SITE Working Paper No. 1, 2009
Early vs. Late in Aid Partnerships
and Implications for Tackling Aid
Stockholm Institute of Transition Economics
Stockholm School of Economics
Sveavägen 65, P.O. Box 6501
SE – 113 83 Stockholm
Early vs. Late in Aid Partnerships and
Implications for Tackling Aid Fragmentation
Development aid donors disburse aid to many developing coun-
tries. This paper shows that whether a partnership is established
early or late matters significantly for aid quantities. Donor countries
allocate larger shares of their aid budgets to recipients that entered
early in their portfolios. This effect is large compared to variations
due to recipients’ income differences, and matters even in the long
run. Entry dates are weakly related to GDP per capita, but are in-
fluenced strongly by colonial past. On the other hand, colonial rela-
tionships explain only a small part of the observed variation in entry
dates. These findings imply that donors, while continuously increas-
ing their number of recipients, have allocated smaller aid quantities
to new partnerships. This has direct consequences for aid fragmen-
tation, with many donors disbursing small amounts to a recipient. I
study a simple reform that eliminates “small” partnerships, but leaves
unaffected donor aid budgets and developing countries receipts. The
reform reshuffles only about 20 percent of all the aid disbursed in a
year but drastically reduces fragmentation.
JEL Codes: F35, O19.
Keywords: Aid, Fragmentation.
∗SITE, SSE, P.O. Box 6501, SE-11383 Stockholm, Sweden.
Developed countries have been disbursing foreign aid to developing countries
for many years. Yet little is known about how aid partnerships evolve over
time, whether or how donors shift priorities, or whether they keep a small
core set of partners, or reach out to new partnerships as their a