Elasticity and revenue: a reappraisal

It seems from what is explained in some textbooks that, for a given demand function, price and total revenue move in the same direction when elasticity is smaller than one and move in opposite directions when elasticity is greater than one, with elasticity from point (p, q) to point (p', q') of a demand function defined as – [(q ' – q) / q] / [(p' – p) / p]. Since these two results turn out to be false, this comment clarifies the relationship between elasticity (as previously defined), price movements and changes in total revenue. This paper is the result of the following allocation of tasks: Ted Bergstrom did the part that is interesting, correct and worth reading, whereas I did the rest.