan, Durch and Gilmore (2006).


                                                    24
As a result, postconflict countries often are obliged to devote a substantial share of scarce
government resources to the payment of debt service. As shown in Figure 14, in 2006
debt-service payments were seven times greater than public health expenditure in
Guinea-Bissau, more than five times greater in Burundi, and more than four times greater
in Angola. In the average sub-Saharan African country, the ratio was considerably lower
although debt service still exceeded public health expenditures. Along with its human
cost, the diversion of African government resources into debt service may impose a
political cost by weakening the state-society contract that underpins peace.

[INSERT FIGURE 14 HERE]

One way that donors can ease debt burdens is by offering debt relief. Some relief has
been provided by bilateral creditors. The Heavily Indebted Poor Countries (HIPC)
initiative has provided some temporary and provisional relief for countries that have
passed its “decision point,” but relatively few countries have reached the “completion
point” at which time debt relief becomes irrevocable. At present, debt burdens in many
countries remain high as shown Figures 13 and 14.

Moreover, the conditions countries are required to meet to qualify for debt relief may be
problematic, if not simply infeasible, in the wake of violent conflict. Announcing an
emergency postconflict loan to the Central African Republic in 2004, for example, a
senior IMF official remarked: “It will be important to resist political pressures to
increase spending ahead of the elections, and to sustain and deepen the effort to contain
the wage bill, given the large share of public expenditure absorbed by salary payments.” 33
Noting the singular importance during postconflict transitions of efforts “to improve the
well-being of public servants and other groups that need to feel part of the peace
process,” del Castillo (2008, p. 71) comments that such business-as-usual demands from
donors “have not contributed to national reconciliation, and have even endangered
peace.”

Another potential way to ease debt burdens is by the selective erasure of “odious” debts,
defined as debts contracted for purposes that did not benefit the people of the country,
without their consent, in situations where creditors knew or should have known that this
was the case. The establishment of an international arbitration body where claims of
odious debt could be adjudicated at low cost – a step proposed by the government of
Norway in 2005 – would make this policy more feasible.34




33
  “IMF Approves US$8.2 million Emergency Post-Conflict Assistance for the Central African Republic,”
IMF Press Release No. 04/158, July 23, 2004.
34
     For discussion, see Ndikumana and Boyce (2010).


                                                   25
                                      7. Conclusions

This review of financing for fragile and conflict-affected states has revealed a variety of
weaknesses in current arrangements. The performance and eligibility criteria that have
been introduced for much of ODA are not well-suited to addressing the challenges posed
by weak state-society relationships and the risk of violent conflict. The large number of
bilateral and multilateral donors provides diversity of funding sources but impedes policy
coordination and strategy coherence, while placing substantial burdens on recipient
countries with limited administrative and managerial capacity. Multi-donor trust funds
and other forms of pooled funding have in some instances resulted in greater
collaboration among donors and recipients, but lack of agreement on strategic goals and
competition among donors with diverse national and institutional interests continue to
impede coordination and progress on the ground. International assistance, conditioned by
donors’ postcolonial and strategic interests, is characterized by an uneven distribution of
resources across countries and regions. Short-term mandates, a paucity of multi-year
funding, volatility in aid flows and delays in disbursements impede both planning and
timely implementation.

The lack of credible and consistent data on aid flows impedes systematic analysis,
evaluation, and learning, especially in fragile and conflict states where studies are forced
to rely on incomplete and anecdotal evidence. Despite on-going attempts to improve data
collection and recording, it remains difficult to track the gaps between pledges,
commitments and disbursement of aid, and impossible to document its final resting place,
whether in the administrative bureaucracies of funders or recipients, external service
providers and consultants, or actual program deliverables on the ground. OECD
recording codes are not universally agreed, nor are non-eligible ODA activities
systematically recorded. At present, neither donors nor recipients have anything close to
an adequate idea of how aid monies are being spent, let alone the impacts of this
spending on peacebuilding and statebuilding.

All too often, funding is geared to priorities other than strengthening the state-society
contract and building durable peace. Donors often target projects that provide banner
visibility and reflect the perceived interests of constituents in donor countries rather than
pressing postconflict needs that build confidence in recipient countries. Off-the-shelf
“templates” for postconflict reconstruction substitute for the hard work of designing
projects around local political, economic, environmental and socio-cultural needs.
Substantial amounts of international aid fail to touch ground in the recipient countries,
instead paying for foreign consultants and the administrative costs of the international
presence. External funds can crowd out incentives for domestic revenue generation,
deferring steps toward national self-sufficiency and fiscal sustainability. Most aid
bypasses the government budget, failing to build capacity in resource allocation, public
expenditure management, and transparent and accountable governance. The international
presence itself creates fleeting and detached economic enclaves that fail to provide the
building blocks for local enterprise, leaving the national economy on the sidelines in
processes of procurement, service provision, and employment generation.



                                             26
Recommendations

With this daunting constellation of problems in mind, we offer the following
recommendations:

