Do Spillovers Stimulate Incremental or Drastic Product Innovations? Evidence from German Establishment Data*
| Published date | 01 August 2011 |
| Author | Kornelius Kraft,Uwe Jirjahn |
| Date | 01 August 2011 |
| DOI | http://doi.org/10.1111/j.1468-0084.2010.00618.x |
509
©Blackwell Publishing Ltd and the Department of Economics, University of Oxford, 2011. Published by Blackwell Publishing Ltd,
9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA.
OXFORD BULLETIN OF ECONOMICS AND STATISTICS, 73, 4 (2011) 0305-9049
doi: 10.1111/j.1468-0084.2010.00618.x
Do Spillovers Stimulate Incremental or Drastic
Product Innovations? Evidence from German
Establishment Data*
Uwe Jirjahn† and Kornelius Kraft‡
†Universität Trier, Fachbereich IV, Lehrstuhl für Arbeitsmarktökonomik, Universitätsring 15,
54286 Trier, Germany (e-mail: jirjahn@uni-trier.de)
‡Technische Universität Dortmund, Lehrstuhl für Wirtschaftspolitik, Germany
(e-mail: kornelius.kraft@tu-dortmund.de)
Abstract
Despite the generally recognized importance of knowledge spillovers, the empirical litera-
ture is essentially silent on the type of innovation stimulated by spillovers. Weestimate the
determinants of product innovations differing in their degree of newness to the adopting
firm. Knowledge spillovers from rivals have a positive impact on incremental innovations.
This impact is largely independent of participation in R&D cooperation. Spillovers exert
no such independent influence on drastic innovation activities. The results support the
hypothesis that establishments face difficulties in using knowledge that comes from areas
they are not familiar with. Establishments exploit spillovers for incremental innovations
rather than for drastic innovations. To some degree R&D cooperation can help to over-
come the difficulties in using spillovers for drastic innovations. Furthermore, our estimates
provide evidence that the firm’s own R&D effort and the use of outside information are
substitutes.
I. Introduction
Knowledge spillovers play a key role in economics. First, endogenous growth models
assume that if the knowledge produced by individual firms spreads industry-wide, pro-
ductivity throughout the whole industry is enhanced.1Second, microeconomic analyses
emphasize that spillovers may influence the incentives to engage in innovative activities.
The firm’s own innovations in turn influence both its productivity and profitability.2Third,
the number of inter-firm strategic alliances increased dramatically during the 1990s. R&D
ÅThe authors thank the editor Jonathan Temple,the anonymous referees and seminar participants at the University
of Wisconsin-Milwaukee and the Leibniz University of Hannover for helpful comments.
JEL Classification numbers: L10, L20, L60, O31, O32.
1See, for example, Romer (1986) and Sena (2004).
2See Geroski, Machin and VanReenen (1993) and Hanel and St-Pierre (2002).
510 Bulletin
cooperation among firms is an important part of these alliances.3A substantial theoretical
literature examines the role spillovers are playing in R&D cooperation.4
Despite the generally recognized importance of knowledge spillovers, previous em-
pirical research is essentially silent on the type of innovation stimulated by spillovers. To
date, there has been almost no econometric study that examines whether or not a firm
can use the knowledge from other firms for all kinds of innovations. Using unusually rich
data on a sample of manufacturing establishments in Germany, we consider product inno-
vations differing in their degree of newness to the adopting firms. Incremental product
innovations are captured by follow-up products, qualitatively improved products and pro-
ducts with additional functions, while drastic innovation activities are captured by patent
applications and the launch of completely new products. Controlling for a variety of other
influences, we examine whether spillovers stimulate incremental or drastic innovations.
We also investigate if a firm’s own R&D effort or cooperation in R&D facilitates the use
of knowledge spillovers.
Multivariate probit estimates show that knowledge spillovers from rivals have a pos-
itive impact on incremental innovations. This impact is largely independent of participa-
tion in R&D cooperation. Spillovers exert no independent influence on drastic innovation
activities. The results support the hypothesis that establishments face difficulties in using
knowledge that comes from areas they are not familiar with. Establishments exploit
spillovers primarily for innovations that are closely related to their existing activities.
The results also show that R&D cooperation to some extent helps to overcome the diffi-
culties in using spillovers for drastic innovations. Yet, we do not find evidence for Cohen
and Levinthal’s (1989) influential hypothesis that the firm’s own R&D effort facilitates
the use of spillovers.5Quite the contrary, our estimates show that own R&D effort and
the use of outside knowledge are substitutes. This supports theoretical contributions by
Jovanovic and MacDonald (1994) and Eeckhout and Jovanovic (2002). Establishments
with high R&D effort are closer to the frontier of technology and product development.
Since they have less to learn from others, outside knowledge is less useful to them.
The rest of the article is organized as follows. Section II provides a theoretical back-
ground discussion. Section III describes the data and variables. Section IV presents the
results of the empirical analysis. Section V presents the robustness checks. Section VI
concludes.
II. Hypotheses
Learning and reorganization of production
Learning and the capabilities to learn are at the heart of innovative activities. The strategic
management literature stresses that learning is predominantly cumulative and closely re-
lated to the firm’s previous activities. New knowledge builds upon experience accumulated
3See Caloghirou, Ioannides and Vonortas(2003).
4See de Bondt (1996) for a survey. While early contributions treated spillovers as exogenous determinants
(d’Aspremont and Jacquemin, 1988), recent analyses stress that the use of spillovers depends on whether or not
firms form cooperative agreements (Katsoulacos and Ulph, 1998; Kamien and Zang, 2000).
5This fits with recent studies by Kneller (2005) and Kneller and Stevens (2006).
©Blackwell Publishing Ltd and the Department of Economics, University of Oxford 2011
To continue reading
Request your trial