Knowledge Spillovers, Competition and Innovation Success

AuthorSpyros Arvanitis,Florian Seliger,Martin Woerter
Published date01 October 2020
DOIhttp://doi.org/10.1111/obes.12365
Date01 October 2020
1017
©2020 The Department of Economics, University of Oxford and JohnWiley & Sons Ltd
OXFORD BULLETIN OF ECONOMICSAND STATISTICS, 82, 5 (2020) 0305–9049
doi: 10.1111/obes.12365
Knowledge Spillovers, Competition and Innovation
Success*
Spyros Arvanitis, Florian Seliger and Martin Woerter
KOF Swiss Economic Institute, ETH Zurich, Leonhardstrasse 21, Zurich 8092, Switzerland.
(e-mail: sarvanit@retired.ethz.ch; e-mail: seliger@kof.ethz.ch; e-mail: woerter@kof.ethz.ch)
Abstract
We study the relationship between a patent-based measure of knowledge spillovers that
calculates technological proximity based on technologically relevant firms and innovation
success. We find – for a representative sample of Swiss firms – that knowledge spillovers
have a positive and significant association with the commercial success of innovative
products. The paper shows the importance of market conditions for the relationship of
spillovers with innovation performance: It is only positive and significant in markets with
a medium number of competitors in the main product market, but not in monopolistic or
polypolistic market structures.
I. Introduction
Since the two seminal papers by Jaffe (1986) and Jaffe, Trajtenberg and Henderson (1993)
patent-based measures of knowledge spillovers have become the workhorse in micro-level
studies. Although Bloom, Schankerman andVan Reenen (2013) substantially extended the
original Jaffe measure and made an effort to include spillovers from the product market,
the original approach to measure knowledge spillovers as suggested by Jaffe has sustained
its attractiveness. With the paper at hand, we study the relationship between a modified
version of the Jaffe spillover measure and a survey-based measure of innovation success.
Moreover weshow that market structure influences the relationship between spillovers and
innovation performance.
An important contribution of this study is the analysis of the spillover-innovation rela-
tionship in different competition regimes. Although the competition-innovation relation-
ship has been investigated extensively, the relationship with knowledgespillovers is incon-
clusive so far.Apar t from Bloom et al. (2013) who distinguished between spillovers from
technologies and from the product market, there have not been any systematic attempts
to study the relationship in different competitive environments.As the abundance of new
JEL Classification numbers: O31, O33, O34.
*Wethank two anonymous referees. Weare g rateful for comments receivedin the KOF Brownbag Seminar, at the
VPDE-BRICK Workshopin Economics of Innovation, Complexity, and Knowledge at the Collegio CarloAlber to in
Torino,the EARIE Conference in Lisboa, the EPIP Conference in Oxford, and the Globelics International Conference
in Athens.
1018 Bulletin
knowledge depends on the R&D intensity of competitors, a higher number of R&D intense
firms means in principle more knowledge spillovers. However, whether a firm’s innovation
activities can actually benefit from ‘more knowledge’in the market depends on its incen-
tives to innovate.We argue that the incentivesare lower in a market with many competitors
as leakage of own knowledge is more likely. Concretely, we expect that spillovers are more
likely to be relevantwith a ‘medium’ number of competitors where knowledge leakage due
to spillovers does not impede innovation incentives, but where still a sufficient number of
firms are in the market to generate a significant amount of spillovers.
To include all firms from all over the world that might be directly technologically
relevant and thus likely to generate spillovers for the focal firm, our spillover measure is
based on backward citation links. We use a sample of firms with patent activities from
the KOF Swiss Innovation Survey and look for citations to other firms’ patents in order to
identify the firms that are technologically relevant for the sample firms. For this exercise,
the technological landscape of Switzerland is an ideal subject because it is a small country
with a strongly internationalized economy where a significant portion of knowledge might
be generated abroad.
We test the relationships discussed above in the framework of an innovation equa-
tion and apply three different estimation procedures (Random Effects, Fixed Effects,
and a dynamic panel estimator (GMM)) to account for unobserved heterogeneity. The
spillover proxyshows a positive and highly significant association with innovation success.
Furthermore, the econometric analysis indeed shows that product market competition is an
important determinant of the relationship between spillovers and innovation performance:
The relationship between spillovers and innovation success is insignificant both in markets
with a relatively large number of competitors in the main product market and in markets
with a low number of competitors. In markets with a medium number of competitors
(6 to 10 according to our survey-based measure of competition), however, spillovers are
significantly positively associated with innovation success.
The paper is structured as follows: In Section II, we discuss the conceptual background
of the study. In Section III, we describe the empirical specification and corresponding
econometric issues. Section IV describes the data that are used and Section V presents the
results. Section VI summarizes and concludes.
II. Conceptual background
Knowledge spillovers: concept and measurement
Overview
A crucial aspect of innovativeactivity is the generation of knowledge, which to some extent
has the character of a public good. This gives rise to externalities (‘spillovers’) that are a
central theme in the literature on innovation in industrial economics (see, e.g. Griliches,
1979; Spence, 1984; Cohen and Levinthal, 1989; Geroski, 1995;Aghion and Jaravel, 2015).
A general though rather simplistic way to address this externality problem is to assume
the diffusion of new private knowledge leading to a ‘spillover pool of knowledge’ from
which other economic actors can drawinformation useful for their own innovativeactivities.
A general formulation for the spillovers as a (weighted) sum of the knowledge capital of a
©2020 The Department of Economics, University of Oxford and JohnWiley & Sons Ltd

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT