European Journal of Purchasing & Supply Management 8 (2002) 135–149

Innovation as a determinant of suppliers’ roles and performances: an

empirical study in the food machinery industry
Alberto Petroni*, Barbara Panciroli
Dipartimento di Ingegneria Industriale, Facolta" di Ingegneria, Viale delle Scienze 181/A, 43100 Parma, Italy
Received 14 May 1999; received in revised form 31 May 2001; accepted 6 December 2001

Abstract

The basic premise of this study is that suppliers, in order to compete successfully in the marketplace, have to develop certain
innovative capabilities that, in turn, depend on the type of relationship that is established with customers (in other words, the
strategic role assumed by the supplier in the supply chain). Based on a study of 198 suppliers operating in the food packaging
machinery industry, structural equation modelling has revealed that customers actually assign suppliers different roles and give them
varying levels of responsibility in the product development and manufacturing stages. These roles are correlated to the supplier’s
distinctive innovation capabilities. Clustering analysis is used to divide the suppliers into homogeneous groups. Innovative
capabilities are not only associated with traditional competencies in R&D and product/process innovation but also in supportive
capabilities in the form of absorptive capacity, technological scanning, innovation-oriented culture, skills and know-how of
individuals and managerial practices. The impact of different innovative capabilities (both technological and managerial) on critical
performance factors such as price, quality, time and flexibility is measured for each type of supplier in turn. The research findings
lend support to the hypothesis that the level of investment in technology and the acquisition of specific managerial capabilities are,
to a large extent, a determinant of supplier–customer interdependence. r 2002 Elsevier Science Ltd. All rights reserved.

Keywords: Supplier–customer relationships; New product development; Technological competencies; Food machinery industry

1. Introduction siveness, policy co-ordination and symbiotic relation-

Increasingly, the need to establish long-term purchas-
ing relationships has been recognised, and consequently,
more manufacturing companies have turned to their
suppliers (Lindberg and Trygg, 1991; Imrie and Morris,
1992; Lamming, 1993; Nishiguchi, 1994; Helper and
Sako, 1995).
Relationships between customers and suppliers are
usually viewed as being positioned along a continuum
with two extreme ‘‘adversarial’’ and ‘‘collaborative’’
poles (Macbeth and Ferguson, 1994; Miller and Kelle,
1998). Relationships are considered to be evolving
towards more collaborative forms, such as the so-called
Japanese-style keiretsu partnership (JSP) (Dyer and
Ouchi, 1993; Martin et al., 1995; Nishiguchi and
Brookfield, 1997). Imai (1990) defines a keiretsu as a
‘‘web of relationships ranging from tight to loose among
companies working together’’. Internal control, cohe-
*Corresponding author. Tel.:+39-0521-905861; fax: +39-0521-
905705.
E-mail address: petroni@ied.unipr.it (A. Petroni).
ships combine to form a keiretsu which links firms into
groups.
The achievement of high-level performances in terms
of cost, quality, and time-to-market appears ever more
dependent on the quality and effectiveness of the supply
network (De Toni and Nassimbeni, 1995; Dyer, 1996c).
The relationship between manufacturers and their parts
suppliers has been recognised as a key element in the
improvement process (Butcher, 1991).
Suppliers may be required to take charge of develop-
ing new products in cooperation with the supplier’s
design staff (starting from the simple functional
requirements) and innovating their manufacturing
process (Krause, 1999). In fact, numerous studies have
confirmed that supplier–customer collaboration in new
product development (NPD) helps to improve quality,
and reduces costs and lead times (Clark, 1998; Clark and
Fujimoto, 1991). Project strategy, which implies deci-
sions being taken both on project complexity and scope
(including decisions concerning the level of suppliers
involvement), has a relevant effect on the overall project
performance (Clark and Fujimoto, 1991). As far as the

0969-7012/02/$ - see front matter r 2002 Elsevier Science Ltd. All rights reserved.
PII: S 0 9 6 9 - 7 0 1 2 ( 0 2 ) 0 0 0 0 4 - 7
136 A. Petroni, B. Panciroli / European Journal of Purchasing & Supply Management 8 (2002) 135–149

