Does private aid follow the flag? An empirical analysis of humanitarian assistance

• Author: Andreas Fuchs,Hannes Öhler
• DOI: http://doi.org/10.1111/twec.13021
• Published date: 01 March 2021
• Date: 01 March 2021

World Econ. 2021;44:671–705.

|

671

wileyonlinelibrary.com/journal/twec

Received: 23 January 2020

|

Revised: 26 July 2020

|

Accepted: 6 August 2020

DOI: 10.1111/twec.13021

ORIGINAL ARTICLE

Does private aid follow the flag? An empirical

analysis of humanitarian assistance

AndreasFuchs1,2

|

HannesÖhler3

1University of Göttingen, Göttingen, Germany

2Kiel Institute for the World Economy (IfW), Kiel, Germany

3German Development Institute/Deutsches Institut für Entwicklungspolitik (DIE), Bonn, Germany

Funding information

Deutsche Forschungsgemeinschaft, Grant/Award Number: FU 997/1-1

KEYWORDS

aid allocation, corporations, disaster relief, foreign aid, humanitarian assistance, non-governmental organisations, private

donors, private foundations

1

|

INTRODUCTION

Non-state donors are becoming increasingly more important for providing funds in response to

humanitarian needs, fostering economic development and shaping development policy (Desai &

Kharas,2008, 2018; Esser & Bench, 2011; Metzger, Nunnenkamp, & Mahmoud,2010; Werker &

Ahmed,2008). The Bill and Melinda Gates Foundation, for example, contributed over US$26.1bil-

lion over the 2009–2017 period, which is a tenth of the official aid budget of the United States, the

foundation's home country (OECD,2019). Policymakers, development scholars and recipient govern-

ments relish such private aid flows, which originate from non-governmental organisations (NGOs),

companies and charitable individuals, among others (Lundsgaarde,2013; White,2012). In paragraph

41 of the 2030 Agenda for Sustainable Development, the international community “acknowledges[s]

the role of the diverse private sector, ranging from micro enterprises to cooperatives to multinationals,

and that of civil society organizations and philanthropic organizations in the implementation of the

new Agenda” (United Nations,2015).

This paper focuses on the role of private donors in humanitarian assistance. The growing frequency

and severity of natural disasters and militarised conflicts raises demand for additional financial re-

sources in response to humanitarian crises (Global Humanitarian Assistance,2018). Official donors

are often unable or unwilling to provide the required funds. Becerra, Cavallo, and Noy (2014) and

This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and

distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are

made.

© 2020 The Authors. The World Economy published by John Wiley & Sons Ltd

672

|

FUCHS and ÖHLER

Becerra, Cavallo, and Noy (2015) show that official aid surges in the aftermath of natural disasters are

low compared to the economic damages caused. A prominent example is the aftermath of the 2004

Indian Ocean earthquake and tsunami where the majority of humanitarian assistance originated from

private sources.1 In fact, US companies alone mobilised more than US$565 million (Thomas &

Fritz,2006). Furthermore, official bilateral aid is generally perceived as bureaucratic, slow and polit-

ically driven (e.g., Lancaster,2007). Previous empirical research finds that the allocation patterns of

official aid are not only determined by recipient countries’ needs and performance in terms of their

development policies but also by the political and economic interests of the government of the donor

country (e.g., Alesina & Dollar,2000; Faye & Niehaus,2012; Hoeffler & Outram,2011). This finding

has been confirmed by studies that focus on humanitarian aid exclusively (e.g., Annen &

Strickland, 2017; Bommer, Dreher, & Perez-Alvarez, 2018; Fink & Redaelli, 2011; Raschky &

Schwindt,2012).

Hence, private donors, and in particular NGOs, have often been perceived as more need-oriented

than official donors. For example, Desai and Kharas (2008: 161) highlight that “while official donor

allocations are influenced by, among other things, political coalitions, policy concerns, and colonial

ties, NGO allocations are assumed to be influenced by need”. In addition, private donors are supposed

to have a comparative advantage in difficult environments as they can more easily circumvent corrupt

governments and deal with local target groups directly (Riddell, Bebbington, & Peck,1995).2 However,

critics suspect that private donors rather imitate the allocation of official aid. In the case of NGOs,

being financially dependent on official financiers is expected to undermine the autonomy of NGOs in

allocating aid. According to Edwards and Hulme (1996: 970), the relations of NGOs with state agen-

cies are “too close for comfort”—with NGOs often becoming “the implementer of the policy agendas”

of governments. Furthermore, official financiers may predominantly cofinance NGO projects that are

located in their favoured countries (Dreher, Nunnenkamp, Öhler, & Weisser,2012). Along similar

lines, corporations may provide humanitarian assistance to their government's favoured locations to

obtain favours in return. In this spirit, Bertrand, Bombardini, Fisman, and Trebbi (2020) describe

corporate philanthropy as an alternative to campaign contributions and lobbying activities for compa-

nies that seek to influence policies.

