Explaining development aid allocation by growth
| Pages | 21-41 |
| Published date | 19 April 2013 |
| Date | 19 April 2013 |
| DOI | https://doi.org/10.1108/20452101311318657 |
| Author | Hristos Doucouliagos,Martin Paldam |
| Subject Matter | Strategy |
Explaining development aid
allocation by growth
Hristos Doucouliagos
School of Accounting, Economics and Finance, Deakin University,
Burwood, Australia, and
Martin Paldam
School of Economics and Management, University of Aarhus,
Aarhus, Denmark
Abstract
Purpose – The purpose of this paper is to study a little researched relation: the relation from
economic growth in a less developed country to the development aid it receives. Does economic growth
influence donor aid allocation decisions?
Design/methodology/approach – The authors’ apply two different methodologies. First, a
quantitative and systematic review is presented of the literature of 30 empirical studies of aid
allocation where a growth coefficient is estimated. Second, a primary study is presented of the data
using a panel of 147 countries for the period 1967-2004.
Findings – The growth-aid relation should be negative if humanitarian motives dominate aid
allocation decisions. The result from both the meta-analysis and the primary data analysis suggests
a very small effect between lagged growth and aid allocations, with a dominating positive sign. This
result appears to be driven partly by the large development banks.
Originality/value – No attempt has previouslybe en made to summarize the literature on growth as a
motive for giving aid. This paper offers the first attempt to do so, by presenting a meta-analysis of the
empirical literature, as well as analysis of the primary data.
Keywords Aid allocation, Growth, Meta-analysis, Economic growth, International aid,
Developing countries, Banks
Paper type Research p aper
1. Introducing a little known relation
Entrepreneurship, innovation, and capital formation are indispensible drivers of
economic growth and economic prosperity. Developing countries often lack the
necessary institutional arrangements, legal and regulatory framewo rk, and policies to
provide sufficient incentives to promote entrepreneurship and investment. Many
developing countries rely on development aid, which for many forms a significant
part of the institutional environment in which entrepreneurs, innovators, and investors
operate. It is now apparent that development aid has largely been ineffective in
stimulating investment and growth (e.g. Doucouliagos and Paldam, 2011). Rather than
development aid creating growth, it is possible that economic growth itself may
influence donor aid allocation decisions.
A large empirical literature has emerged to explore the motives behind aid
allocations. One of the more important donor motives is the recipient’s humanitarian
needs, typically measured by per capita income or population, and less commonly so
by economic growth. This paper considers the causal relation from economic growth in
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/2045-2101.htm
Journal of Entrepreneurship and
Public Policy
Vol. 2 No.1, 2013
pp. 21-41
rEmeraldGroup PublishingLimited
2045-2101
DOI 10.1108/20452101311318657
The authors’ cooperation has been assisted by a “Globalization Grant” from the University of
Aarhus. Pia Wichmann Christensen has been a very competent research assistant.
21
Development
aid allocation
a recipient country to the development aid it receives: the growth-aid relation, with the
slope, j.The literature on aid allocations has developed four hypotheses about j:
H1. Humanitarian interests: poor development is a humanitarian concern. Predict
jo0.
H2. Development banking interests: development banks are charged with the
financing of worthwhile development projects. Economic growth generates
many such projects. Predict j40.
H3. Commercial interests: donor country business sees aid as a public investment in
their future business. Growth makes countries more promising. Predict j40.
H4. It is the result of modeling biases: typical prediction jo0 (see Section 2).
The importance of humanitarian interests (H1) is of particular concer n, given the
primary stated objectives of development aid. The sign on jpredicted by the four
hypotheses differs, so that the net outcome is undetermined and remains an empirical
issue. As the signs differ we suspect that the average size of jis very small.
No attempt has previously been made to summarize neither this literature nor the data
to test these hypotheses. Does economic growth influence donor aid allocation
decisions? This paper offers the first attempt to do so, by presenting a meta-analysis of
the empirical literature, as well as analysis of the primary data.
The meta-analysis provides a quantitative synthesis and systematic review of the
relevant results in the extant empirical studies. We have found 30 studies where jis
estimated[1]: the estimates of jare typically small and positive. Research is a process
of truth searching where new results are produced by innovation, and confidence
is build by independent replication, which is replication by other researcher s on new
data sets. Meta-analysis is a quantitative study of this process, it asks:
Q1. Do the findings of the research process converge to something we can term the
true value of j?
Q2. What factors explain the heterogeneity in the reported empirical results?
To analyze these questions, meta-analysis uses all results reported in the literature as
the data. To study Q1, methods have been developed to study convergence as data
expands, and models and estimators improve. To study Q2, each data point is provided
with a string of information characterizing the way the said estimate is reached. The
string covers data, model specification, and estimation differences. The meta-study
thus analyzes if results change over time, vary across countries, exhibit structu ral
shifts due to innovations, etc.
Our primary data analysis is based on panel data analyses of aid allocations to
147 countries for the period 1967-2004. This involves a larger and more comprehensive
data set than that used in the extant empirical studies. The results from the panel data
analysis confirm those of the meta-analysis.
Section 2 looks at the correlation between growth and aid, and at the causal
structure in the aid-growth-income nexus, and discusses H4. Section 3 discusses the
three hypotheses, H1-H3. The meta-analysis is presented in Sections 4 and 5, where
22
JEPP
2,1
To continue reading
Request your trial