extent and type of collaboration in NPD processes is different extent to the development of the final product.
concerned, the dominant research area in this field Supplier involvement may, thus, increase the complexity
draws on the pioneering studies by Womack et al. of managing development projects.
(1990), Clark and Fujimoto (1991), Clark and Wheel- One of the key responses to reduce this complexity is
wright (1992) and Lamming (1993). The following to determine which type of involvement a manufacturer
elements are identified as being critical for achieving should have with the various suppliers who might be
effectiveness and efficiency in product development: engaged simultaneously in a project (Kamath and Liker,
1994). The segmentation of suppliers into different tiers
* manufacturing capabilities enabling the firm to
of sourcing and role profiles (Dyer, 1996a, b, 1998)
achieve effective links between R&D and manufac-
fosters the reconfiguration and integrated management
turing;
of the supply chain (Hartley et al., 1997) which, in turn,
* engineers capabilities to carry out a wide spectrum of
may help to prioritise activities thus making the
activities and tasks;
involvement of suppliers more manageable.
* superior capabilities of concurrent engineering;
As a consequence, there might be a whole range of
* capacity of product managers to act as coordinators
postures that customers and suppliers can adopt within
of multi-functional teams and project supervisors
a relationship which can broadly be assumed to vary
(so-called ‘‘heavyweight program manager’’);
from one extreme, a durable-arm’s-length relationship,
* mutual involvement of suppliers and customers in the
to the other extreme, a strategic partnership. All these
product concept stage.
relative postures are characterised by different degrees
There is general agreement among scholars and of supplier collaboration in product development stages,
practitioners that the innovative capacity of suppliers R&D integration, inter-firm knowledge, risk sharing
is a critical factor of their ability to respond to the and mutual trust.
increasing demands and challenges set by customers. The central premise of this paper is that the technolo-
The shift in the intrinsic nature of supplier–customer gical capability profile of suppliers (both in terms of
relationships and the establishment of more complex investment in R&D/acquisition of technology and
forms of collaboration has been accompanied by a
parallel evolution of the suppliers’ technological base. It
is, in fact, the integration of supplier technology that is STEP 1
the basis of successful long-term relationships (Kamath
and Liker, 1994). For advanced manufacturing technol-
ogies (AMT) to be fully exploited, strong, long-term Type of
links must exist between manufacturers and their Innovative supplier-customer
capabilities interdependence
component suppliers. These links are themselves
affected by new technologies such as CAD, FMS and
CIM to the extent that new developments in the
technical field provide opportunities for cooperation in
the operational areas (Lamming, 1986). As a prerequi- STEP 2
site for the greater involvement of suppliers along the
whole design/manufacturing process, customers may
often be induced to stimulate changes in the competence SUPPLIER TYPE 1
base of their suppliers (Carren and Pearson, 1999). In
their efforts to change, suppliers may be supported Innovative Performance
through consultancy and training services in order to capabilities
develop the required capabilities. Or alternatively, the
choice to invest in technology and knowledge may be SUPPLIER TYPE 2
the result of a deliberate and conscious strategic
decision. In either case, firms that have invested in the Innovative Performance
capabilities
relationship with their customers have capitalised upon
the acquisition of both technical and managerial skills
SUPPLIER TYPE 3
for bringing about product development.
Supplier involvement in product development may
Innovative Performance
range from giving minor design specifications (e.g. to capabilities
improve the manufacturability of a component) to being
responsible for the complete development, design and
engineering of a specific part or sub-assembly. Suppliers
of different parts and components contribute to a Fig. 1. The research framework.
 A. Petroni, B. Panciroli / European Journal of Purchasing & Supply Management 8 (2002) 135–149
complementary assets) is, to a large extent, a determinant
of the supplier–customer interdependence in NPD.
More specifically, the purpose of this study is to
investigate the research questions: do firms with a high
level of innovative capabilities tend to establish strategic
partnerships with customers to a greater extent than
firms with lower levels of innovative capabilities do?
How do the specific innovative capabilities that are
developed by suppliers, having different levels of
integration with customers, impact on performances?
To answer these questions, the following research
steps (Fig. 1) were performed:
(a) testing causal relationships between innovative
capabilities and supplier–customer interdepen-
dence;
(b) identifying different ‘‘role profiles’’ for suppliers on
the basis of the specific level of supplier–customer
interdependence;
(c) measuring the effects on performances of the
innovative capabilities that distinguishes each ‘‘role
profile’’.
2. Constructs and measures
2.1. Customer–supplier relationship in new product
development
The stream of research on customer–supplier inter-
dependence in NPD has considered different issues from
different perspectives. Of specific relevance to this paper
are the studies that have addressed a classification of
supplier types and role profiles. At a general level of
analysis, supplier classifications have been carried out by
focussing on:
(a) the distinctive features of the products supplied
(Clark and Fujimoto, 1991; Lamming, 1993;
Hsuan, 1999);
(b) the specificity of tasks in NPD processes (Von
Hippel, 1990; Holmen and Kristensen, 1998);
(c) the extent and type of collaboration in the processes
of NPD (Kamath and Liker, 1994; Roy and Potter,
1996; Calderini and Cantamessa, 1997; Bensaou,
1999; Wynstra and ten Pierick, 2000).
Clark and Fujimoto (1991) classified the suppliers
based on the product type: supply proprietary part,
black box part and detailed controlled part. Supply
proprietary parts are standard products totally devel-
oped by the supplier and sold through catalogues. Black
box parts are the typical outcome of the joint effort
between the customer and the original equipment
manufacturer (OEM). The OEM defines the desired
standards of cost, performance, interface details and
other basic features of the product. The supplier is in
137
turn responsible for both functional specifications and
detailed engineering. A similar concept is the ‘‘grey box’’
(Lamming, 1993). As for the detailed controlled parts,
much of the product design is performed by
the OEM, who is responsible for both functional speci-
fications and detailed engineering. The supplier is
responsible for process engineering and manufacturing.
Hsuan (1999) examines the effects of supplier–buyer
relationships (assumed to vary from ‘‘durable-arm’s
length relationship to strategic partnership) on product
modularisation by evaluating the opportunities of
modularisation and corresponding interface constraints
at the level of component, module, sub-system and
system. This paper posits that the degree of supplier–
buyer interdependency might heavily influence the
outcome of modularisation.
In the classification carried out by Kamath and Liker
(1994), the OEM assigns suppliers different roles
and gives them varying levels of responsibility for
product development. The OEM can be actively
involved in the design stage. Its influence is exercised
through an extensive effort in supplier development,
both in the form of bilateral deployment of resources,
(engineers, capital, technology and facilities) and
training/information sharing (Krause, 1999). Only an
elite corps of first-tier suppliers have the appropriate
human and physical resources, the technologies and
capabilities to enjoy a full-blown partnership with their
customers. These suppliers are called ‘‘full-service
providers’’. A second type of supplier (‘‘full-system
supplier’’) takes major responsibility with close custo-
mer guidance. The so-called ‘‘child’’ supplier acts
primarily in response to customer design requirements
(behaving above all as an executor). Finally, the
‘‘contractual’’ supplier is used as an extension of
customer manufacturing capability: the customer
gives detailed blueprints or orders from a catalogue
which the supplier then executes. Implicit in the
classification by Kamath and Liker (1994) is the idea
that suppliers might follow a growth path (from a
contractual relationship to a full partnership) through
sound investments in R&D facilities and manufacturing
systems. A similar analytic approach is followed by Roy
and Potter (1996): the spectrum of the supplier–buyer
relationships as a function of the NPD activities is split
into five classes (‘‘in-house’’, ‘‘tender-based’’, ‘‘colla-
boration’’, ‘‘interaction’’, ‘‘dedication’’). A different
classification is proposed by Calderini and Cantamessa
(1997), whose taxonomy is based on the temporal
relationship between the design and the transactional
stage. Three categories of suppliers are thus identified:
firms supplying standard products through catalogues
(when products are designed before commercialisation),
firms supplying highly customised products on the basis
of a tough tender-based competition (when production
and commercialisation are simultaneous) and suppliers
138 A. Petroni, B. Panciroli / European Journal of Purchasing & Supply Management 8 (2002) 135–149
who are vertically integrated within the value chain
of their customers (when commercialisation precedes
design).
From a more theoretical perspective, Von Hippel
(1990) adds a further argument for the discussion. The
efficiency achieved in the product development stage by
splitting the tasks along the supply chain depends on the
degree of specificity of the said tasks. Following these
suggestions, the study by Holmen and Kristensen (1998)
posits that suppliers can be divided into those who could
contribute to product development through task parti-
tioning, and those with whom an interactive develop-
ment process would have to be initiated. More
specifically, their research illustrates how quality func-
tion deployment can be used effectively to identify the
correlations between the characteristics of a planned
product as a practical tool for discriminating between:
(i) the suppliers with whom the product concept calls for
development interaction between two or more parties
(through the formation of a joint development team);
(ii) the suppliers who can contribute to the development
on the basis of task partitioning (through a modular
design process, avoiding the costly process of develop-
ment interaction) and (iii) the standard suppliers of
proprietary parts.
Although without elaborating specific supplier pro-
files, Bensaou’s portfolio model (Bensaou, 1999)
provides a description of supplier roles based on the
distinction between different buying situations. In the
market exchange situation, suppliers manufacture to
buyer specifications with little interaction, while a
captive buyer situation calls for ‘‘broadband’’ commu-
nication in design. Finally, the strategic partnership case
is based on standardised rules and procedures such as
electronic data interchange and schemes for exchange of
guest engineers.
The supplier involvement portfolio by Wynstra and
Pierick (2000) distinguishes four types of suppliers on
the basis of two variables: the degree of responsibility
for product development that is contracted out to the
supplier and the development risk. In a strategic
development situation, the supplier is asked to prepare
a ‘‘global design’’ of the building block using functional
specifications as a starting point. The OEM is con-
fronted with a high development risk and the supplier
perceives a high level of uncertainty. In the case of
critical development, supplier involvement is charac-
terised by a considerable development risk and a low
degree of development responsibility. In the case of
arm’s-length development, a large part of the develop-
ment is contracted out to the supplier, but the
development risk is considerably lower than in the case
of strategic development. Finally, routine development
is characterised by low development risk and little
responsibility for the development being held by the
supplier.
Using the themes outlined above as a basis, six aspects
(constructs) of supplier–customer relationships in NPD
were considered in this study for classification purposes.
Even though the constructs considered are somehow
‘‘transversal’’ to the classification criteria that have been
found in the literature, they are used in an attempt to
capture the rich variety and complexity of relevant
aspects of supplier–customer relationships in product
development.
These constructs are:
(a) customer’s influence in product performance speci-
fications;
(b) mutual involvement in product development;
(c) customer’s influence in manufacturing specifica-
tions;
(d) R&D integration;
(e) stability of relationship;
(f) supplier development.
Table 1 reports these constructs and the relative
measures that were considered in the study.
Customer’s influence in product performance specifica-
tions: This construct was measured by the influence of
customer in the definition of product specifications in
terms of (i) cost, (ii) time and (iii) quality.
Mutual involvement in product development: As
discussed above, the establishment of novel models of
supply might be accompanied by an increasing degree
of supplier responsibility in product development. Two
measures were then considered: mutual involvement in
product concept generation and mutual involvement
in product design and testing.
Customer’s influence in manufacturing specifications:
This construct was measured by (i) customer influence in
defining supplier capacity planning and (ii) process
engineering requirements.
R&D Integration: This construct was operationalised
by considering three measures: the extent to which (i)
engineers, (ii) technical physical assets and (iii) critical
development information are shared between the
supplier and buyer.
Stability of relationship: Four measures were consid-
ered: the length of relationship (years), the customer’s
propensity to involve the supplier in multiple product
development projects, the use of non-tender-based
(Dyer, 1996a-c) and non-price based selection proce-
dures (Roy and Potter, 1996).
Supplier development: Two measures were used: the
level of customer commitment to supplier training and
the use of flexible contracts (Krause, 1999).
2.2. Supplier innovation capabilities
Innovation crucial determinants reside in the interac-
tion between technological and organisational
 A. Petroni, B. Panciroli / European Journal of Purchasing & Supply Management 8 (2002) 135–149 139

