Recognizing that it is difficult to prove a negative, this paper marsh
als evidence in support of the hypothesis that the widely documented f
lypaper effect of federal grants to state and local governments is pur
ely a statistical artifact. A review of previous studies and a monte c
arlo investigation suggest that the use of an inappropriate functional
form may generate an illusory flypaper effect. A local expenditure eq
uation is then estimated in alternative specifications with ten years
of state level data. Empirical results confirm the sensitivity of the
flypaper effect to specification, and tests of fit unambiguously favor
one functional form. That specification yields no statistical evidenc
e of a flypaper effect.