       Create an international, comprehensive database on aid to fragile and conflict
              states in order to facilitate monitoring, evaluation and accountability for
              both donors and recipients.
       Create national-level databases on aid commitments and actual and planned
              disbursements, including mechanisms (e.g., direct budget support, service
              delivery through NGOs, etc.) and sectoral allocations, a step that would
              require that donors track this information internally and share it.
       Develop criteria for international funding for fragile and conflict-affected states
              that do not meet generic performance criteria.
       Strengthen aid for conflict prevention, not only by devoting more resources to
              preventive action but also by mainstreaming conflict impact assessment in
              aid programming and project appraisal in fragile states.
       Strengthen the capacity of regional and sub-regional organizations and
              development banks to address the regional causes and effects of conflict
              and seek programmatic solutions.
       Capitalize multilateral funding mechanisms that can accommodate “aid orphans”
              – including both countries and sectors within countries – independent of
              donor interests and priorities.
       Establish a “stand-by” window for rapid needs assessment, deployment of
              personnel and disbursement of funds to meet immediate stabilization and
              conflict-recovery needs beyond humanitarian relief.
       Provide funding, including budget support, nationally to support locally
              determined plans for recovery and stabilization and help build public
              confidence in the state’s capacity to fulfill the social contract.
       Structure aid in ways that develop national public finance and management
              capacity with attention to fiduciary responsibility and accountability,
              including through civil society watchdog mechanisms.
       Explore ways to bring aid and the associated income flows into the tax base of
              recipient countries – so as to increase domestic revenue, build revenue-
              collection capacities, and affirm the principles of good fiscal governance –
              an effort that could begin with clarification of what is now taxable and non-
              taxable so as to eliminate blurred lines that currently leave much scope for
              tax evasion.
       Design and implement procurement policies that favor – or, at a minimum, do not
              exclude – local contractors, service providers, and labor, with an eye to
              stimulating economic sectors that ensure local livelihoods and generate
              employment.




                                           27
Acknowledgements

The authors are grateful to Sarah Cliffe, Gary Milante and their colleagues on the WDR
team for valuable comments and the provision of data. We also benefitted from very
useful comments provided by reviewers William Byrd, Susan Woodward, and the
OECD/DAC International Network on Conflict and Fragility. For research assistance we
are grateful to Matthew Benson, Grace Chang, Megan Gleason, and Jesse Sanes. The
usual caveats apply.



                                      References

Alvarez-Plata, Patricia and Tilman Brück. 2007. “Postwar Debts: Time for a New
Approach.” In James K. Boyce and Madalene O’Donnell, eds. 2007. Peace and the
Public Purse: Economic Policies for Postwar Statebuilding. Boulder, CO: Lynne
Rienner, pp. 245-269.

Andersen, Louis. 2007. “Democratic Governance in Post-Conflict Liberia: An Interview
with Dr. Amos Sawyer.” Working Paper 2007/20, Danish Institute for International
Studies (DIIS), Copenhagen.

Ball, Nicole, and Mariska van Beijnum. 2009. “Review of the Peacebuilding Fund.”
Paper, United Nations Peacebuilding Fund, New York.

Baunsgaard, Thomas and Michael Keen. 2005. “Tax Revenue and (or?) Trade
Liberalization.” Washington, DC: IMF Working Paper No. WP/05/112.

Birdsall, Nancy. 2007. “Do No Harm: Aid, Weak Institutions, and the Missing Middle in
Africa.” Washington, DC: Center for Global Development, Working Paper No. 113,
March.

Boyce, James K. 2002. Investing in Peace: Aid and Conditionality after Civil Wars.
Oxford: Oxford University Press.

Boyce, James K. 2008. “Post-Conflict Recovery: Resource Mobilization and
Peacebuilding.” Working Paper Series No. 159, Political Economy Research Institute,
University of Massachusetts Amherst, Amherst.

Boyce, James K. and Madalene O’Donnell, eds. 2007. Peace and the Public Purse:
Economic Policies for Postwar Statebuilding. Boulder, CO: Lynne Rienner.

Boyce, James K., ed. 1996. Economic Policy for Building Peace: The Lessons of El
Salvador. Boulder, CO: Lynne Rienner.



                                          28
Boyce, James K., and Ndikumana, Léonce. Forthcoming. “Capital Flight from Sub-
Saharan Africa: Linkages with External Borrowing and Policy Options.” International
Review of Applied Economics.

Brautigam, Deborah, Odd-Helge Fjeldstad, and Mick Moore, eds. 2008. Taxation and
State-Building in Developing Countries: Capacity and Consent. Cambridge: Cambridge
University Press.

Bulir, Ales and A. Javier Hamann. 2003. “Aid Volatility: An Empirical Assessment.”
IMF Staff Papers 50(1): 64-89.

Call, Charles T. and Vanessa Wyeth, eds. 2008. Building States to Build Peace. Boulder:
Lynne Rienner.

Carnahan, Michael. 2007. “Options for Revenue Generation in Post-Conflict
Environments.” New York: Center on International Cooperation, and Amherst: Political
Economy Research Institute. May.

Carnahan, Michael, Scott Gilmore, and Monika Rahman. 2005. “Economic Impact of
Peacekeeping.” Phase I Interim Report. Ottawa and New York: Peace Dividend Trust.

Carnahan, Michael, William Durch and Scott Gilmore. 2006. “Economic Impact of
Peacekeeping.” Final Report. Ottawa and New York: Peace Dividend Trust.

Chandran, Rahul, Bruce Jones and Natasha Smith. 2008. “Recovering from War: Gaps in
Early Action.” Report for the UK Department of International Development, Center on
International Cooperation, New York.

Chesterman, Simon. 2004. You, the People: The United Nations, Transitional
Administration, and State-Building. New York: Oxford University Press.

Clay, Edward, Matthew Geddes and Luisa Natali. 2009. “Untying Aid: Is it Working? An
Evaluation of the Implementation of the Paris Declaration and of the 2001 DAC
Recommendation of Untying Aid to the LDCs.” Copenhagen: Danish Institute for
International Studies

Clements, Benedict, Sanjeev Gupta, Alexander Pivovarsky, and Erwin R. Tiongson.
2003. “Foreign aid and Revenue response: Does the Composition of Aid Matter?”
Working Paper WP/03/176, International Monetary Fund, Fiscal Affairs Department,
Washington, DC.

Collier, Paul and Anke Hoeffler. 2004. “Aid, Policy and Growth in Post-Conflict
Societies.” European Economic Review 48(5): 1125-1145.

Collier, Paul et al. 2003. Breaking the Conflict Trap: Civil War and Development Policy.
A World Bank Policy Research Report. New York: Oxford University Press.

                                           29
Del Castillo, Graciana. 2008. Rebuilding War-Torn States: The Challenge of Post-
Conflict Economic Reconstruction. Oxford: Oxford University Press.

