Euroma 2005 - Operations and Global Competitiveness : Operations and Global Competitiveness 9781846632518, 9781846632501

This e-book contains a selection of papers from the 12th international conference of the European Operations Management

152 70 1MB

English Pages 134 Year 2007

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

Euroma 2005 - Operations and Global Competitiveness : Operations and Global Competitiveness
 9781846632518, 9781846632501

Citation preview

ijopm cover (i).qxd

29/01/2007

11:53

Page 1

ISSN 0144-3577

Volume 27 Number 1 2007

International Journal of

Operations & Production Management EurOMA 2005: Operations and global competitiveness Guest Editor: Krisztina Demeter

www.emeraldinsight.com

International Journal of

ISSN 0144-3577

Operations & Production Management

Volume 27 Number 1 2007

EurOMA 2005: Operations and global competitiveness Guest Editor Krisztina Demeter

Access this journal online _________________________

3

Editorial advisory board __________________________

4

Guest editorial ___________________________________

5

The pervasive human resource picture in interdependent supply relationships Marie Koulikoff-Souviron and Alan Harrison________________________

8

Greening the automotive supply chain: a relationship perspective Dayna Simpson, Damien Power and Daniel Samson __________________

28

Managers’ perceptions of learning in new product development Ursula Koners and Keith Goffin __________________________________

49

How do subcontractors evolve? Andrea Furlan, Roberto Grandinetti and Arnaldo Camuffo ____________

Access this journal electronically The current and past volumes of this journal are available at:

www.emeraldinsight.com/0144-3577.htm You can also search more than 150 additional Emerald journals in Emerald Management Xtra (www.emeraldinsight.com) See page following contents for full details of what your access includes.

69

CONTENTS

CONTENTS

The impact of forecast information quality on supply chain performance

continued

Helena Forslund and Patrik Jonsson _______________________________

90

Structuring front office and back office work in service delivery systems: an empirical study of three design decisions Leonieke G. Zomerdijk and Jan de Vries ____________________________

108

Note from the publisher ___________________________

132

The 2nd Annual Emerald/EFMD Outstanding Doctoral Research Awards 2006___________________

134

www.emeraldinsight.com/ijopm.htm As a subscriber to this journal, you can benefit from instant, electronic access to this title via Emerald Management Xtra. Your access includes a variety of features that increase the value of your journal subscription.

Structured abstracts Emerald structured abstracts provide consistent, clear and informative summaries of the content of the articles, allowing faster evaluation of papers.

How to access this journal electronically

Additional complimentary services available

To benefit from electronic access to this journal, please contact [email protected] A set of login details will then be provided to you. Should you wish to access via IP, please provide these details in your e-mail. Once registration is completed, your institution will have instant access to all articles through the journal’s Table of Contents page at www.emeraldinsight.com/0144-3577.htm More information about the journal is also available at www.emeraldinsight.com/ ijopm.htm

Your access includes a variety of features that add to the functionality and value of your journal subscription:

Our liberal institution-wide licence allows everyone within your institution to access your journal electronically, making your subscription more cost-effective. Our web site has been designed to provide you with a comprehensive, simple system that needs only minimum administration. Access is available via IP authentication or username and password.

E-mail alert services These services allow you to be kept up to date with the latest additions to the journal via e-mail, as soon as new material enters the database. Further information about the services available can be found at www.emeraldinsight.com/alerts

Emerald online training services Visit www.emeraldinsight.com/training and take an Emerald online tour to help you get the most from your subscription.

Key features of Emerald electronic journals Automatic permission to make up to 25 copies of individual articles This facility can be used for training purposes, course notes, seminars etc. This only applies to articles of which Emerald owns copyright. For further details visit www.emeraldinsight.com/ copyright Online publishing and archiving As well as current volumes of the journal, you can also gain access to past volumes on the internet via Emerald Management Xtra. You can browse or search these databases for relevant articles. Key readings This feature provides abstracts of related articles chosen by the journal editor, selected to provide readers with current awareness of interesting articles from other publications in the field. Non-article content Material in our journals such as product information, industry trends, company news, conferences, etc. is available online and can be accessed by users. Reference linking Direct links from the journal article references to abstracts of the most influential articles cited. Where possible, this link is to the full text of the article. E-mail an article Allows users to e-mail links to relevant and interesting articles to another computer for later use, reference or printing purposes.

