This paper warns against the use of Marshallian welfare loss in applied ana
lysis of market power. We show how to compute the Hicksian deadweight loss
from an ordinary demand. Then, from an experiment using real data on twenty
-one sectors of the French food industry, we find that the Marshallian dead
weight loss poorly approximates the exact Hicksian measure. Hence, it is ad
visable to use the latter in applied welfare analysis.