employment (expenditure) relative to either total employment (turnover). Measuring
the intensity of knowledge sourcing through firms' supply-chain and non supply-chain
innovation linkages is more experimental, and here we follow LR (2001) who develop
intensity scores for the extent of firms' external contacts. More specifically, Love and
Roper (2001) construct an intensity score which is proportionate to the number of
innovation activities where the firm is involved in outsourcing3. Here we adopt a
similar approach constructing intensity scores for each firms' knowledge sourcing
through supply-chain and non supply-chain collaboration based on the number of
types of organisation with which the firm is undertaking collaborative innovation
activity. For example, we identify six types of potential supply-chain partners
(customers, suppliers, competitors, other group companies, joint ventures and ???) for
companies undertaking innovation collaboration with three of these types of partner
'score' 50 per cent, firms collaborating with all six types of partner score 100 per cent.
If firms' expectations about post-innovation returns are rational, i.e. involve no
systematic errors, and we regard
πi = β0+ β1MPOSi + β2RBASEi + β3ITECHk +ηi
(2)
We can substitute for expected post-innovation returns in equation (1) to obtain
reduced form knowledge sourcing equations:
RKSi =θ10 +θ12MPOSi +θ13RBASEi +γ14RIS +θ15ITECHk+λ1
i 10 12 i 13 i 14 j 15 k 1
SCKSi =θ20 +θ22MPOSi +θ23RBASEi +γ24RIS +θ25ITECHk+λ2 (3)
i 20 22 i 23 i 24 j 25 k 2
NSCKSi =θ30+θ32MPOSi+θ33RBASEi+γ34RIS +θ35ITECHk +λ3
i 30 32 i 33 i 34 j 35 k 3
where: θ12=γ12+γ11β1 and λ1=ε1+γ11η etc.
Knowledge sourced through R&D, supply-chain or non-supply chain collaboration
will then be combined into a form which can be commercially exploited, i.e.
innovation. Locational and industry-specific factors may also be important - along
with the resource base of the firm - in determining the efficiency with which acquired
knowledge is translated into commercially exploitable outputs or innovations
(INNOVi). The potential for such effects suggests a general form of innovation
production function (Geroski, 1990; Harris and Trainor, 1995):
INNOVi = φ0 + φ RDi + φ2 XNETi + φ3 INETi + χ1MPOS
+ χ2 RBASE + χ3 RIS + χ4 ITECH + μi
(4)
where we allow for the possibility of regional (RIS) (Audretsch and Feldman, 1996)
and industrial efficiency (ITECH) effects on as well as plant specific variables
(MPOS, RBASE).
Different measures of the outputs from the innovation process are possible reflecting
the potential technological, commercial and organisational outcomes. The percentage
of sales derived from innovative products (used by LH, CDM and LR), for example,
3. LR (2001) consider firms' networking during seven activities which form part of
the product development process: the identification of new or improved products;
prototype development; final product development; product testing; production
engineering; market research and marketing strategy.