Xtra resources and collections When you register your journal subscription online, you will gain access to Xtra resources for Librarians, Faculty, Authors, Researchers, Deans and Managers. In addition you can access Emerald Collections, which include case studies, book reviews, guru interviews and literature reviews.

Emerald Research Connections An online meeting place for the research community where researchers present their own work and interests and seek other researchers for future projects. Register yourself or search our database of researchers at www.emeraldinsight.com/ connections

Choice of access Electronic access to this journal is available via a number of channels. Our web site www.emeraldinsight.com is the recommended means of electronic access, as it provides fully searchable and value added access to the complete content of the journal. However, you can also access and search the article content of this journal through the following journal delivery services: EBSCOHost Electronic Journals Service ejournals.ebsco.com Informatics J-Gate www.j-gate.informindia.co.in Ingenta www.ingenta.com Minerva Electronic Online Services www.minerva.at OCLC FirstSearch www.oclc.org/firstsearch SilverLinker www.ovid.com SwetsWise www.swetswise.com

Emerald Customer Support For customer support and technical help contact: E-mail [email protected] Web www.emeraldinsight.com/customercharter Tel +44 (0) 1274 785278 Fax +44 (0) 1274 785201

IJOPM 27,1

EDITORIAL BOARD Professor David Bennett Aston Business School, Birmingham, UK Professor Will Bertrand Eindhoven University of Technology, The Netherlands

4

Dr Ruth Boaden Manchester School of Management, UMIST, Manchester, UK Professor Harry Boer University of Aalborg, Aalborg, Denmark Professor Ken Boyer Michigan State University, USA Professor Richard B. Chase Marshall School of Business, USA Professor T.C. Edwin Cheng Hong Kong Polytechnic University, Hong Kong Professor Henrique Luiz Correa Crummer Graduate School of Business, Rollins College, Florida, USA Professor Paul Cousins Manchester Business School, UK Dr Simon Croom University of Warwick, UK Professor Robert Handfield College of Management, North Carolina State University, USA Professor Christer Karlsson Copenhagen Business School, Denmark Professor Bob Lowson East Anglia University, UK Professor Bart MacCarthy Nottingham University Business School, UK

International Journal of Operations & Production Management Vol. 27 No. 1, 2007 p. 4 # Emerald Group Publishing Limited 0144-3577

Professor Douglas Macbeth University of Southampton School of Management, UK Professor Jose´ Machuca University of Seville, Seville, Spain Professor Jack Meredith Wake Forest University, USA Professor Andy Neely Cranfield University, Bedfordshire, UK Professor Colin New School of Management, Cranfield University, Cranfield, UK Dr Carol Prahinski Ivey Business School, University of Western Ontario, Canada Professor Jaume Ribera IESE Business School, Spain Dr Nick Rich Lean Enterprise Research Centre, Cardiff University, UK Professor Giovani da Silveira University of Calgary, Canada Professor Nigel Slack Warwick Business School, Coventry, UK Professor Amrik S. Sohal Department of Business Management, Monash University, Caulfield East, Victoria, Australia Professor Andrea Vinelli University of Pa´dova, Vicenza, Italy Professor Chris Voss London Business School, UK

Guest editorial About the Guest Editor – Krisztina Demeter obtained her PhD degree at Budapest University of Economic Sciences and Public Administration in 2001 (today Corvinus University of Budapest, CUB). She has been associate professor since 2002. She played an important role to establish courses in operations and service management at CUB. She has numerous publications both in English and in Hungarian. She has responsibilities in the European Operations Management Association and the International Society for Inventory Research. Her major research fields are manufacturing strategy and supply chain management. She takes part in international (IMSS and GMRG) and Hungarian research programs (e.g. research on competitiveness, supplier capabilities in the automotive industry).

