Review of Industrial Organization 19: 335–350, 2001.
© 2001 Kluwer Academic Publishers. Printed in the Netherlands.
Product Innovation, Process Innovation, and Size
MICHAEL FRITSCH and MONIKA MESCHEDE
Technical University Bergakademie Freiberg, Faculty of Economics and Business Administration,
Lessingstraße 45, D-09596 Freiberg, Germany
Abstract. We test the hypothesis that large ﬁrms devote a higher proportion of their research and
development (R&D) expenditure on process innovation than smaller ﬁrms. According to the estim-
ates, process- and product R&D expenditure rise less than in proportion to size. The size effect is
somewhat stronger for process R&D but the difference to product R&D is in no way dramatic. This
difference with regard to size elasticity of process- and product R&D is somewhat more pronounced
when accounting for possible interrelationships between expenditure on process- and product R&D
but remains statistically non-signiﬁcant.
Key words: Enterprise size, process innovation, product innovation, R&D expenditure.
JEL Classiﬁcations: D21, O31.
The relationship between innovation activity and ﬁrm size has been subject to
theoretical speculation and particularly to empirical investigation for a number of
decades (see Cohen, 1995, for an overview). This research was mainly motivated
by the implications that the ﬁndings have for the development of market structure
and the long-term stability of a market economy. Recently, Cohen and Klepper
(1996a, b) have developed a model for the relationship between R&D activities and
business size. One important element of their approach is the hypothesis that the
return of an innovation is positively related with the size of the respective business
unit and that this relationship is stronger for process innovation than for product
innovation. As a consequence, large ﬁrms may be expected to devote a higher
proportion of their R&D expenditure towards process innovation than smaller
In this paper, we test this hypothesis by investigating the relationship between
size and the amount of resources spent on product- and on process innovation using
data for German enterprises.
First, we outline the argument raised by Cohen and
We are indebted to anonymous referees for valuable comments and suggestions that helped us
to improve previous drafts.
The term ‘enterprise’ here refers to a productive unit at a speciﬁc location that may be part of
a ﬁrm with several units at different locations. This kind of unit of observation is sometimes also
referred to as ‘establishment’.