De Renzio, Paolo, Vitus Azeem and Vivek Ramkuman. 2006. “Budget Monitoring as an
Advocacy Tool: Uganda Debt Network.” Brighton: Institute of Development Studies.
Case study prepared for the research project on Lessons from Civil Society Budget
Analysis and Advocacy Initiatives. April.

De Soto, Alvaro and Graciana del Castillo. 1994. “Obstacles to Peacebuilding.” Foreign
Policy 94: 69-83.

Development Initiatives. 2006. “Review of Trust Fund Mechanisms for Transition
Financing.” Phase 2 Report, Somerset, UK.

Di John, Jonathan. 2006. “The Political Economy of Taxation and Tax Reform in
Developing Countries.” Helsinki: World Institute for Development Economics Research,
Research Paper No. 2006/74. July.

Di John, Jonathan. 2008. “Conceptualising the Causes and Consequences of Failed
States: A Critical Review of the Literature.” London: LSE Crisis States Research Centre,
Working Paper No, 25. January.

Department of Peace Keeping Operations and The World Bank, New York and
Washington, DC.

Durch, William J. 2009. “The Economic Impact of Peacekeeping: An Update.”
Washington, DC: Henry L. Stimson Center.

Fearon, James D. and David D. Laitin. 2003. “Ethnicity, Insurgency, and Civil War.”
American Political Science Review 97(1): 75-90.

Fjeldstad, Odd-Helge and Mick Moore. 2008. “Tax Reform and State-Building in a
Globalised World.” In Deborah Brautigam, Odd-Helge Fjeldstad and Mick Moore, eds.,
Taxation and State-Building in Developing Countries: Capacity and Consent.
Cambridge: Cambridge University Press, pp. 235-260.

Flores, Thomas E. and Irfan Nooruddin. 2009. “Financing the Peace: Evaluating World
Bank Post-Conflict Assistance Programs.” Review of International Organizations 4(1): 1-
27.

Foreign Investment Advisory Service (FIAS). 2009. Taxation as State-Building:
Reforming Tax Systems for Political Stability and Sustainable Economic Growth. A
Practitioner’s Guide. Washington, DC: International Finance Corporation.

Forman, Shepard and Stewart Patrick, eds. 2000. Good Intentions: Pledges of Aid for
Post-Conflict Recovery. Lynn Rienner Publishers.
                                           30
Forman, Shepard, Stewart Patrick and Dirk Salomons. 2000. “Recovering From Conflict:
Strategy for an International Response.” New York: Center on International Cooperation,
Paying for Essentials: A Policy Paper Series.

Galtung, Fredrik and Martin Tisné. 2009. “A New Approach to Postwar Reconstruction.”
Journal of Democracy 20(4): 93-107.

Ghani, Ashraf, Clare Lockhart, Nargin Nehan and Baqer Massoud. 2007. “The Budget as
the Linchpin of the State: Lessons from Afghanistan.” In James K. Boyce and Madalene
O’Donnell, eds. 2007. Peace and the Public Purse: Economic Policies for Postwar
Statebuilding. Boulder, CO: Lynne Rienner, pp. 153-183.

Ghani, Ashraf and Clare Lockhart. 2008. Fixing Failed States: A Framework for
Rebuilding a Fractured World. Oxford: Oxford University Press.

Hammergren, Lisa. 2008. “Balanced Justice and Donor Programs: Lessons from Three
Regions of the World.” New York: Open Society Institute. November.

Hartzell, Caroline A. and Matthew Hoddie. 2010. “Economic Liberalization via IMF
Structural Adjustment: Sowing the Seeds of Civil War?” International Organization 64:
339-356.

Hoeffler, Anke and Sarah v Billerbeck. 2010. “Post-conflict Recovery and Peace
Building: Background Paper for the World Development Report 2011.”

Hout, Will. 2007. The Politics of Aid Selectivity: Good governance criteria in World
Bank, US and Dutch Development Assistance. Abingdon: Routledge.

International Development Association. 2007. Operational Approaches and Financing in
Fragile States, Operational Policy and Country Services (OPCS) and Resource
Mobilization Department (FRM). Washington, DC.

International Monetary Fund (IMF). 2009. “Democratic Republic of Timor-Leste:
Selected Issues.” Washington, DC: IMF Country Report No. 09/220. July.

Kang, Seonjou and James Meernik. 2004. “Determinants of Post-Conflict Economic
Assistance.” Journal of Peace Research 41(2): 149-166.

Karl, Terry Lynn. 1997. The Paradox of Plenty. Stanford: Stanford University Press.

Kharas, Homi. 2008. “Measuring the Cost of Aid Volatility.” Washington, DC:
Brookings Institution, Wolfensohn Center for Development, Working Paper No. 3.

Koeberle, Stefan G. 2003. “Should Policy-Based Lending Still Involve Conditionality?”
World Bank Research Observer 18(2): 249-273.

                                           31
Krafchik, Warren. 2001. “Can Civil Society Add Value to Budget Decision-Making?”
Washington, DC: International Budget Project.

Leader, Nicholas and Peter Colenso. 2005. “Aid Instruments in Fragile States.” London:
Department for International Development (DfID) Policy Division, Poverty Reduction in
Difficult Environments Working Paper No. 5. March.

Le Billon, Philippe. 2008. “Resources for Peace? Managing Revenues from Extractive
Industries in Post-Conflict Environments.” New York: Center on International
Cooperation, and Amherst: Political Economy Research Institute. February.

Levin, Victoria and David Dollar. 2005. “The Forgotten States: Aid Volumes and
Volatility in Difficult Partnership Countries (1992-2002).” Paper prepared for the DAC
Learning and Advisory Process on Difficult Partnerships, January.

