Transcript Dirk-Jan Koch`s presentation

Blind spots on the aid map: clustering &
other curious geographic choices of NGOs
DANISH INSTITUTE FOR
INTERNATIONAL
STUDIES
MAY 19, 2008
Dirk-Jan Koch,
Axel Dreher,
Peter Nunnenkamp
and Rainer Thiele
Set-up of presentation
1. Hypotheses
2. Data
3. Method
4. Results
5. Conclusion
Why this research ?
• NGOs are the new aid giants
– 26 billion US dollar in 2005 (OECD)
• Many studies on geographic choices
– of NGOs within countries;
– of other aid actors;
– but not on NGO choices between countries.
• Between country differences are substantial
• Why quantitative approach ?
– Exploratory chapter of larger research;
– NGOs operate within resource transfer paradigm.
Hypothesis 1: Poverty
NGO aid is focused on the needy countries
(economic and social dimension of poverty, both in
relative and/or absolute terms, such as assessed by
GDP, HDI or GINI)
Hypothesis 2: Governance
NGOs are relatively strongly engaged in countries
with weak institutions in order to exploit their
comparative advantage of working in “difficult”
environments
(on technical and political aspects as assessed by
Freedom House Indicators, Kaufmann indicators and
Polity IV democracy measures)
Hypothesis 3: Back-donor preferences
The preferences of official backdonors affect the
allocation of NGO aid. Yet NGOs are not affected
by geo-political interests of backdonors
(backdonor preferences assessed by aid flows, geopolitical interests by trade to recipient counties and UN
General Assembly Voting)
Hypothesis 4: Concentration
Hypothesis 4: NGOs locate where other NGOs are
active.
(assessed by the numbers of other international NGOs
that are active in the country and the amounts those
other NGOs disburse)
Hypothesis 5: Likemindedness
Hypothesis 5: NGOs are more strongly engaged in
countries characterized by similarities with their
own organization.
(assessed by religious and colonial matches)
Data
1. 100 largest international NGOs were invited to
provide data; 61 did:
2. Total budget of NGOs in the sample: appr. 7 billion
€
NGO aid per capita per continent
0
.02
.04
.06
in Euro in 2005
Africa
Asia
Middle East
Latin America
Europe
NGO aid per capita
0
.02
.04
.06
in Euro in 2005
LDC
OLIC
LMIC
UMIC
LDC = Least Developed,OLIC = Other Low Income,LMIC = Lower Middle Income,UMIC, Upper Middle Income
Top recipients
International NGO
International NGO
expenditures in million Euro expenditures in Euro per
(2005)
capita (2005)
India
262 Palestine
12,3
Ethiopia*
175 Lesotho*
11,4
Sudan*
150 Zimbabwe
9,7
Indonesia
135 Nicaragua
9,1
Kenya
126 Haiti*
8,8
Zimbabwe
126 Swaziland
8,3
Bangladesh*
118 Zambia*
7,6
Uganda*
109 Malawi*
7,3
Sri Lanka
95 Honduras
6,7
Malawi*
93 El Salvador
6,6
* Indicates that a country was labelled a Least Developed Country by the OECD/DAC in 2005
Methods
Multivariate regression analysis
• What determines if NGOs become active in a
country?
• What determines how active NGOs become in a
country ?
Results regarding poverty focus not clearcut
• NGOs do not consider levels of poverty when
deciding on becoming active in a country.
• NGOs do consider poverty when determining level
of intervention in a country (1% increase in GDP
leads to a decrease of 0.15% of aid)
Strong results: 75% of choices can be explained
Governance determinant: rejected
NGOs do not work more in countries with poor governance
Back-donor preferences: accepted
NGOs become (more) active in countries where backdonors are
active (but not susceptible to other donor interests).
Concentration determinant: accepted
NGOs become (more) active in countries where other NGOs are
active
Likemindedness determinant: accepted
NGOs become active in those countries with which they share
key-characteristics.
Conclusion
• Good news:
– Some poverty targeting
– Not influenced by geo-strategic interests.
• Bad news:
– Concentration
– Donor led behavior
– Likemindedness determines NGO choices
Questions, comments ?
[email protected]
NGO donor darlings (e.g. Guatemala, Kenya, Malawi and Sri
Lanka, Zambia and Uganda) receive > 100 million USD annually
NGO donor orphans (e.g. Côte d’Ivoire, Congo-Brazzaville,
Guinea, Yemen and CAR receive < 10 million USD annually)