Operations and global competitiveness The 12th International Conference of the European Operations Management Association (EurOMA) took place in Budapest in June 2005 with the theme “Operations and Global Competitiveness”. This was the first occasion that EurOMA chose a venue in Central and Eastern Europe. The timing was perfect. Hungary had just joined the EU together with nine additional new members, thus extending the economic borders of Europe. This change inevitably has impacted on the operations of European companies by forcing them to internationalize their operations. This special issue is a selection of papers from the conference. The more than 300 participants could listen to almost 250 papers during the two days. The primary criterion for selection for this special issue was the quality of the papers rather than the exact topic since the numerous parallel sessions provided opportunities to present a paper on any topic within the operations management field. Selection was extremely difficult since many papers reached the required standards. The diversity of papers in this special issue correctly represents the variegation in the conference. It includes some rarer but up-coming topics as well as some traditional issues. The paper by Marie Koulikoff-Souviron (CERAM Sophia Antipolis) and Alan Harrison (Cranfield University) discusses human resource issues, which is still a very under developed field of research within OM. On the basis of 48 interviews the authors analyze the role of HR in the supply relationship of two companies. They conclude that human issues play a pervasive role in these relationships and we have to deal with it in more depth in order to get better understanding of the contribution of HR to OM success. Another burgeoning topic is environmental protection. Dayna Simpson, Damien Power and Daniel Samson (University of Melbourne) investigate it in the Australian automotive industry, using case studies. According to their analysis, in order to green supply chains, customers first have to commit themselves to environmental protection. They conclude that the chance of greening the supply chain is higher if:

Guest editorial

5

International Journal of Operations & Production Management Vol. 27 No. 1, 2007 pp. 5-7 q Emerald Group Publishing Limited 0144-3577

IJOPM 27,1

. .

.

6

they rely more on a given supplier, they buy a specific product on a contractual basis as opposed to buying commodities on the market, or they strictly monitor the environmental activities of their suppliers.

But again, the customers have to set the example. The third paper by Ursula Koners and Keith Goffin (Cranfield University) also cannot be considered as a traditional topic. When project management is taught in classes, we emphasize the importance of post-project reviews for learning. However, there is very little knowledge in the literature about the number of companies who consciously do it, the benefits they obtain, the way they do it and so on. This paper gives insights into this issue, based on data from five case studies. Knowledge codification is a crucial point in the paper by Andrea Furlan, Roberto Grandinetti and Arnaldo Camuffo (University of Padova). They involved 417 SMEs in their survey to find typical profiles of the level of design and marketing capabilities. Then they selected ten firms to see, through cross-case analysis, how new capabilities are developed. They find that faster development can be achieved with conscious knowledge codification, which helps SMEs to improve their design, marketing and supply management capabilities. Another supply chain related paper was that by Helena Forslund (Va¨xjo¨ University) and Patrik Jonsson (Chalmers University of Technology). They analyze the value of customer forecast information quality in relation to supply chain performance. The authors develop a construct for forecast information quality, and test it with 136 Swedish companies. They come to the conclusion that the forecast information quality has to be improved considerably in order to have a real impact on supply chain performance. Beside so many manufacturing issues, fortunately there is one paper from the service sector in this issue. Leonike Zomerdijk (London Business School) and Jan de Vries (University of Groningen) discuss the design of service delivery systems. In their analysis, based on five case studies, they investigate and justify three major decision areas and their trade-offs: the level of customer contact, the contact between front and back office (decoupling), and the grouping of employees. The rich analysis provides an excellent basis for further thinking and research. Although there are only six papers in this conference special issue, some further conference papers are still on the way, to be published in later IJOPM issues. And finally some acknowledgements. First of all, thanks to the Editors of IJOPM, Margaret Webster and Andrew Taylor who supported my Guest editorial work with their cautious remarks, suggestions and control. I would like to thank the EurOMA board for the trust and help they provided in the organizational issues before, during and after the conference. Special thanks to Chris Voss, Harry Boer and Jan Olhager who gave several suggestions for papers to be included. Finally, the referees of the special issue really increased the level of quality with their deep and wise advice in their reviews. I think they deserve to be mentioned, in alphabetic order: Nuran Acur, Per Ahlstrom, Juan Alberto Arago´n Correa, John Bessant, Harry Boer, Peter Bruun, Raffaella Cagliano, Henrique Luiz Correa, Mariano Corso, Simon Croom,

