Aid effectiveness in Latin America and the Caribbean (LAC) has been little studied, despite the fact that it is the developing region receiving foreign aid with the highest per capita income and inequality levels. This paper uses a growth regression model to analyze the impact of Official Development Assistance (ODA) in LAC. We evaluate ODA effectiveness in relation to the growth rate of an ‘inequality-adjusted GDP per capita’ in order to precisely define the desired impact of aid in a region with high levels of inequality. The estimation produces three main results: aid is effective, in aggregated terms, once we deal with the effect of income inequalities; the impact of concessional loans seems to be greater than the impact of grants; and, aid may be more effective in less corrupt countries.
Economic research has paid significant attention to the connection between foreign
aid for development and economic growth. Since the 1960s several development
economists, starting with Paul Rosenstein-Rodan and Hollis Chenery, have
claimed that the effectiveness of public international development policies — socalled Official Development Assistance, ODA — should be evaluated in terms of
the stimulus finally exerted on the developing countries’ growth rate of per capita
income. Nevertheless, after more than 50 years of research it is still debatable to
conclude that ODA stimulates growth in aggregate terms. In contrast with previous
studies, this paper deals with three issues that have been insufficiently considered
in the aid effectiveness literature:
Firstly, the majority of aid-effectiveness studies have analyzed the whole
group of developing countries, paying limited attention to the analysis of regional
experiences. Specifically, aid impact in Latin America and the Caribbean (LAC)
— the developing region with the highest per capita income, but also with the
highest inequality levels among those which receive aid — has been little studied,
and it has only been pointed out that this region is an outlier in the general models
of aid effectiveness.1
Secondly, preceding studies have analyzed the impact of aid on recipient
countries’ GDP per capita growth rate. However, this method involves a ‘target
problem’, which has not been previously considered in the literature: it is not
possible to distinguish if aid benefits the income growth of the relatively richer or
poorer citizens, precisely when the latter citizens are, indeed, the target population
of aid policies.2 In this context, the intra-national distribution of foreign aid
should help to reduce income inequalities. This is the reason why we evaluate aideffectiveness
in terms of the growth rate of the GDP per capita for the population
with lower incomes, which we call the ‘inequality-adjusted GDP per capita’.
Thirdly, previous studies have analyzed the aggregate ODA impact, thus
neglecting the fact that different aid modalities may have dissimilar impacts
on growth. In contrast, we disaggregate into ODA grants and ODA loans, and
analyze the financial characteristics and productive incentives that may affect their
potential impact on growth.
All in all, this article aims to analyze the impacts of ODA grants and loans on
the inequality-adjusted rate of growth of LAC countries’ income per capita duringthe period 1992–2007. The second section briefly reviews the recent economic
literature on aid effectiveness. The third section proposes an analytical model of
aid impact on growth, adapted to the peculiarities of the American region, and
based on growth theory. The fourth section presents the results obtained from the
estimation of the ODA’s impact model. The last section summarizes the main
conclusions of this piece of research and suggests some economic policies that
may increase the effectiveness of aid disbursed to LAC.
LAC countries are among the developing economies with highest levels of
income per capita, but also the economies with highest levels of inequality. The
region has taken part in the international co-operation system since its origins,
receiving since 1960 a total ODA flow that amounts to 0.48% of the regional
GDP. Obviously, from these limited resources it is not reasonable to expect an
outstanding impact, but invested strategically they should effectively support the
development strategies of these countries.
The aim of this paper is to quantify the impact that ODA has exerted on the
LAC countries’ pace of growth during the period 1992-2007. For this reason, we
run a regression model of the aid impact on growth, adapted to the peculiarities
of the region and based on modern growth theory. The model evaluates aid
effectiveness in relation to the growth rate of the GDP per capita within the
population with income lower than that of the ninth decile (‘inequality-adjusted
GDP per capita’), in order to precisely define the desired impact of aid in a regionwhich is characterized by high inequality levels. In contrast with other previous
studies, we limit our analysis to the sample of Latin American countries in order
to capture the peculiarities of their growth dynamics, and we distinguish between
two aid modalities — ODA grants and ODA loans — that may have different
impacts on growth.
The econometric estimation of the model points out three relevant results in
relation to aggregate aid effectiveness for the period 1992-2007. Firstly, ODA has
been effective in stimulating the rate of growth of the inequality-adjusted GDP per
capita. Furthermore, aid impact dilutes when we consider the GDP per capita for
all income deciles. This result suggests that aid has been effectively concentrated
in those Latin American citizens with lower incomes (within each country), which
may reflect a progressive ‘within country’ distribution of aid.
Secondly, the analysis suggests that ODA loans have exerted a greater stimulus
on growth than ODA grants (with average estimated coefficients of 0.27 and
0.44, respectively). This result supports the use of both aid grants and loans in a
middle income region such as LAC, despite its long record of debt unsustainability
problems. Obviously, from this result we cannot infer that concessional loans are
preferable than grants; on the contrary, grants should still be concentrated in those
LAC countries with lesser repayment capacities and more restricted access to
credit. But the use of concessional loans should be enhanced in those economies
which offer guaranties of repayment and need resources for financing productive
activities. Yet this piece of research leaves unresolved the questions about which
socio-economic conditions are more appropriate for the use of loans or grants, and
which kind of development activities — with potentially different growth impacts
— are primarily financed by loans or grants.
Thirdly, aid seems to be more effective in countries with better mechanisms of
corruption control. This result backs up the thesis of previous studies that claimed
the importance of institutions for the effectiveness of aid (such as Burnside and
Dollar 2004; Chauvet and Guillaumont 2004; and Tezanos et al. 2009).
All in all, this paper tries to contribute to the aid-effectiveness literature; a
literature that, after 50 years of research, still offers controversial conclusions
about the potential mechanisms of aid impact on growth. Actually, our study onlytests the ‘macroeconomic effectiveness’ of aid in relation to economic growth, and
not in relation to progress in other dimensions of human development. Therefore,
it should be interpreted as a partial evaluation of aid effectiveness, exclusively
referred to the economic dimension of development.