Moreno Torres, Magui and Michael Anderson. 2004. “Fragile States: Defining Difficult
Environments for Poverty Reduction.” London: Department for International
Development (DfID) Policy Division, Poverty Reduction in Difficult Environments
Working Paper No. 1. August.

Ndikumana, Léonce and James K. Boyce. 1998. “Congo's Odious Debt: External
Borrowing and Capital Flight in Zaire.” Development and Change 29(2): 195-217.

Ndikumana, Léonce and James K. Boyce. 2010. “Capital Flight from Sub-Saharan
Africa: Linkages with External Borrowing and Policy Options.” International Review of
Applied Economics (forthcoming).

Ndikumana, Léonce and Justine Nannyonjo. 2007. “From Failed State to Good
Performer? The Case of Uganda.” In James K. Boyce and Madalene O’Donnell, eds.
2007. Peace and the Public Purse: Economic Policies for Postwar Statebuilding.
Boulder, CO: Lynne Rienner, pp. 15-54.

O’Brien, David, Charles Graybow and Maurizio Iacopetta. 2000. “A Geographically
Based Introduction to Regional Intergovernmental Organizations.” New York: Center on
International Cooperation.

OECD. 1997. DAC Guidelines on Conflict, Peace and Development Co-operation. Paris.

OECD. 2007. “Principles for Good International Engagement in Fragile States &
Situations.” Paris. April.

OECD. 2008. “Concepts and Dilemmas of State Building in Fragile Situations: From
Fragility to Resilience.” Journal on Development. 9 (3).




                                           32
OECD. 2009a. Resource Flows to Fragile and Conflict-Affected States: Annual Report
2008. Paris.

OECD. 2009b. Aid Effectiveness Survey 2008. Data online at
http://www.oecd.org/dataoecd/21/11/41564900.xls.

OECD. 2010a. Do No Harm: International Engagement for Statebuilding. Paris. January.

OECD. 2010b. Transition Financing: Building a Better Response. Paris. May.

Paris, Ronald and Timothy D. Sisk. 2007. Managing Contradictions: The Inherent
Dilemmas of Postwar Statebuilding. New York: International Peace Academy.

Patrick, Stewart, and Kaysie Brown. 2007. Greater than the Sum of its Parts?: Assessing
“Whole of Government” Approaches to Fragile States. New York: International Peace
Academy.

Peace Dividend Trust (PDT). 2009. Spending the Development Dollar Twice: The Local
Economic Impact of Procurement in Afghanistan. Ottawa and New York: PDT.

Pettersson, Gunilla, Todd Moss, and Nicolas van de Walle. 2006. “An Aid-Institutions
Paradox? A Review Essay on Aid Dependency and State Building in Sub-Saharan
Africa.” Working Paper 74, Center for Global Development, Washington, DC.

Picciotto, Robert. 2008. “State Fragility and Human Security in Africa: A Background
Paper.” London: King’s College. Background paper for JICA/UNDP project on conflict
prevention and development cooperation.

Radin, Beryl A. 2007. “Performance Measurement and Global Governance.” Global
Governance 13: 25-33.

Rice, Susan and Stewart Patrick. 2008. Index of State Weakness in the Developing World.
Washington, DC: Brookings Institution.

Robinson, Mark. 2006. “Budget Analysis and Policy Advocacy: The Role of Non-
Governmental Public Action.” Brighton: Institute of Development Studies, Working
Paper No. 279. September.

Rotberg, Robert (ed.). 2004. When States Fail: Causes and Consequences. Princeton:
Princeton University Press.

Scanteam. 2007. Review of Post-Crisis Multi-Donor Trust Funds. Final Report for The
World Bank, Norad, CIDA, Netherlands Ministry of Foreign Affairs, UKDfID. Oslo.




                                          33
Schiavo-Campo, Salvatore. 2003. “Financing and Aid Management Arrangements in
Post-Conflict Situations.” Washington, DC: World Bank, Conflict Prevention and
Reconstruction Unit, Working Paper No. 6. June.

Stoddard, Abby. Forthcoming. “Operational Consequences of Reform: Updated Financial
Analysis.” Humanitarian Outcomes. New York and London.

Suhrke, Astri and Julia Buckmaster. 2005. “Post-war Aid: Patterns and Purposes.”
Development in Practice 15(6): 737-746.

Suhrke, Astri, Espen Villanger and Susan L. Woodward. 2005. “Economic Aid to Post-
Conflict Countries: A Methodological Critique of Collier and Hoeffler.” Conflict,
Security and Development 5(3): 329-361.

Tilly, Charles. 2009. “Foreword.” In William Martin, Ajay K. Mehrota and Monica
Prasad, eds., The New Fiscal Sociology: Taxation in Comparative and Historical
Perspective. Cambridge: Cambridge University Press.

United Nations Development Program. 2007. “Multi-Donor Trust Funds: Guidance Note
to UN Country Teams.” Report, UNDP, New York.

Wood, Adrian. 2008. “How donors should cap aid in Africa,” Financial Times,
September 4.

World Bank. 2007. “Speed of Delivery: Analysis of Projects under Bank-Administered
Multi-Donor Trust Funds (MDTFs).” Report. Washington, DC.

World Bank. 2008. Assessing Aid: What Works, What Doesn’t, and Why. Washington:
World Bank.

Zoellick, Robert B. 2008. “Fragile States: Securing Development.” Survival 50(6): 67-84.




                                          34
                                         [Sidebar]

                              RESOURCE BLESSING?
                           The Timor-Leste Petroleum Fund

When Timor-Leste became an independent nation in 2002 it ranked among the poorest
countries in the world, with an annual per capita income of less than US$1/day. With the
onset of offshore petroleum production two years later, oil and natural gas suddenly
became the country’s largest source of national income and government revenue.

In an effort to ensure that this new wealth will turn out to be a blessing rather than a
“resource curse,” the Timorese Parliament established a Petroleum Fund (PF) in 2004,
modeled on Norway’s, with technical assistance from the IMF.