Table 1 and renewing process (Teece and Pisano, 1994; Saviotti,

Constructs and measures used in the questionnaire. All items were 1994).
measured using 7-point likert-type scales
In order to be functional to this purpose, any kind of
Constructs measures knowledge must be embodied in specific ‘‘action
Customer–supplier relationship in new product development
producing’’ forms, that is, using Leonard-Barton’s
Customer’s influence in product performance specifications (1992) categories, individual skills (IS), managerial
price specifications resources (routines, procedures, incentives) and social
time specifications values and norms. Innovation capabilities (IC) are, thus,
quality specifications not only associated with traditionally measured innova-
Mutual involvement in product development
mutual involvement in product concept generation
tion efforts in R&D and product/process innovation but
mutual involvement in product design and testing also in supportive organisational capabilities in the form
Customer’s influence in manufacturing specifications of innovation-oriented culture, skills and know-how of
supplier capacity planning individuals and managerial practices. Hence, supplier
definition of process engineering requirements. innovation capacity should not be limited to efforts
R&D Integration
co-location of engineers
traditionally evaluated in terms of R&D investments
sharing of equipment and technology acquisition. These investments are
sharing of critical information essential but without complementary assets such as
Stability of relationship a supportive absorptive capacity (AC) (Cohen and
length of relationship Levinthal, 1990; Cohen and Levinthal, 1994), appro-
supplier’s involvement in multiple projects
use of non-tender-based selection procedures
priate organisational and managerial capabilities
use of non-price-based selection procedures (Lefebvre et al., 1993) and creative skills, they might
Supplier development not be sufficient to reach a successful result. This is
customer’s commitment to supplier training in line with the wealth of indications that comes from
use of flexible contracts the literature on learning organisations. A learning
Innovative capabilities
company is an organisation dedicated to knowledge
Absorptive capacity creation, collection, and control. Contribution to
R&D expense to annual sales knowledge is a key criterion for all activities, albeit
Personnel training expense to annual sales not the only one. In a learning laboratory or factory,
Technological penetration tremendous amounts of knowledge and skills are
number of advanced manufacturing technologies
relative importance of advanced manufacturing technologies
embedded in physical equipment and processes and
Technological scanning embodied in people. More important, however, are the
technological awareness non-technical aspects, the managerial practices and
commercial availability of information on technologies underlying values that constantly renew and support
perception of technology’s commercial impact the knowledge bases. In other words, learning requires
Individual skills
problem solving capabilities
creation and control of both external and internal
attitude to knowledge exchange knowledge for both current and future operations.
exclusive know-how related to products Therefore, each distinguishing learning activity (such
Innovation-supporting managerial competence as, for instance, problem solving in current operations,
project management competence internal knowledge integration across functions and
human resources management competence
quality management capabilities
projects, innovation and experimentation and integra-
Innovation culture tion of external information flows) is the operational
sharing of research-oriented values expression of an underlying value and is strongly
diffusion of result-oriented tacit norms supported by a compatible managerial system of
Performance measures procedures and incentives (Leonard-Barton, 1991,
Price
Quality
1992). Thus each operational activity, social value, and
Flexibility managerial system functions as an internally consistent
Time subsystem. These types of assets cannot be dissociated
from the innovation capacity of a firm and as such
represent a form of innovation effort. In other words,
processes. The former influence the creation, shape and successful firms must seek to provide a supportive
renewal of a firm’s skills and capabilities, thus repre- organisational environment conducive to the full rea-
senting the basic foundation for the latter, which in lisation of the benefits that the new technologies can
turn, can be a powerful driver for re-shaping a provide.
firm’s competence base. Knowledge is then the basic Bearing these considerations in mind, the supplier IC
foundation of capabilities and problem-solving repre- profile consists of six aspects (constructs): absorptive
sents the driver of the capabilities building, nurturing capacity, technological penetration (TP), technological
140 A. Petroni, B. Panciroli / European Journal of Purchasing & Supply Management 8 (2002) 135–149

scanning (TS), individuals skills, innovation-supporting formula:

