A number of empirical studies have shown that there is a negative asso
ciation between population:physician ratio and utilization of medical
services. However, it is not clear whether this relationship reflects
supplier-inducement, the effect of lower prices on patient demand, a s
upply response to variation in health status, or improved availability
. In Norway, patient fees and state reimbursement fees are set central
ly. Therefore, the correlation between utilization and population:phys
ician ratio either reflects supplier-inducement, a supply response or
an availability effect. We applied a theoretical model which distingui
shed between an inducement and an availability effect. The model was i
mplemented on a cross-sectional data set which contained information a
bout patient visits and laboratory tests for all fee-for-service prima
ry care physicians in Norway. Since population:physician ratio is pote
ntially endogenous, an instrumental variable approach is used. We foun
d no evidence for inducement either for number of visits or for provis
ion of laboratory services. (C) 1998 John Wiley & Sons, Ltd.