All petroleum earnings must be deposited into the PF, to be invested in U.S. Treasury
bills or other sovereign debt instruments. By 2009, the Fund had grown to more than
US$5 billion. The government can withdraw money from the PF for public expenditure,
but annual withdrawals are limited to the PF’s “estimated sustainable income.” Any
withdrawal beyond this limit would require the government to provide Parliament with a
detailed explanation of why it is in the long-term interests of the country, along with a
report certified by the PF’s independent auditor on the impact on future sustainable
income.

The PF aims:

       to preserve the value of the country’s petroleum wealth for future generations;
       to limit the risk of “Dutch disease” from exchange-rate appreciation;
       to insulate the national budget from the impact of oil price instability; and
       to ensure transparency in the use of petroleum revenues.

Transparency is important for statebuilding as well as for economic reasons. An IMF
report (2009, p. 5) remarks that the PF aims at “avoiding that rent seeking leads to
economic and social divisions and weakened institutions.” The IMF (p. 13) cautions,
however, that “transparency helps, but easy access to petroleum revenue still alters the
balance between governments and taxpayers and reduces the pressure for accountability.”




                                            35
                                        [Sidebar]

              SPENDING THE DEVELOPMENT DOLLAR TWICE:
                     Local Procurement in Afghanistan

The Afghan Compact, signed by more than 60 nations in early 2006, committed aid
donors to increasing the share of assistance channeled through the government budget
and increasing local procurement of goods and services. A study of ODA disbursements
by the Peace Dividend Trust (2009) found that in the following year approximately 38%
of donor spending entered the Afghan economy.

“Using local goods and services to carry out project work,” the study observes (p. 30),
“allows a development dollar to be spent twice – providing much-needed goods and
services to Afghan citizens and communities while simultaneously creating jobs,
generating revenue, and promoting a more sustainable marketplace – all of which can
ultimately reduce the likelihood of a relapse into conflict.”

Among the eight large ODA donors surveyed by PDT researchers, the share of local
content in their spending ranged from 23% (in the case of the United States) to 71% (in
the case of the United Kingdom).

The study found that 70-80% of aid channeled to the government in direct budget support
or through multi-donor trust funds went into the local economy, compared to 10-20% of
aid disbursed via contracts with international companies and international NGOs for
construction, technical assistance, service delivery, and the supply of specific goods.
Smaller amounts of ODA were disbursed through contracts with local providers, mostly
for construction and rehabilitation of buildings and other infrastructure; in these cases,
local content was found to be 35-50%, since many inputs were imported.

Three main recommendations emerged from this analysis:

       First, donors should increase the share of aid channeled into budget support and
       trust funds, so as to increase spillover benefits to the local economy.

       Second, donors should make efforts to increase local procurement, among other
       ways by translating tenders into local languages, supporting seminars for local
       small and medium enterprises on procurement procedures, and educating
       international staff on the economic benefits of local spending.

       Finally, donors should monitor and collect systematic data on the share of local
       procurement in their aid disbursements.




                                            36
                                     Table 1: Aid Volatility
             (Coefficient of variation of real ODA disbursements, 1989-2007)


                  Country         Volatility                    Country              Volatility
         Post Conflict              0.65              Not Fragile, Not Post
                                                                                        0.39
         Angola                     0.55              Conflict
         Burundi                    0.53              Benin                             0.24
         Central African Republic   0.39              Botswana                          0.50
         Congo, Dem. Rep.           1.38              Burkina Faso                      0.36
         Cote d'Ivoire              0.69              Cape Verde                        0.21
         Guinea-Bissau              0.35              Gabon                             0.73
         Liberia                    0.98              Ghana                             0.34
         Mozambique                 0.30              Kenya                             0.39
         Rwanda                     0.37              Lesotho                           0.37
         Sierra Leone               0.58              Madagascar                        0.52
         Somalia                    0.77              Malawi                            0.20
         Sudan                      0.91              Mali                              0.34
         Fragile, Not Post                            Mauritania                        0.22
                                    0.62
         Conflict                                     Mauritius                         0.67
         Cameroon                   0.62              Namibia                           0.24
         Chad                       0.26              Senegal                           0.28
         Comoros                    0.33              Seychelles                        0.44
         Congo, Rep.                1.40              South Africa                      0.50
         Equatorial Guinea          0.47              Swaziland                         0.36
         Eritrea                    0.56              Tanzania                          0.38
         Ethiopia                   0.44              Togo                              0.53
         Gambia, The                0.36              Zambia                            0.47
         Guinea                     0.29
         Niger                      0.31
         Nigeria                    2.27
         Sao Tome and Principe      0.35
         Togo                       0.53
         Zimbabwe                   0.47

Definitions:
Postconflict: The country is classified as postconflict (by the formal peace definition) at some
time during the period.
Fragile: The country is classified as fragile for three or more years during the period.

Sources:
Postconflict and fragile status from World Bank data files. ODA volatility calculated from
OECD-DAC Creditor Reporting System database.


                                                37
                                  Table 2:
                Agricultural share of labor force and ODA in
              low-income sub-Saharan African countries, 2005

                                      Agriculture share of labor   Agriculture share
               Country
                                              force (%)              of ODA (%)

   Non-fragile countries
          Mauritania                               57.1                  31.8
          Malawi                                   89.4                  14.7
          Kenya                                    77.3                  13.3
          Burkina Faso                             92.1                  12.1
          Mozambique                               86.5                   6.5
          Tanzania                                 82.6                   6.4
          Senegal                                  81                     4.5
          Zambia                                   74.5                   4.2
          Madagascar                               76.1                   2.9
          Benin                                    60                     2.6
          Ethiopia                                 78.8                   2.4
          Uganda                                   84.6                   2.2
          Ghana                                    65                     1.9
    Non-fragile average                            77.3                   8.1
   Fragile (but not postconflict) countries
          Gambia                                   88.9                  15.1
          Zimbabwe                                 71.5                   1.2
          Guinea                                   86.1                   2.1
          Eritrea                                  88.4                   2.1
          Togo                                     61.1                   1.4
          Chad                                     75.8                   0.9
    Fragile average                                78.6                   3.8
   Postconflict countries
          Central African Rep.                     68                    10.8
          Sierra Leone                             64.5                   5.4
          Congo, Dem. Rep.                         65.5                   1.2
          Burundi                                  92.3                   1.2
          Somalia                                  80.1                   1.1
          Guinea-Bissau                            90.9                   1.1
          Liberia                                  60.6                   0.2
    Postconflict average                           74.6                   3.0

Source: Calculated from WDI and OECD Creditor Reporting System databases.



