Modeling the Effects of Promotional Efforts on Aggregate Pharmaceutical Demand: What We Know and Challenges for the Future - Innovation and Marketing in the Pharmaceutical Industry - page 591

third column indicates whether or not the model accounts for endogeneity of the
promotional variables. When a study accommodates endogeneity in other variables
than promotional variables (e.g., price) we classify this study as partly accommodat-
ing endogeneity. Studies that investigate endogeneity, and fi nd that this is not a
concern, are labeled as “Investigated.” Column four shows whether heterogeneity of
the promotional effects across brands/categories is accounted for. Some studies do
not estimate brand-specifi c parameters, but report brand-specifi c marginal responses
to marketing variables (induced by, e.g., interaction effects). Such studies are labeled
as partly accommodating heterogeneity. The fi fth column describes the type of data
used and the geographical location of the study. The last column summarizes the
key fi ndings for each study. We categorize these studies according to two levels of
demand (Leefl ang et al. 2000 ): category- and brand level. It is important to distin-
guish between these two levels of demand as pharmaceutical promotions are likely
to affect them to a different extent. Brand level sales are primarily driven by “own”
marketing efforts and to a lesser extent by competitive expenditures. The resulting
sales effect can be seen as a sum of “market stealing” (brand switching) effects and
“market making” (category expansion) effects (see, e.g., De Laat et al. 2002 ),
whereas sales at the product category level are the result of market making effects
only. In Sect. 20.2.1 we review studies that examine product category level effects.
In Sect. 20.2.2 we focus on the effects of marketing activities on brand level demand.

20.2.1 Product Category Level Demand Effects

of Pharmaceutical Promotion

One feature of the product category studies in Table 20.1 is that promotional expen-
ditures have only small effects on (category) demand (Berndt et al. 1995 , 1997 ;
Chintagunta and Desiraju 2005 ; Rosenthal et al. 2003 ). Fischer and Albers ( 2010 )
report primary demand elasticities that are averaged across a large number of
categories. The average short-term elasticities are smaller than 0.05 for all instru-
ments. Long-term elasticities are larger, but still well below one, indicating inelastic
response. The conclusion is that pharmaceutical marketing activities typically have
only moderate effects on the prescription behavior of physicians, consistent with
Manchanda et al. ( 2005 ) and Kremer et al. ( 2008 ). On the one hand, this does not
provide much support for the argument that pharmaceutical marketing is generating
unwanted demand effects by stimulating overprescription and misuse (see, e.g.,
Avorn and Solomon 2000 ). On the other hand, this fi nding may support the claim
that marketing expenditures contribute to increased costs of pharmaceutical treat-
ment (Spurling et al. 2010 ) because large amounts of marketing effort are needed to
generate sizeable sales effects. We discuss the relationship between marketing and
drug prices in greater detail in Sects. 20.2.3 and 20.2.4 .
The question of which marketing instrument is most effective in stimulating prod-
uct category demand is highly relevant to policy makers, because this would identify
the main driver for category expansion. This could, in turn, guide the selection of