managerial competence (MC) and innovation-oriented
norms and values. X
24
DPI ¼ i j rj ;
Absorptive capacity: This construct refers to the firm’s i¼1
ability to evaluate, assimilate, and exploit extramural
technological developments (Cohen and Levinthal, where DPI is the degree of process innovativeness, ij ¼ 0
1990, 1994). It helps explain the extent of technological or 1, depending on whether the specific jth technology
capability transfer, at least in some forms of partnership has been implemented or not, rj the degree of innova-
(Mowery et al., 1996). Expenditure on R&D as a tiveness of technology j as evaluated by three academic
fraction of annual sales is a common measure used for experts who ranked each innovation on a seven point
estimating a firm’s absorptive capacity. Another com- Likert-type scale.
mon measure is the investment in personnel training Technological scanning (TS):This construct captures
(Liu and White, 1997). Both these measures were thus the extent to which firms are sensitive to their
considered in the questionnaire. technological environment. It is intended to assess to
Technological penetration (TP): In order to capture what extent firms are concerned with the search for new
the full variety of a firm’s innovation efforts, recent ways of making better products at a better price. It
research has proposed going beyond strict R&D specifically refers to the ability of a firm to identify
evaluation to consider other aspects such as the opportunities, to understand and foresee the technolo-
penetration of technologies within the firm. The adop- gical strategies of competitors as well as to evaluate
tion of AMT may reveal a firm’s expertise as well as its emerging technologies. Three measures were selected to
commitment and ability to better meet client require- operationalise this construct: supplier awareness of
ments. The number of AMT and methodologies was technological development, availability of information
therefore used as a factual measure of a firm’s IC. A on production technologies (i.e. using patent databases,
complete list of the AMTs that were considered is technology transfer programmes, internet or other
reported in Table 2. sources) and awareness of the competitive advantage
Not only the absolute number of technologies was which may be derived from the use of new technologies.
considered but also their relative importance. The Individual skills (IS): In a context where specialisation
degree of process innovativeness (the second measure gradually becomes the only way of getting business from
for TP) for each firm was computed via the following customers, suppliers cannot avoid upgrading the skills
of their employees. Not only should those skilled
employees make it easier to adopt innovation but they
should also stimulate and facilitate the creation of
Table 2 specific know-how and/or new products. Three further
List of manufacturing/logistic and design-related technologies and
measures were therefore considered: problem solving
methodologies considered
capabilities, attitude of personnel to exchange of
Manufacturing/logistic and Manufacturing/logistic and knowledge and the level of exclusive know-how related
design technologies design methodologies
to products.
Automated handling PERT Innovation-supporting managerial competence: This
Bar code system CPM construct concerns a firm’s capability to acquire and
Computer-assisted design Product development process manage its technological resources in order to meet
(CAD) modelling
Computerized numerical Value analysis and engineering
customer expectations. Managerial capability involves
control (CNC) tools the ability to coordinate several activities both inside
Direct numerical control Taguchi (design of experiment) and outside the firms’ boundaries. Concern for manage-
(DNC) rial quality in the NPD process, in project management
Electronic data interchange Quality function deployment
practices or in the implementation of novel technologies
(EDI) (QFD)
Integrated CAD/CAM Design for assembly/ are domains where the appropriate managerial capabil-
manufacturing ities can be exercised (Wynstra et al., 1999). Three
Manufacturing resource Reliability analysis (FTA, further measures were thus introduced that were
planning (MRPII) Fmea) finalised to assess the implementation and development
Net needs planning (MRPI) Computerized quality of managerial practices and systems in the areas of: (i)
inspection and control costing
Rapid prototyping technique Employee accounting project management, (ii) human resource management
(RPT) and (iii) quality management.
Inventory management Innovation culture (INC): Two measures were con-
Job order costing sidered: the diffusion of innovation-oriented values and
Just-in-time systems of tacit results-oriented norms and behaviours within
Statistical process control
the organisation. To establish behaviours and values
 A. Petroni, B. Panciroli / European Journal of Purchasing & Supply Management 8 (2002) 135–149
able to foster and nurture a working climate oriented
to innovation and results requires the building of an
organisation where winning as a team takes precedence
over individual ownership of ideas, and knowledge is
liberally shared. Also, a positive attitude towards risk
has to permeate the company. If everyone experiments,
learns, and innovates, then neither success nor failure
can be heavily personalised. If individuals are singled
out for praise, they have an incentive to protect ideas as
intellectual property rather than seek embellishment
from friendly critics and co-developers (Leonard-
Barton, 1992).
2.3. Performance
Four different measures of performance were con-
sidered. In the context of selecting suppliers, customers
usually retain firms that produce the highest aggregate
score on:
* price;
* quality;
* flexibility of production and
* delivery times.
In order to approximate this aggregate score,
respondents were asked to evaluate their company’s
main assets in comparison to those of their closest
competitors on these four measures. Further, cross-
validation was also made on specific cases by matching
the scores assigned by the respondents with those
provided by the customers. In order to carry out such
a validation, six purchasing managers of six large firms
producing complete bottling and packaging lines were,
after the questionnaire collection stage was completed,
identified as important customers for a number of
respondents. They were interviewed and asked to rate
their suppliers along the same performance measures
considered in the questionnaire. Data matching proved
to be quite satisfactory, showing that respondents
do generally know their own strengths and weaknesses
well.
This evaluation method is fairly common and the four
criteria are amongst the most widely reported in the
literature (Chao et al., 1993; Wilson, 1994; Mummala-
neni et al., 1996; Verma and Pullman, 1998), and its
appropriateness was confirmed during extensive pre-
liminary discussions with prime contractors.
3. Sample and data collection
The results presented in this paper come from an in-
depth questionnaire survey carried out among suppliers
operating in the food machinery industry in northern
Italy. Prior to the survey, in order to provide more
complete information for the hypothesis and render the
141
preliminary version of the questionnaire valid, ‘‘round-
table’’ meetings were organised with eight CEOs from
the sample. This stage proved to be particularly useful
since it provided a direct insight into how specific
phenomena, which came to the fore during the literature
survey, found actual application in the industrial
practice within the context under investigation. This
activity pursued two distinct objectives: building a
framework for a structured research hypothesis con-
cerning the relationships between the phenomena
investigated and, secondly, to decide upon which data
should be collected in a subsequent questionnaire
survey. The results of this process were edited for
overlaps and repetitions to produce a questionnaire with
a final list of more than 40 items. The structure of the
questionnaire consisted of three parts: the first referred
to the collection of general information about the
company, its competitive environment, the production
process and the type of products supplied; the second
was aimed at investigating the characteristics of the
relationship with customers; and the third concerned the
assessment of the respondent’s innovative profile.
An overall population of 542 firms was obtained from
two complementary data sources: the CIBUS database
(an annual exhibition of food and beverage machinery
manufacturers) and the chambers of commerce of five
different provinces. All 542 firms were sent the
questionnaire, and were contacted (by mail or phone)
to check for their willingness to participate in the survey.
Among them, 206 suppliers confirmed their interest in
participating in the research, but 80 of them did not
return the questionnaire (or returned it incomplete). A
final number of 198 usable questionnaires was obtained,
with a satisfying response rate of 36% (198 out of 542).
The questionnaire was submitted to managers who
were identified as being responsible for the strategic and
operational management of the manufacturing/logistic
processes. Typical respondents included: chief technical
officers, head of operations, logistic managers, distribu-
tion managers, and quality managers. In the adminis-
tration of the questionnaire, respondents were always
assisted by at least one interviewer who, given the
prevailing perceptual nature and the purpose of the
research, took the maximum possible care to avoid
exerting any influence on the respondent. The presence
of a research assistant (who received thorough training
prior to the interview) helped avoid misinterpretation
of questions.
4. Results
4.1. LISREL and structural equation modelling
LISREL is a vigorous method for testing causal
models with both observable and latent variables, as it
142 A. Petroni, B. Panciroli / European Journal of Purchasing & Supply Management 8 (2002) 135–149

is able to simultaneously evaluate the measurement and
causal components of complex models. LISREL conse-
quently is becoming preferred to correlation, regression,
or path analysis by researchers for testing causal models
(Dillion and Goldstein, 1984). In structural equation
modelling, it is preferable to have several indicators of
a construct as opposed to a single indicator (Hair et al.,
1995).
The relationships to be tested with structural equation
modelling are shown in Fig. 2.
The composite scores for the AC, TP, TS, IS,
innovation-supporting MC and INC constructs are
shown as the observable indicators of the endogenous
latent variable, IC. The composite scores for customer
influence in product performance specifications (CIP),
mutual involvement in product development (MIP),
customer influence in manufacturing specifications
(CIM), R&D integration (R&D), stability of relation-
ship (SR), and supplier development (SD) are shown as
the observable indicators of the endogenous latent
variable, customer–supplier relationship (C–SR). The
composite measures were calculated by summing the
individual scores for each item in a construct and then
dividing by the number of items. For example, the
responses to AC1 and AC2, were summed and then
divided by two to determine the composite measure AC.
The responses to the four performance items were used
as the observable indicators of the endogenous latent
variable, PER.
The goodness-of-fit index (GFI) is used to evaluate
the appropriateness of the models tested. It is relatively
robust against departures from normality and appraises
all of the model’s parameters, including measurement
items, directional relationships, and error terms, at the
same time. GFI provides a measure ranging from 0 to 1.
GFI will be close to 1 if a ‘‘good’’ model to data fit is
detected (Dillion and Goldstein, 1984).
All of the 198 responses were submitted to LISREL to
evaluate the model in Fig. 2. The GFI of 0.921 indicates
a good model to data fit. The goodness-of-fit index
adjusted for degrees of freedom (AGFI) was 0.919,
which is also good. The computed t-values, which
evaluate the statistical significance of the indicators
measurement portion of LISREL, ranged from 6.931
to 14.346. These are well above the minimum acceptable
t-value of 2.00 (at a ¼ 0:05). The top portion Table 3
displays a summary of the data related to testing the
hypothesised relationships shown in Fig. 2.