                                              38
                                 Table 3:
ODA disbursements for governance and budget support in sub-Saharan African
                                 countries
                        (annual average 2006-2008)

                                              Governance     Governance   Budget support   Budget support
                  Country
                                              ($ million)    (% of ODA)     ($ million)     (% of ODA)
   Non-fragile countries
               Benin                              65.2           10.66        57.6              9.83
               Botswana                             8             4.89          0                 0
               Burkina Faso                       89.3            8.16        152.1            14.35
               Cape Verde                         25.1           13.79        12.8              6.64
               Ethiopia                          335.8           10.98         0.2                0
               Gabon                               8.9           10.11         0.3              0.43
               Ghana                             141.6            9.11        201.2            13.14
               Kenya                             127.6           10.07        19.7              1.46
               Lesotho                            23.9           20.28          1               1.12
               Madagascar                         100             9.43         63               6.26
               Malawi                             97.4            7.38        92.9               7.4
               Mali                              119.1            9.96        91.9              6.94
               Mauritania                          40            10.16        10.5              2.95
               Mauritius                           7.1             6.7        71.1             51.19
               Mozambique                        223.2           11.09        339.6            16.82
               Namibia                             22            11.59          0                 0
               Niger                              59.4             8.8        44.6              6.55
               Senegal                            97.6            8.56        28.8              2.78
               South Africa                      183.6           19.87         0.1              0.01
               Swaziland                           4.4            8.62          0                 0
               Tanzania                          285.2            9.53        472.4            15.89
               Uganda                            199.3            9.34        136.2             5.33
               Zambia                             100             7.94        150.3            12.02
   Non-fragile average                           102.8           10.31        84.6              7.87
   Fragile (but not postconflict) countries
               Cameroon                           53.7            3.29        141.5             9.11
               Chad                               25.8            6.56         8.7              2.82
               Congo, Rep.                        15.4            5.16        19.1              4.75
               Equatorial Guinea                   5.3           15.87         0.1              0.25
               Eritrea                            15.9           11.65          0                 0
               Gambia, The                        13.5           15.55          0                 0
               Guinea                              37             12.7          0                 0
               Nigeria                           201.4             7.8         8.8              0.45
               Sao Tome and Principe               8.8           15.11         0.1              0.12
               Togo                               50.4           26.15         8.2              3.48
               Zimbabwe                            61            14.48         0.1              0.03
   Fragile average                                44.4           12.21        17.0              1.91
   Postconflict Countries
               Angola                             70.3           21.21         3.5              1.03
               Burundi                            57.4           12.42        42.9              9.14
               Central African Republic           38.7           19.29        13.2              6.72
               Congo, Dem. Rep.                  230.2           13.03         2.1              0.11
               Cote d'Ivoire                       78            13.64        17.6              5.24
               Guinea-Bissau                      17.1           13.54        11.8              9.66
               Liberia                           155.6           21.76         3.1              0.46
               Rwanda                            110.1           11.88        112.1            12.54
               Sierra Leone                       87.8           20.24        43.9              9.81
   Postconflict average                           93.9           16.33        27.8              6.08


 Source: Calculated from WDI and OECD Creditor Reporting System databases. http://bit.ly/cps66x




                                                            39
Table 4: ODA disbursements for rule of law in sub-Saharan African countries
                       (annual average 2006-2008)

                                                                             Rule of law
                                                      ODA for rule of law
                             Country                                        share of ODA
                                                         ($ millions)
                                                                                (%)
                 Non-fragile countries
                        Benin                                  9                1.58
                        Botswana                              0.5               0.47
                        Burkina Faso                          7.7               0.74
                        Cape Verde                            1.1               0.68
                        Ethiopia                             13.5               0.43
                        Gabon                                 0.1               0.15
                        Ghana                                 4.4               0.31
                        Kenya                                21.1               1.68
                        Lesotho                               1.7               1.69
                        Madagascar                            4.3               0.44
                        Malawi                                8.9               0.65
                        Mali                                  2.5               0.22
                        Mauritania                            0.8               0.23
                        Mauritius                             0.2               0.21
                        Mozambique                            16                 0.8
                        Namibia                               1.5               0.81
                        Niger                                 3.1                0.5
                        Senegal                               3.2               0.21
                        South Africa                          14                1.61
                        Swaziland                             0.3               0.61
                        Tanzania                             11.7               0.41
                        Uganda                               30.4               1.59
                        Zambia                                5.5               0.42
                  Non-fragile average                        7.02               0.71
                 Fragile (but not postconflict) countries
                        Cameroon                              3.4               0.17
                        Chad                                  1.4               0.38
                        Congo, Rep.                           0.6               0.25
                        Equatorial Guinea                     1.3               3.75
                        Eritrea                                1                0.72
                        Gambia                                0.7               0.88
                        Guinea                                0.6               0.17
                        Nigeria                              18.9               0.89
                        Sao Tome & Principe                   0.1               0.21
                        Togo                                  1.3               0.88
                        Zimbabwe                             11.9               2.88
                  Fragile average                             3.7               1.02
                 Postconflict countries
                        Angola                                4.2               1.31
                        Burundi                               4.9               1.04
                        Central African Rep.                   2                1.01
                        Congo, Dem. Rep.                     23.5               1.37
                        Cote d'Ivoire                         1.8               0.42
                        Guinea-Bissau                         1.9               1.55
                        Liberia                               2.8               0.36
                        Rwanda                               22.9               2.49
                        Sierra Leone                         16.9               3.76
                  Postconflict average                        9.0               1.48


