5. The aid-growth literature
The literature on aid-growth linkages started during the 1
960s with the rising question of whether aid works. «There are some
periods when assessing aid effectiveness becomes particularly intensive. These
coincide with cycles of doubt on the efficacy of aid as an instrument of
foreign development.» (Kanbur, 2003, p 15) Over time, the successive
empirical studies have given some strong contradictions. Some authors have
tried to structure chronologically the debate. This seems to be an important
step to clarify the ideas on this issue. We will here discuss two of these
reviews.
5.1. The optimism of Hansen and Tarp (2000)
In a quite optimistic article, Hansen and Tarp (2000)
distinguish three generations among 131 cross-country regressions. They
essentially consider the methodology applied to measure the efficiency of aid.
The first generation looked at the impact of foreign aid on domestic savings.
Already, the question turned into debate. After more than a decade of
consensual optimism, Griffin (1970), Griffin and Enos (1970) and Weiddkopf
(1972) launched the discussion with their pessimistic conclusions. They found
rates of growth negatively linked with the amount of foreign capital inflows.
They argued that aid may retard development by decreasing domestic savings. To
retort to this «revisionist» literature, Papanek (1972) was the first
to regress growth on aid. After disaggregating capital flows into foreign aid,
private capital and other inflows, he found a strongly significant positive
correlation between aid and growth for a sample of 51 countries observed from
1950 to 1965. Provided the convincing arguments of Papanek, Hansen and Tarp
(2000, p7) conclude: «The overwhelming evidence from these studies is that
aid lead to an increase in total savings, although not by as much as the aid
flow... aid spurs growth.»
The second generation of studies switched to the
aid-investment relationship to estimate the impact on growth. This period was
also considered by Hansen and Tarp (2000) to be optimistic since all but one
article show a positive impact of aid on growth. The only exception was the
much cited Mosley et al. (1987, 1992). In consequence, the conclusion of the
authors was that the aid-investment link was positive, leading probably to a
higher growth rate. As we will see later, Hansen and Tarp's positive perception
of the early literature is a bit surprising. Other authors conclude at best
that the first decades of research about the aid-growth linkages was rather
inconclusive9.
Finally, by the end of the nineties, analyses became more and
more sophisticated. With one emblematic article of Burnside and Dollar (1997),
this can be considered as the third and last generation of studies. Hansen and
Tarp (2000) review a few of these recent papers10 but found only one
outlier11 concluding that aid has no effect on growth. All the
remaining articles they analyse found a significant impact of aid on growth
either conditional on appropriate economic policy or not. This allows them to
say about aid that: «There is a significant effect on growth, either alone
or in combination with a policy variable.» (Hansen and Tarp, 2000, p16).
Before providing our own discussion on all this literature we would like to
account for another way to analyse its evolution.
| 