PR FLEX

AC
PER
TP
QUAL T
TS

IS
IC
MC CIP MIP CIM
INC

C-SR

SR SD
R&D

Results: LEGEND:
AC-Absorptive capacity
GFI = 0.921
TP-Technological penetration
AGFI = 0.919
TS-Technological scanning
n = 198
IS-Individual skills
MC-Managerial competence
INC-Innovation culture
CIP- Customer's influence in PR-Price
product performance specifications
MIP- Mutual involvement in FLEX-Flexibility
product development
CIM-Customer's influence in QUAL-Quality
manufacturing specifications
R&D-R&D Integration T-Time
SR-Stability of relationship
SD-Supplier development

Fig. 2. The impact of IC on supplier–customer relationship and performance.

 A. Petroni, B. Panciroli / European Journal of Purchasing & Supply Management 8 (2002) 135–149 143

Table 3 by IC. To examine these relationships at the factor level,

Summary of LISREL generated results the model in Fig. 3 was submitted to LISREL.
Relationship t-value Significant LISREL coefficient The GFI of 0.915 indicates a clearly acceptable model
to data fit. The AGFI is 0.895, which is also good. The
IC-C-SR 4.394 Yes 0.342
IC-PER 6.256 Yes 0.502
computed t-values for the indicators (measurement
IC-CIP 4.176 Yes 0.332 portion of LISREL) ranged from 7.765 to 16.985. The
IC-MIP 3.765 Yes 0.364 bottom portion of Table 3 displays a summary of
IC-CIM 0.659 No 0.092 the data generated by LISREL related to the testing
IC-R&D 5.198 Yes 0.417 of the hypothesized relationships between the con-
IC-SR 5.002 Yes 0.392
IC-SD 1.876 No 0.102
structs. It is worth noting that no causal relationship
was found between IC/Customer’s influence in manu-
facturing specifications and IC/Supplier Development.

4.2. Identification of suppliers’ roles

CIP
AC
MIP
TP
TS
IS
IC
CIM
MC
INC
SR
Results:
GFI = 0.915
SD R&D
AGFI = 0.895
n = 198
Fig. 3. Examining relationships between IC and C-SR.
The computed t-values judge the statistical signifi-
cance of each theorised relationship, and they are well
above the minimum acceptable value of 2.00. LISREL
coefficients give an indication of the relative strength of
each relationship (at a ¼ 0:05).
A significant positive relationship is shown between
IC and S–CR. We may then posit that there is a causal
relationship between the innovative profile of a firm and
the role that is assigned to it by customers. A positive
and significant relationship is also demonstrated be-
tween IC and PER. Firms with high levels of IC are high
performing as measured by customer satisfaction and
market performance. By combining these results, it may
be appropriate to claim that IC positively affect
organisational performance through their impact on
supplier–customer relationships.
While it is certainly useful to understand that IC may
determine a firm’s role profile within the supply chain or
have a positive impact on a firm’s performance, it seems
important to know which factors of S–CR are impacted
The six constructs (CIP, MIP, CIM, R&D, SR, SD)
defining customer–supplier relationship were subjected
to a cluster analysis using the SPSS+Quick cluster
routine (Norusis, 1988). Since no initial cluster centres
are specified, Quick Cluster proceeds by selecting k
cases with well-separated, non-missing values as initial
centres with which to begin classifying cases, where k is
the number of clusters requested (in the specific case k
was equalled to 3). As cases are added to each cluster,
the program updates the selection criteria to the value
means for the cases (firms) thus far placed in a cluster.
The program uses a squared Euclidean distance measure
to assign a company to its nearest cluster.
Table 4 summarises the final mean value cluster
centres of the companies assigned to each of the three
supplier role profiles. The cluster profiles show the
differences in the responses to the six attributes. Thus
the study has revealed that customers do actually assign
suppliers different roles and give them varying levels of
responsibility in the product development and manu-
facturing stages. In general, positive cluster means
suggest that a supplier’s role has that characteristic
and negative cluster means suggest the opposite.
From the analysis of data related to the firms’
productive and competitive environment, it is possible
to draw a profile of the firms included in the three
distinct clusters.
A-type suppliers (n ¼ 38) typically manufacture pro-
ducts of high complexity (complete bottling and
packaging lines). Their relationship with customers
most resembles some form of fully blown partnership
for a number of reasons. First, intense communication
and exchange of early information about new products
and/or product modifications characterises customer–
supplier relationships. A-type suppliers are often in-
volved in product development even in the concept
creation stage. Second, R&D integration is brought
about mainly through the co-location of the suppliers’
engineers and joint utilisation of technical equipment
(such as rapid prototyping systems). Third, the nature of
the collaboration is to a great extent based on flexible
144 A. Petroni, B. Panciroli / European Journal of Purchasing & Supply Management 8 (2002) 135–149
Table 4
Final cluster means of suppliers roles
Characteristics of the relationships
Customer’s influence in product performance specifications
Mutual involvement in product development
Customer’s influence in manufacturing specifications
R&D Integration
Stability of relationship
Supplier development
contracts rather than tender-based and/or price-based
selection. On the other hand, the customer commitment
to SD is not prominent. This is probably due to the fact
that the supplier is considered to be of equal standing
(peers relationship).
B-type suppliers (n ¼ 51) have responsibilities which
include the provision of proprietary products or
complete subsystems such as machines for different
uses. These companies typically develop moderately
complex products or standardised components to be
assembled at a later date into final packaging lines (i.e.
wrappers, blister packers, automatic net loaders, etc).
The customer usually define the product performance
specifications in terms of cost, time and quality. A
distinctive trait of this cluster is that B-type suppliers
feel a strong pressure to reduce time-to-market,
although their product development responsibility is
lower than A-type companies. Unlike A-type firms,
NPD is largely driven by the customer and B-type firms
need to fine tune actions and programs to customer
specifications. B-type firms design and manufacture
complex assemblies, but because they lack developed
technological capabilities, they have less influence on
design. The customer provides critical specifications for
product performance and the supplier develops the
subsystem on its own. Intensive communication reaches
a peak in the design and production stages. The
competitive environment for B-type firms is very tough
since tender-based competition forces these firms to be
evaluated not only on the quality of the final product
but also upon the overall process followed to develop it.
B-type firms reported the highest score with regard to
the customer’s influence in manufacturing specifications.
The products that are supplied are, in fact, generally
more customised and specialised than simple basic
components and their timely supply and availability
thus becomes a critical condition for customers. As a
consequence, R&D integration is lower than for A-type
companies, as is the stability and continuity of relation-
ship with the customer. As expected, SD effort is the
highest for this cluster of firms. The customer, in fact,
solicits the supplier to fine-tune its operational proce-
dures in line with its own and to increase the quality of
operations and manufacturing processes.
Suppliers’ roles
A-type (n ¼ 38) B-type (n ¼ 51) C-type (n ¼ 109)
0.44 0.19
0.71
0.67 0.35
0.12

0.45 0.18
0.34
0.56
0.39
0.61
0.58 0.21
0.64
0.47 0.73
0.40
C-type firms (n ¼ 109) are manufacturers of basic
electronic and mechanical components (mainly pneu-
matic and magnetic parts). Because of the less specia-
lised nature of the work required, the pool of firms
capable of providing product and services is usually
larger at this level and many firms of this cluster are
largely de-specialised (operating in a number of different
industries). This type of supplier simply sells catalogue
products and seldom manufacture parts designed by the
customer. In this case, they usually receive detailed
drawings or even process plans from the customer’s
designers so that their main and only concern is to
manage shop floor activities. This supplier profile is
effective for customers that need a ‘‘contractual’’
extension of manufacturing capacity. On this basis,
there is little need for exchange of information in the
concept creation and design stages. Communication is
thus less intensive than for other supplier profiles. Long-
term customer–supplier relationships are usually
adopted in order to minimise transaction costs and
maximise procurement efficiency.
4.3. Subgroup regression analysis
A third area of results concerns the identification of
the attributes of the best performing suppliers in each
role profile. To this extent, the investigation of the
linkages between IC and performances was carried out
for each supplier type. Separate multiple regression
models were examined in each group with the IC
constructs as independent variables (predictor variables)
and the performance scores as the dependent variables.
A first analysis was conducted to uncover problems of
multicollinearity between independent variables. Table 5
reports means and correlation coefficients among all
independent variables. The data presented reveal no
problem of multicollinearity. The strongest correlation
coefficient (R ¼ 0:58; po0:001) relates to AC and TP.
Thus minimal redundancy exists between all indepen-
dent variables.
Table 6 presents the results of multiple hierarchical
regression analyses. The research findings make it
possible to draw a comparison among the innovative
profiles of the three clusters of firms in the sample. The
 A. Petroni, B. Panciroli / European Journal of Purchasing & Supply Management 8 (2002) 135–149 145