Source: Calculated from WDI and OECD Creditor Reporting System databases. http://bit.ly/cps66x




                                                      40
   Table 5: ODA Disbursements: Total, Postconflict Countries, and Fragile States,
                                      2007
                                             ODA to       Percentage total      ODA to        Percentage
                                 Total
                                           postconflict       ODA to            fragile      total ODA to
          Donor                 ODA ($
                                           countries ($     postconflict      countries ($      fragile
                                million)
                                             million)        countries          million)       countries
United States                    13698         2222             16.2              3733           27.2
EC                                9018         2186             24.2              2135           23.7
IDA                               7450         1479             19.9              1653           22.2
United Kingdom                    3545         1335             37.6              1116           31.5
Japan                             4477          684             15.3              849            19.0
Germany                           6391          549              8.6              1559           24.4
Norway                            1750          503             28.7              580            33.2
France                            5206          496              9.5              1234           23.7
Netherlands                       2317          427             18.4              534            23.0
Belgium                           782           333             42.5              390            49.9
Spain                             2197          327             14.9              247            11.2
Global Fund (GFATM)               1600          295             18.4              374            23.4
Sweden                            1524          285             18.7              520            34.1
Canada                            1995          254             12.8              753            37.8
Denmark                           1257          239             19.0              283            22.5
AfDF (African Dev. Fund)          1100          230             20.9              105             9.5
UNICEF                            692           196             28.4              262            37.9
Switzerland                       704           186             26.4              230            32.6
Ireland                           613           174             28.4              193            31.5
Australia                         1768          106              6.0              456            25.8
UNDP                              428           101             23.5              164            38.2
Austria                           1178          86               7.3              116             9.9
Italy                             1121          73               6.5              153            13.6
Arab Countries                    538           69              12.9               83            15.4
Finland                           374           67              18.0              130            34.8
Korea                             431           67              15.6              103            24.0
IFAD                              322           67              20.8               98            30.4
Portugal                          229           62              27.2              122            53.4
Arab Agencies                     425           62              14.6               55            12.9
WFP                               230           60              25.8               77            33.3
UNFPA                             214           57              26.5               72            33.6
UNTA                              278           54              19.5               83            29.7
Greece                            174           47              27.0               45            26.0
UNHCR                             114           38              33.2               36            32.0
Luxembourg                        212           34              15.9               38            17.9
AsDF (Asian Dev. Fund)            1182          33               2.8              313            26.5
GEF                               150           24              16.1               15            10.1
Slovak Republic                    23           19              82.4               19            84.8
Turkey                            522           19               3.7              165            31.6
MONTREAL PROTOCOL                  94           19              19.8               0              0.0
Czech Republic                     61           14              23.7               21            35.0
New Zealand                       164           10               6.0               54            33.1
Nordic Dev. Fund                   68            7              10.6               7             10.8
Iceland                            23            3              13.0               5             21.8
Hungary                            33            1               3.9               8             25.2
EBRD                               6             1              15.3               1             16.7
CarDB (Caribbean Dev.              37            0               0.0               0              0.0
Bank)
UNRWA                             700            0               0.0               0               0.0
Poland                            154            -1             -0.7               12              7.8
IDB Spec. Fund                    206           -15             -7.2               94             45.8
SAF+ESAF+PRGF(IMF)                -72           -35             48.9               49            -68.8
Total                            77703        13548             17.4             19346           24.9
                           Source: OECD Creditor Reporting System database.


                                                  41
                                                                      Figure 1

                                                      Aid to post-conflict countries
                                                         (% of ODA worldwide)
             30




             25




             20
Percentage
    %




             15




             10




             5




             0
                  1989   1990   1991   1992   1993   1994   1995   1996   1997   1998     1999   2000      2001   2002   2003   2004   2005   2006   2007

                                                                                   Year

                                                               Formal peace               Negative peace




                                  Source: Calculated from OECD-DAC database.




                                                                              42
                                                                 Figure 2

                                              Aid to selected postconflict countries
                                        ODA $/person/year averaged over first 5 post-conflict years
                    400



                    350



                    300



                    250
Constant 2007 US$




                    200



                    150



                    100



                    50



                     0
                           Guatemala        Burundi         El Salvador     Sierra Leone     Mozambique     Bosnia and
                          (1997-2001)     (2003-2007)      (1992-1996)      (2000-2004)      (1993-1997)   Herzegovina
                                                                                                           (1996-2000)




                                        Source: Calculated from OECD-DAC database.




                                                                      43
                                                             Figure 3


                                               Timing of post conflict aid
                      450



                      400



                      350



                      300
(constant 2007 US$)
   Aid per capita




                      250                                                                  Bosnia and Herzegovina
                                                                                           Burundi

                      200                                                                  El Salvador
                                                                                           Guatemala
                                                                                           Mozambique
                      150
                                                                                           Sierra Leone


                      100



                      50



                       0
                            1   2     3    4       5          6           7   8   9   10

                                               Years after peace accord




                                    Source: Calculated from OECD-DAC database.




                                                                   44
                                                  Figure 4

                      Average aid to “fragile” versus other countries
                              ($ per capita in 2007; low-income SSA countries)
120




100




80




60




40




20




 0
      Non-fragile countries                Fragile countries (not postconflict)   Postconflict countries
             (n=16)                                      (n=10)                          (n=4)




                 Source: Net ODA disbursements from OECD-DAC database;
                           “fragile” as per World Bank definition.




