J. Heckman, INSTRUMENTAL VARIABLES - A STUDY OF IMPLICIT BEHAVIORAL ASSUMPTIONS USED IN MAKING PROGRAM EVALUATIONS, The Journal of human resources, 32(3), 1997, pp. 441-462
This paper considers the use of instrumental variables to estimate the
mean effect of treatment on the treated, the mean effect of treatment
on randomly selected persons and the local average treatment effect.
It examines what economic questions these parameters address. When res
ponses to treatment vary, the standard argument justifying the use of
instrumental variables fails unless person-specific responses to treat
ment do not influence decisions to participate in the program being ev
aluated. This requires that individual gains from the program that can
not be predicted from variables in outcome equations do not influence
the decision of the persons being studied to participate in the progra
m. In the likely case in which individuals possess and act on private
information about gains from the program that cannot be fully predicte
d by variables in the outcome equation, instrumental variables methods
do not estimate economically interesting evaluation parameters. Instr
umental variable methods are extremely sensitive to assumptions about
how people process information. These arguments are developed for both
continuous and discrete treatment variables and several explicit econ
omic models are presented.