Table 5
Correlations and descriptive statistics

Absorptive Technological Technological Individual Managerial Innovation

capacity penetration scanning skills competence culture

Absorptive capacity 1.00

Technological penetration 0.59**** 1.00
Technological scanning 0.21 0.50 1.00
Individual skills 0.25 0.38** 0.15 1.00
Managerial competence 0.13*** 0.36* 0.16 0.26** 1.00
Innovation culture 0.36 0.15** 0.45 0.58*** 0.17 1.00
Mean 4.004 2.465 3.398 4.857 4.845 2.938

*po0:10; **po0:05; ***po0:01; **** po0:001:

Table 6
Results of regression analysis (n ¼ 198): standardised betas reported

Suppliers role

A-type B-type C-type

Independent variables Price Quality Flexibility Time Price Quality Flexibility Time Price Quality Flexibility Time

Absorptive capacity — 0.21* 0.15* — 0.09* — — — — — — —

Technological penetration 0.08* 0.23* — — 0.19* 0.07* 0.16* — 0.14** — — 0.13*
Technological scanning — 0.18* — — — 0.10** — — — — — —
Individual skills — 0.18* 0.05* — — 0.12* 0.24* 0.03* 0.09* — 0.13*** 0.09*
Managerial competence — 0.14* 0.03** — — 0.03* 0.25*** — 0.13*** — — 0.09*
Innovation culture — 0.18*** 0.09* — — 0.02* 0.34**** — 0.11* — — 0.10
Number of variables in equation 1 6 4 0 2 5 4 1 4 0 1 4
R2 5.25% 41.3% 35.6% — 11.6% 39.8% 37.7% 4.2% 30.7% — 6.45% 31.5%

*po0:10; ** po0:05; ***po0:01; **** po0:001:

Basic assumptions for conducting regression analysis are met: because of the large sample size, the assumption of multivariate normality is not
rejected; the independent variables are not highly correlated. Finally, the analysis of residues indicated no violation of basic assumptions.