Cipriano Forza, Brian Fynes, Matthias Holweg, Christer Karlsson, Peter Kelle, Jose´ Machuca, John Meredith, Peter Milling, Guido Nassimbeni, Jan Olhager, Ken Platts, Jaume Ribera, Daniel Samson, Nigel Slack, Dave Tansik, Ann Vereecke, Bart Vos, Chris Voss, John Wacker, Clay Whybark. Have a good reading.

Guest editorial

Krisztina Demeter Guest Editor

7

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0144-3577.htm

IJOPM 27,1

The pervasive human resource picture in interdependent supply relationships

8

Marie Koulikoff-Souviron CERAM Sophia Antipolis, Sophia Antipolis, France, and

Alan Harrison Cranfield School of Management, Bedford, UK Abstract Purpose – This paper aims to explore the influence of human resource (HR) practices in shaping inter-organisational supply relationships, and to describe the extent to which supply partners seek to adapt their internal HR practices to the external relationship. Design/methodology/approach – Inter-organisational relationships are known to be multi-dimensional in nature. Using seven dimensions to collect empirical evidence, the focus was on a dyadic supply relationship as the unit of analysis. A total of 48 interviews was undertaken from informants in both organisations. Findings – It is shown that the supply relationship tends to be taken for granted at operational levels, and not subject to the relatively high levels of strategic alignment to be found at senior levels in both organizations. Operational aspects of the relationship are tested primarily in failure mode when poorly prepared to respond, for example, as a result of facility breakdowns and quality problems. Inter-organisational HR practices were relatively weakly prioritised and – despite their acknowledged importance at operator level – were not clearly identified and supported in the way that intra-organisational practices were. Originality/value – An in-depth, operations-based study of HR practices has been undertaken in a reciprocally interdependent supply relationship from the perspectives of both partners. Keywords Supply chain management, Adaptabilty, Human resource strategies Paper type Research paper

International Journal of Operations & Production Management Vol. 27 No. 1, 2007 pp. 8-27 q Emerald Group Publishing Limited 0144-3577 DOI 10.1108/01443570710714510

Introduction Studies which link human resource (HR) issues with operations management (OM) are still relatively few in number. The two sets of literature tend to “end up talking past each other” (Harrison and Storey, 1996). The OM literature has “paid little attention to HR issues” which is surprising given the “critical role” of HR in achieving superior performance in operations competitive priorities (Ahmad and Schroeder, 2003, p. 20). From an operations perspective, HR practices can be roughly categorised into those relating to internal consistency with a firm’s competitive priorities (Hayes and Wheelwright, 1984, p. 130), and those relating to external consistency with the priorities the firm has agreed with its trading partners. The need for internal consistency has been widely commented on, largely in disciplines other than OM. For example, the “typology of organisations” of Adler and Borys (1996, p. 78) proposes a “fit” between the type of formalisation (enabling or coercive) and the relative routineness of the task. Other studies of internal consistency develop the argument for “bundles” of HR practices (McDuffie, 1995; Delery, 1998), or for human resource management (HRM) “systems” (Becker and Huselid, 1998).

