Conventional treatments of fungibility, such as in Assessing Aid, are conce
rned with evidence that aid recipients do not increase sufficiently (that i
s, by the amount of the aid) expenditure on specific areas favoured by dono
rs. In other words, fungibility implies that recipients divert aid to expen
diture on areas donors did not wish to fund. However, there is evidence tha
t aggregate expenditure, and even spending on donor-supported areas, rises
by more than the value of the aid inflow. This contribution, using insights
from public choice research on fiscal illusion, provides a range of theore
tical scenarios to explain this outcome. Included are scenarios where, even
where all the features of fungibility are present, expenditure on areas fa
voured by the donor can increase by more than the value of the aid inflow.
The study concludes by suggesting new directions for research on aid policy
and the impact of aid on the public sector in developing countries.