subgroup comparison has been carried out by means of
t-test and the differences that have been reported are
significant at the 0.10 level or below.
As expected, suppliers that have established less stable
relationships with the customers are less inclined to
invest in technology and R&D. Moreover, the perfor-
mance attributes that are more influenced by IC (except
for C-type suppliers) are, in descending order of
importance: (i) quality, (ii) flexibility, (iii) price, and
(iv) lead-time. This conclusion can be drawn by looking
at the number of variables in the regression equations
and the variance explained for each performance
indicator. The impact of IC on performance is not
uniform. In general terms, IS are the strongest predictor
of performance (since this attribute has a statistically
significant influence on 8 out of 12 performance
indicators for the three types of suppliers considered),
followed by TP, a strong innovation-oriented culture,
MC, AC and, lastly, TS. The relative importance of IC
vary according to the type of suppliers considered.
Absorptive capacity, in other words making appropriate
investment in technological resources in order to
increase the ability to evaluate, assimilate, and exploit
extramural technological developments has some rele-
vance for A-type suppliers only and is almost negligible
for the other two types of suppliers. The number of
AMT and methodologies that have been successfully
adopted and their impact in terms of process innova-
tiveness seem to play a more essential role in explaining
the performances of B-type suppliers rather than those
of A-type and C-type suppliers. As expected, the ability
of a firm to identify opportunities, to understand and
foresee the technological strategies of competitors and
to evaluate emerging technologies is of some importance
for A-type suppliers only (TS was found to explain 18%
of quality performances). The tacit or explicit knowl-
edge and know-how of employees seem to represent a
major driver of performance for B and C-type suppliers.
For MC (which hints at the strategic role that is played
by managers in the acquisition, as well as the deploy-
ment and utilisation, of the technological resources that
are critical for sustaining the competitive position of the
firm over time), no substantial differences are noted
among the three clusters of firms. The diffusion of
innovation-oriented values and results-oriented culture
has been found to impact more heavily on the
performance of B-type suppliers rather than of A-type
suppliers.
146 A. Petroni, B. Panciroli / European Journal of Purchasing & Supply Management 8 (2002) 135–149
5. Discussion of the results
The relationship between A-type suppliers and
customers can be well described as a ‘‘partnership
between equals’’. Customers and suppliers have equal
standing in terms of their understanding of products and
technologies. A-type suppliers have, in fact, the global
reach that enables them to suggest solutions to meet
their customers’ price, time, quality and flexibility
objectives. In particular, the investment in R&D and
in training of technical personnel has been found to have
a strong positive effect on product quality and flexibility
of operations. This evidence lends support to the
hypothesis that this cluster includes larger suppliers
with organisational competencies that facilitate a full
integration of R&D and manufacturing processes. Few
suppliers can make the investments in training, technol-
ogies and research capabilities that a full-blown partner-
ship with their customers requires. Their distinctive
competence is mainly related to the ability to undertake
large projects and calls for capabilities in the domain of
integrated product design, manufacturing development
and product support. For this reason these companies
show outstanding performances and represent a ‘‘fron-
tier of excellence’’ in NPD that is largely boosted by the
successful adoption of carefully chosen leading-edge
technologies and methodologies. This small group of
more prominent suppliers has capitalised upon learning
methodologies, and thus developed a customised ‘‘tool-
box’’ of manufacturing technologies, well tailored to the
customer and its product lines. For instance, computer-
aided technologies are not considered merely as turnkey
appliances, but rather as tools supporting the organisa-
tional learning process. In other words, the adoption of
innovative manufacturing technologies and supporting
methodologies is generally aimed (either consciously or
otherwise) at improving the quality of products and to a
lower extent to obtain a cost-leadership advantage. The
competitive advantage is reached through the establish-
ment of long-term relationships with customers based
on mutual involvement and the overall ‘‘quality’’ of the
provision rather than on contingent price differentials.
According to this approach, TS activities are targeted at
obtaining technical information that may translate into
quality advantage. More than the employees’ tacit
knowledge of product technologies (which is common
to all supplier profiles), it is the individual capacity to
exchange relevant information together with a problem-
solving mind in the product development stages that
distinguishes these firms. Supplying good quality pro-
ducts increases the supplier’s reputation and owning
specific IC are a guarantee for effective partnerships.
On the other hand, for B-type suppliers, the results of
Table 6 show that their strategic imperative is ‘‘flex-
ibility’’ and product customisation. The highest stan-
dardised betas are in fact related to the impact of INC,
MC and IS on the flexibility performance. To create
a ‘‘flexible’’ organisation, these firms emphasise
the human side and the implementation of appropriate
managerial systems. It is, in fact, essential to have a
working climate that fosters result-oriented values and
motivates individuals to adopt a problem-solving-
oriented mind. Concurrently, the best performing firms
of this cluster show a distinctive profile in terms of
project management competence. One of the main
challenges for these firms is to find a balance between
formal managerial systems and innovation-oriented
culture, since rigidities in the project management
system may limit and prevent profitable customer
relationships in NPD and, on the other hand, response
flexibility may gradually decrease as the ‘‘project
culture’’ becomes dominant. Another distinctive char-
acteristic of B-type suppliers is that successful firms tend
to prefer to concentrate on the implementation of ‘‘soft’’
(methodologies) rather than ‘‘hard’’ technologies. A
possible explanation for this is that response flexibility
and product customisation are usually attained via
product modularisation (which often implies a high level
of coordination of operational activities) rather than
through actual engineering design process. In principle,
the ideal supplier for staging flexible integration with
their customers should be very skilled at managing the
product development process and have a large ‘‘tool-
box’’ of technological aids and methodologies. How-
ever, most of these suppliers have leaned quite heavily
upon the adoption of innovative technological tools,
albeit without caring too much about their integration.
Finally, with regard to C-type companies, the picture
is significantly different. These suppliers do not have
true product development responsibilities, so their main
concern is to manage shop floor operations. As
expected, IC are less critical for performances, while
MC has a limited impact on price and time performance.
This type of supplier is often too lean to locate engineers
in their customers’ facilities or to pledge major resources
to the development of products they will manufacture.
But this fact does not prevent them from reaching a
substantial success in their operations. The strategic
imperative for these suppliers is the overall efficiency of
operations, which is generally the result of employees’
tacit skills and know-how. These findings hint at the
importance of specialised know-how and of result-
oriented values for these firms. AMT (even if of limited
scope) translate into price and delivery time perfor-
mances that are critical to competition. Nonetheless,
technological integration at organisational level is very
limited. There is, in fact, a poor effort to exploit the
‘‘organisational integration potential’’ of technologies,
particularly when technologies are highly domain-
specific. This leads to make investments being made in
‘‘self-contained’’ technologies rather than in modular
resources to be installed and incrementally interfaced.
 A. Petroni, B. Panciroli / European Journal of Purchasing & Supply Management 8 (2002) 135–149
The majority of respondents reported that specific
methodologies are looked upon very suspiciously and
conceived mainly as ‘‘buzzwords’’. There is generally
little knowledge concerning the appropriate utilisation
of operational methodologies, their benefits and short-
comings. Whether this is due to the absence of actual
product development responsibilities or to ignorance of
‘‘off the shelf’’ advanced technologies and methodolo-
gies, is still to be investigated.
6. Limitations and generalisations of the study
Food machinery systems are often very complex
products since they are an assemblage of numerous
parts and subsystems. Although most complete lines are
variations on basic features with parts that are
incrementally modified in order to better respond to
the customer’s requirements, fitting all the subsystems
and parts together is crucial. The findings of this
research have thus the potential to provide lessons for
any company that manufactures mechanical products of
similar complexity. But they are probably less relevant
for high-tech industries whose products evolve rapidly
and continuously and for manufacturers of simple
products or highly customised products manufactured
in small batches.
The results should nevertheless be carefully inter-
preted in light of the following limitations. First, since
this study was conducted in a specific industrial sector
and in the particular context of supply firms, compre-
hensive generalisation is precluded. Furthermore, even
though reliability analysis shows that respondents dis-
play internal consistency, use of a single informant for
each firm might introduce some biases. Because of the
relatively limited sample size, statistical relationships
may not be as convincing as in a larger sample.
7. Conclusions and implications
The results of this study reveal a number of interesting
aspects of supplier performance in the food packaging
industry in Italy. Subcontracting firms which are
assessed as ‘‘better than their competitors’’ on four
important measures of manufacturing performance,
namely price, quality, delivery, and flexibility, also
possess distinctive innovative characteristics which are
associated with their performance. These innovative
characteristics have been found to vary significantly
according to the role and the type of product
responsibility that is assumed by the supplier.
There is, in fact, a range of postures that customers
and suppliers can adopt within a cooperative relation-
ship. Suppliers may play different roles for different
customers. Each posture carries fundamentally different
147
responsibilities in the product development and process
engineering stages and the customer-supplier relation-
ship vary considerably in closeness and intensity.
Accordingly, the innovative effort that is required from