This paper studies the determinants of the allocation of humanitarian aid made by NGOs and

corporate donors from a large set of countries. More specifically, we test whether private aid tends to

follow the humanitarian aid allocation pattern made by the government of their home country, that

is, whether private humanitarian aid tends to “follow the flag”. A better understanding of the extent

to which private donors follow their home countries’ official aid allocation pattern enables us to

assess the independence of private aid decisions. If private donors closely follow the aid allocation

made by official donors, they might not be able to exploit their supposed advantage of being more

need-oriented than their official counterparts. Moreover, if official and private donors are clumping

together their aid activities, this cements the grouping of recipient countries into donor “darlings” and

“orphans” (Davies & Klasen,2019) and increases the need for donor coordination within recipient

countries.

Although research on foreign aid has been largely focused on bilateral and multilateral official do-

nors, we are not the first to study the aid allocation of private aid donors. However, previous research

has only analysed private aid giving from single donor organisations, from single donor countries, in

1See Kim, Nunnenkamp, and Bagchi(2016) on an analysis of private donations to NGOs as a response to the tsunami.

2In line with this idea, many donor governments tend to use NGOs as implementers of aid projects to bypass governments in

badly governed recipient countries (Acht, Mahmoud, & Thiele,2015; Dietrich, 2013, 2016).

|

673

FUCHS and ÖHLER

single recipient countries or to single disaster events (e.g., Büthe, Major, & De Mello e Souza,2012;

Dreher, Nunnenkamp, Öhler, et al., 2012; Dreher, Nunnenkamp, Thiel, & Thiele,2012; Metzger

etal.,2010). Our paper is closely related to Fink and Redaelli (2011) who focus on humanitarian as-

sistance specifically, as we do in the present study. They analyse the allocation of aggregated private

humanitarian aid across recipient countries. However, they do not disaggregate non-state aid into

corporate and NGO aid, nor do they take account of the home country of the private entity, that is,

they only study the aggregate “private aid” and its correlates. Likewise, Neumayer (2005) analyses

the allocation of food aid by an NGO aggregate, which prevents the study from testing whether NGOs

follow donor country-specific interests.

We thus contribute to the literature in several ways. First, our study is the first multidonor country

multirecipient country panel analysis that tests whether private aid tends to follow the aid allocation

of the respective official donor. For this purpose, we construct a database that offers information on

the country in which the headquarters of the private donors are located and combine it with human-

itarian aid data provided by the Financial Tracking System (FTS) of the United Nations Office for

the Coordination of Humanitarian Affairs (UNOCHA). Second, we improve the identification strat-

egy compared to previous studies by employing panel data methods controlling for various kinds of

heterogeneity, particularly across recipient country-year pairs. This minimises the risk of spuriously

attributing a positive correlation caused by common factors that affect both official and private hu-

manitarian aid giving to a causal relationship between official aid of the home country and private aid.

Third, to further reduce endogeneity concerns, we use a novel instrumental variable (IV) that relies on

variation in the leadership of ministries responsible for official humanitarian aid. Finally, our study is

the first analysis that compares the aid allocations of NGOs and corporate donors.

We also compare the poverty and need orientation of private donors to that of official donors.

Following humanitarian motives, altruistic donors are expected to provide more humanitarian assis-

tance to needier countries (e.g., Büthe etal.,2012; Fink & Redaelli,2011). Countries with good poli-

cies and good institutions could either get more aid flows as a reward (e.g., Burnside & Dollar,2000)

or less aid since good institutions increase the potential ability of countries to deal with humanitarian

crises themselves (e.g., Fink & Redaelli,2011). However, the comparative advantage of working in

difficult environments may lead NGOs to engage themselves in countries with high corruption and

conflict potential (e.g., Riddell et al., 1995). Finally, private and official donors’ aid patterns are

expected to be shaped by institutional, political and economic self-interests such as vote buying in

international organisations or export promotion (e.g., Alesina & Dollar,2000; Dreher, Nunnenkamp,

& Thiele,2008; Drury, Olson, & Van Belle,2005; Edwards & Hulme,1996).

Our results show that the humanitarian aid allocation of private donors tends to follow the alloca-

tion of the respective official donor. This finding holds when we use a binary variable or a recipient's

share in a donor's total aid allocation as dependent variable. It is robust against the inclusion of vari-

ous fixed effects and tackling remaining endogeneity concerns with IV estimations. A disaggregated

analysis shows that this result holds for the allocations of both NGOs (including non-corporate private

foundations) and corporate donors. Donor country-specific estimations reveal statistically significant

evidence that private donors from China, Sweden, the United Kingdom and the United States tend to

“follow the flag”.

Our paper proceeds as follows. Section2 situates our paper in the burgeoning literature on private

donors, elaborates on potential differences in the aid allocation patterns between private and official

donors and discusses why NGOs and corporations have incentives to follow their government's aid

allocation. In Section3, we introduce a new data set on the home countries of private donors, all other

data used, descriptive statistics and our estimation strategy. Section4 presents our results. We sum-

marise and conclude in Section5.