The managers of modem health services rely on economic modelling in ma
ny ways. Yet these economic models may not be comprehensive enough to
answer the questions asked of them. The economic theories that are use
d to construct these models are often based on an economic orthodoxy w
hich has a questionable claim to reflect reality. This orthodoxy has b
een criticised for its narrowness, its social assumptions, its economi
c assumptions, and its quasi-scientific quantitative techniques. Such
critiques may have succeeded in undermining economic rationalism and o
ther economic formalisms at an academic level, but they have been prac
tically ignored in the health policy debate. The way in which health i
s actually produced and damaged is substantially different from the wa
y in which it is modelled by economic theory. To a very large extent,
this reflects an element of quasi-scientific oversimplification which
is unavoidable in rigid, quantitative models, and which must be temper
ed by an understanding of their faults and strengths, and by large mea
sures of sociological realism in our thought and political flexibility
in our actions. In particular, the hope that economics can provide a
framework for solving all problems of resource allocation - which is t
o say, almost all problems of health policy and health services manage
ment - should be abandoned.