the supplier has been found to vary. Successful supplier–
customer relationships, then, depend on the right
balance between the supplier’s technological capabil-
ities, the customer’s willingness to share information
and both companies’ strategic orientations.
The research that has been presented in this paper has
revealed that three types of suppliers can be identified in
the industry under examination.
For ‘‘partner’’ suppliers, the typical objective is to
improve and enforce integration with customer design,
manufacturing and logistics operations/processes. For
them, expenditure on development facilities, technical
skills and managerial capabilities is an investment in
partnership.
For the second type of supplier, particularly intense
efforts are spent on ensuring that personnel have a
sound attitude towards problem-solving, quality and
responsiveness to different environmental stimuli. These
firms have started broadening their reference market
and are competing on a wider base. This fact urges them
to attempt to become suppliers to a plurality of
manufacturers, since there is a considerable risk that
the effort to reinforce integration with a single customer
could inhibit their ability to search for profitable niches.
Specialisation and the wider market base have to go
hand in hand with response flexibility. This concept of
‘‘flexible specialisation’’ (Piore and Sabell, 1984) is of
particular relevance since it represents one possible
foundation of the capability of initiating intense but
agile customer–supplier relationships.
The prime competitive factor for the third type of
supplier resides in the technical skills and capabilities of
employees, which points to the crucial role of human
resources in achieving, maintaining and improving a
firm’s performance.
The implications of this research are twofold.
First, from a theoretical standpoint, the empirical
evidence tends to support the assumption that tradi-
tional innovative efforts and supportive organisational
capabilities (managerial practices and IS) have a
synergistic effect that largely depends upon the strategic
role assumed by the supplier.
From a practical perspective, the findings of this study
have implications for both customers and suppliers.
From the customer’s point of view, the study reveals
that the favoured Japanese practices suggesting fewer
suppliers and long-term relationships cannot be general-
ised beyond a certain limit. Partnerships with ‘‘contrac-
tual’’ suppliers manufacturing low priority and standard
components are dysfunctional. By choosing inappropri-
ate levels of responsibility for suppliers, a customer
may waste resources, urge suppliers to design highly
148 A. Petroni, B. Panciroli / European Journal of Purchasing & Supply Management 8 (2002) 135–149
customised parts when ‘‘off-the-shelf’’ parts are avail-
able and, most important, require suppliers to play a
role that is beyond the scope that their technological
base and competencies would allow. The paper, on the
other hand, maintains that for a buyer, adopting a set of
differentiated supplier relations is becoming a general-
ised necessity.
From the standpoint of the supplier, the research
findings clearly indicate that for those firms that make
it a goal to move up from a ‘‘contractual’’ to a ‘‘mature’’
or even a ‘‘partner’’ supplier, the development is
expensive and time-consuming. It does not only require
massive investments in new technology, but also to
undergo a deep process of building, nurturing
and developing specific capabilities in terms of TS
abilities, IS and, most important, organisational values.
This evolution is neither fast nor easy since it is highly
path-dependent, requiring a learning process that is
largely based on the firm’s previous experience and
history.
Acknowledgements
This research was funded by a grant from the Italian
National Research Council. The author is grateful to the
anonymous referees for their constructive criticism of
the originally submitted version.
References
Bensaou, M., 1999. Portfolios of buyer-supplier relationships. Sloan
Management Review 40 (4), 35–44.
Butcher, D., 1991. Computer integrated manufacturing in a world of
regions. In: Ayres, R.U., et al. (Ed.), Computer Integrated
Manufacturing. Economic and Social Impacts, Vol. IV. Chapman
& Hall, London, pp. 463–498.
Calderini, M., Cantamessa, M., 1997. Innovation paths in product
development: an empirical research. International Journal of
Production Research 51 (1,2), 1–18.
Carren, Pearson, 1999. Strategically managed buyer-supplier relation-
ships and performance outcomes. Journal of Operations Manage-
ment 17 (5), 497–519.
Chao, C., Scheuing, E.E., Ruch, W.A., 1993. Purchasing performance
evaluation: an investigation of different perspectives. International
Journal of Purchasing and Materials Management 29 (3), 33–39.
Clark, K.B., 1998. Project scope and project performance: the effect on
parts strategy and supplier involvement in product development.
Management Science 35 (10), 1247–1263.
Clark, K.B., Fujimoto, T., 1991. Product Development Performance.
Harvard Business School Press, Boston.
Clark, K.B., Wheelwright, S.C., 1992. Revolutionizing Product
Development: Quantum Leaps in Spead, Efficiency, and Quality.
Free Press, New York.
Cohen, W.M., Levinthal, D.A., 1990. Absorptive capacity: a new
perspective on learning and innovation. Administrative Science
Quarterly 35, 128–152.
Cohen, W.M., Levinthal, D.A., 1994. Fortune favors the prepared
firm. Management Science 40 (2), 227–251.
De Toni, A., Nassimbeni, G., 1995. Supply networks: Genesis, stability
and logistics implications. A comparative analysis of two districts.
Omega 23 (4), 403–418.
Dillion, W.R., Goldstein, M., 1984. Multivariate Analysis: Methods
and Applications. Wiley, New York.
Dyer, J.H., 1996a. How Chrysler Created an American Keiretsu.
Harvard Business Review, 42–56.
Dyer, J.H., 1996b. Does governance matter? Keiretsu alliances
and asset specificity as sources of Japanese competitive advantage:
evidence from the auto industry. Organization Science 6,
649–666.
Dyer, J.H., 1996c. Specialized supplier networks as a source of
competitive advantage: evidence from the auto industry. Strategic
Management Journal 17, 271–291.
Dyer, J.H., 1998. Strategic supplier segmentation: the next ‘‘best
practice’’ in supply chain management. California Management
Review 40 (2), 57–77.
Dyer, J.H., Ouchi, W.G., 1993. Japanese-style partnerships: giving
companies a competitive edge. Sloan Management Review Fall
33 (1), 51–63.
Hair, J.F., Anderson, R.E., Tatham, R.L., Black, W.C., 1995.
Multivariate Data Analysis. Prentice-Hall, Englewood Cliffs,
NJ.
Hartley, J.L., Meredith, J.R., McCutcheon, D., Kamath, R., 1997.
Suppliers’ contributions to product development: an exploratory
study. IEEE Transactions on Engineering Management 44 (3),
258–267.
Helper, S.R., Sako, M., 1995. Supplier relations in Japan and the
United States: are they converging? Sloan Management Review
Spring 36 (3), 77–84.
Holmen, E., Kristensen, P.S., 1998. Supplier roles in product
development: interaction versus task partitioning. European
Journal of Purchasing & Supply Management 4 (2–3), 185–193.
Hsuan, J., 1999. Impacts of supplier-buyer relationships on modular-
isation in new product development. European Journal of
Purchasing & Supply Management 5 (3–4), 197–209.
Imai, K., 1990. The legitimacy of japan’s corporate groups. Japan
Economics 17 (3), 23–28.
Imrie, R., Morris, J., 1992. A review of recent changes in buyer–
supplier relations. Omega-International Journal of Management
Science 20 (5–6), 641–652.
Kamath, R.R., Liker, J.K., 1994. A second look at Japanese product
development. Harvard Business Review 72 (6), 154–170.
Krause, D.R., 1999. The antecedents of buying firms’ efforts to
improve suppliers. Journal of Operations Management 17 (2),
205–224.
Lamming, R., 1986. For better or worse: technical change and
buyer supplier relationships. International Journal of Operations
Production Management 6 (5), 20–29.
Lamming, R., 1993. Beyond Partnership: Strategies for Innovation
and Lean Supply. Prentice-Hall, Englewood Cliffs, NJ.
Lefebvre, E., Lefebvre, L.A., Harvey, J., 1993. Competing interna-
tionally through multiple innovative efforts. R&D Management
23 (3), 227–237.
Leonard-Barton, D., 1991. The role of process innovation and
adaptation in attaining strategic technological capability. Interna-
tional Journal of Technology Management 6 (3,4), 303–321.
Leonard-Barton, D., 1992. Core capabilities and core rigidities: a
paradox in managing new product development. Strategic Manage-
ment Journal 13, 111–125.
Lindberg, Trygg, 1991. Manufacturing strategy in the value system.
International Journal of Operations Production Management
11 (3), 52–62.
Liu, X.R., White, S., 1997. The relative contribution of foreign
technology and domestic inputs to innovation in Chinese
manufacturing industries. Technovation 17 (3), 119–125.
 A. Petroni, B. Panciroli / European Journal of Purchasing & Supply Management 8 (2002) 135–149
Macbeth, D., Ferguson, N., 1994. Partnership Sourcing, An Integrated
Supply Chain Management Approach. Financial Times Pitman
Publishing, London.
Martin, X., Mitchell, W., Swaminathan, A., 1995. Recreating and
extending Japanese automobile buyer–supplier links in North
America. Strategic Management Journal 16 (8), 589–619.
Miller, P.A., Kelle, P., 1998. Quantitative support for buyer–supplier
negotiation in just-in-time purchasing. International Journal of
Purchasing and Materials Management 34 (2), 25–30.
Mowery, D.C., Oxley, J.E., Silverman, B.S., 1996. Strategic alliances
and interfirm knowledge transfer. Strategic Management Journal
17, 77–91.
Mummalaneni, V., Dubas, K.M., Chao, C., 1996. Chinese purchasing
managers’ preferences and trade-offs in supplier selection and
performance evaluation. Industrial Marketing Management 25 (2),
115–124.
Nishiguchi, T., 1994. Strategic Industrial Sourcing. Oxford University
Press, New York.
Nishiguchi, T., Brookfield, J., 1997. The evolution of Japanese
Subcontracting. Sloan Management Review Fall 39 (1), 89–101.
Norusis, M.J., 1988. SPSS/PC+Advanced Statistics V2.0 Chicago,
SPSS.
Piore, M.J., Sabel, C.F., 1984. The Second Industrial Divide.
Possibilities for Prosperity. Basic Books.
149
Roy, R., Potter, S., 1996. Managing engineering design in complex
supply chains. International Journal of Technology Management
12 (4), 403–421.
Saviotti, P.P., 1994. Knowledge, information and organizational
structures. Proceedings of the 11th International Economic History
Congress, Milan, Italy, 12–16.
Teece, D.J., Pisano, G., 1994. The dynamic capabilities of firms:
an introduction. Industrial and Corporate Change 3 (3),
537–556.
Verma, R., Pullman, M.E., 1998. An analysis of the supplier selection
process. Omega 26 (6), 739–750.
Von Hippel, E., 1990. Task partitioning: an innovation process
variable. Research Policy 19, 407–418.
Wilson, E.J., 1994. The relative importance of supplier selection
criteria: a review and update. International Journal of Purchasing
and Materials Management 30 (3), 35–8941.
Womack, J.P., Jones, D.T., Ross, D., 1990. The Machine that
Changed the World. Rawson Ass, New York.
Wynstra, F., Ten Pierick, E., 2000. Managing supplier involvement in
new product development: a portfolio approach. European Journal
of Purchasing & Supply Management 6 (1), 49–57.
Wynstra, F., van Weele, A., Axelsson, B., 1999. Purchasing involve-
ment in product development: a framework. European Journal of
Purchasing & Supply Management 5 (3,4), 129–141.