                                                        45
                                                 Figure 5
                            Aid to low-income West African countries, 2007:
                                    Postconflict versus fragile states
                                                 ($ per capita)


250.0




200.0




150.0
                                                                                        Guinea-Bissau
                                                                                         Postconflict
                                                                                         Fragile
                                                                                        Togo


100.0




 50.0




  0.0
        Guinea-Bissau   Sierra Leone   Liberia      Togo          Guinea   The Gambia




                                                      46
                                                                                Figure 6



                                    Aid to Fragile, Postconflict, and Other Low-Income Countries in
                                                     Sub-Saharan Africa, 1993-2008
                                                                        (average ODA per capita)
                       160



                       140



                       120



                       100
Total ODA per Capita
 (constant 2008 $)




                                                                                                                                      Fragile, Not PostConflict
                       80
                                                                                                                                      PostConflict
                                                                                                                                      Other
                       60



                       40



                       20



                        0
                             1993   1994   1995   1996   1997   1998   1999   2000   2001   2002   2003   2004   2005   2006   2007




                                            Source: Net ODA disbursements from OECD-DAC database;
                                                      “fragile” as per World Bank definition.




                                                                                      47
         Figure 7




Source: OECD-DAC database.




            48
                                                                                           Figure 8
                                  Tax revenue in selected postconflict countries compared to regional averages
                                                                (% of GDP, 2007)
                             25




                             20
 tax revenue/GDP




                             15




                             10




                              5




                              0
                                                                                                  da




                                                                                                                                               l
                                                                                                                              an




                                                                                                                                                                                         a
                                           r




                                                                               a




                                                                                                                                                                                        **
                                                                                                   *
                                        an



                                          a




                                                                                                                               ia




                                                                                                                                                                      ic
                                                                                                                                             pa
                                       do




                                                                                                 go




                                                                                                                                                                                       di
                                        al




                                                                            ric




                                                                                                                                                                  cif




                                                                                                                                                                                      te
                                                                                                                            As
                                                                                                an




                                                                                                                           ist
                                      be




                                                                                                                                                                                    bo
                                                                                                                                          Ne
                                      m


                                    lva




                                                                                              on
                                                                         Af




                                                                                                                                                                                   es
                                                                                                                                                                Pa
                                                                                                                         an
                                   te




                                                                                             Ug




                                                                                                                     h
                                 rib




                                                                                                                                                                                  m
                                                                                            .C




                                                                                                                                                                                 -L
                                Sa




                                                                                                                  ut
                                                                        n
                                ua




                                                                                                                       gh




                                                                                                                                                             &


                                                                                                                                                                              Ca
                              Ca




                                                                     ra




                                                                                                                                                                               or
                                                                                                                So
                                                                                         ep
                             G


                             El




                                                                                                                                                           a
                                                                   ha




                                                                                                                     Af




                                                                                                                                                                             m
                                                                                                                                                            i
                         &




                                                                                       .R




                                                                                                                                                         As




                                                                                                                                                                           Ti
                                                                 Sa
                       ica




                                                                                      m




                                                                                                                                                      st
                                                              b-
                     er




                                                                                   De




                                                                                                                                                    Ea
                                                            Su
                   Am
   tin




                              *DRC estimate is for the year 2005. Source: IMF, "Democratic Republic of the Congo: Poverty Reduction Strategy Paper," September 2007, Table 12.
La




                              **Timor-Leste estimate is for non-oil revenue as % percent of non-oil GDP. Source: IMF, "Democratic Republic of Timor-Leste: 2008 Article IV Consultation," June 2008,
                              T bl 3




                                                                                                  49
                                               Figure 9
                                 Tax revenue in Guatemala, 1995-2006
                                           (taxes as % of GDP)
14




12




10




8
                                                                                              Total tax
                                                                                              revenue
                                                                                              Customs
6




4




2




0
     1995   1996   1997   1998   1999   2000   2001        2002   2003   2004   2005   2006




                                                      50
                                         Figure 10
              Ratio of military expenditure to public health expenditure, SSA, 2006

7




6




5




4




3




2




1




0
    Burundi             Angola              Cote d'Ivoire          Sierra Leone       Average of all other Sub-
                                                                                      Saharan Africa countries




                                              51
             Figure 11



Aid and the state-society contract

                 taxes


 state                   society

              services

             ?


                 aid




                  52
                                                         Figure 12a
                                                        Aid and Taxes:
                                  Low and lower middle income countries in sub-Saharan Africa
                                                          (2006-2008)

                         60

                                            Lesotho


                         50




                         40
Tax revenue (% of GDP)




                         30


                                           Ghana
                                                      Cape Verde
                         20
                                          Kenya     Zambia
                                         Togo              Mali
                                   Cote d'Ivoire Benin
                                                     Niger Uganda                       Sierra Leone*
                         10                Burkina Faso
                                                        Madagascar




                         0
                              0             5           10              15        20   25          30
                                                                 Aid (% of GDP)

                              * Sierra Leone data refer to 2004-2006.



                                  Source: Authors’ calculations, using WDI data.




                                                               53
                      Figure 12b




Source: Moss, Pettersson and van de Walle (2006, p. 12).




                          54
                                                                 Figure 13
                              External debt burdens: Debt/GDP ratios for postconflict countries in Africa, 2007

               4



              3.5



               3



              2.5
current US$




               2



              1.5



               1



              0.5



               0
                    Liberia          Guinea-Bissau     Burundi     Congo, Dem. Rep.   Cote d'Ivoire      Average of all    Angola
                                                                                                      other Sub Saharan
                                                                                                       African countries




                                                                     55
                                               Figure 14
                        Ratio of debt service to public health expenditure, SSA, 2006

8



7



6



5



4



3



2



1



0
    Guinea-Bissau   Burundi         Angola      Congo, Dem. Rep.   Cote d'Ivoire   Sierra Leone   Average of all other
                                                                                                  Sub-Saharan Africa
                                                                                                      countries




                                                    56