Studies on external consistency of HR practices are less common. However, empirical evidence concerning HR issues within inter-organisational relationships has begun to emerge (Hunter et al., 1996; Scarbrough, 2000; Rubery et al., 2004). A recurring theme in these studies has been the influence exerted by customers on the internal organisation of supplier firms. For example, Rubery et al. (2004, p. 1204) contrast internal employment policies with client pressures to show that a number of tensions may be created. Clients may exert pressure on a supplier’s HR practices in areas such as hours worked (to fit with their own), performance assessment and associated bonus payments. Indeed, Guest et al. (2003, p. 293) argue that a focus on external “fit” implies that a distinctive form of HRM, however, it may be characterised, will not be equally effective in organisations with different business strategies. As Scarbrough (2000, p. 16) puts it “the supply chain creates important demands on HRM in securing the necessary skills and behavioural flexibility to support tighter integration”. Such arguments suggest that firms may be willing to adapt, with the result that they affect each other in relatively enduring ways (Halle´n et al., 1991). This view is relevant to the operation of buyer-supplier relationships, and helps to inform the development of OM theory in this complex area. This adaptation may range from unilateral (one partner dominates) to reciprocal (a mutual process). Thompson (1967) and Richardson (1972) refer to the similar concept of “mutual adjustment” in coordinating reciprocally interdependent (RI) supply relationships, which we explain in the next section. The adaptive view assumes that supply partners will accommodate their HR practices in order to manage inter-organisational relationships better (Halle´n et al., 1991; Wilson, 1995; Lei et al., 1997). The aim of this paper is to explore the influence of HR practices in shaping such relationships, and to describe the extent to which supply partners seek to adapt their internal HR practices to the external relationship. We operationalised this aim by investigating a RI supply relationship in depth from the point of view of both partners. We were guided by research questions which probed influence and adaptation in practice. We have organised our paper into four further sections. First, we explore the role of HR practices in shaping and adapting supply relationships and develop our research questions; second, we explain our research methodology and provide the context for the supply relationship we investigated; third, we present our findings; and finally we present our discussion of the findings and our conclusions. Influencing and adapting supply relationships HR practices as an influence on supply relationships A familiar concept in OM is the distinction between “arm’s-length” (ACR) and “obligational” contractual relationships (OCR), (Sako, 1992). Following Sako’s view that these are extremes of a spectrum of possibilities, Speckman et al. (1998, p. 634) illustrate the transition between ACR and OCR as a linear sequence which requires “changes in mindset and strategic orientation” at each step, shown in Figure 1. Moving from left to right, the sequence suggests that supply chain partners become more dependent on one other, i.e. that they become interdependent. One of the architects of organisational theory – Thompson (1967) – identifies three ways in which interdependency develops. The first is pooled interdependency, whereby partners share and use common resources but are otherwise independent. Second is sequential interdependency, where partners work in series and the output from one becomes the

The pervasive human resource picture 9

IJOPM 27,1

10

input to the next. Third is reciprocal interdependency, in which the outputs of each become the inputs for the others, so that each partner poses contingency for the other, and coordination takes place by mutual adjustment (Thompson, 1967, pp. 54-56). Sequential interdependency describes the traditional, unidirectional view of buyer-supplier coordination (source ! make ! deliver). Reciprocal interdependency describes bi-directional coordination (source , make , deliver). Reciprocal interdependence requires the most elaborate form of inter-organisational exchange (Thompson, 1967; Richardson, 1972), which is justified by criticality of supply. Indeed, the more critical the supply, the more the partners will desire a relationship that offers some degree of reliability (manifested by measures such as product quality and process capability). This places increased emphasis on long-term partnering arrangements (Marchington and Vincent, 2004; Gadde and Snehota, 2000), and on mutually beneficial adaptation of existing structures (Richardson, 1972; Halle´n et al., 1991). “Increasingly heavy burdens” (Thompson, 1967) are placed on information sharing and decision-making in moving from pooled to sequential to reciprocal situations. But these increasingly complex settings create barriers to information exchange in the form of difficulty in pinpointing sources of error and constraints (Bendoly et al., 2006, p. 560). Interdependence is a key distinction between ACR and OCR, along with the time span for reciprocity and interaction pattern (Sako, 1992, p. 4). Here, we focus on dyadic, RI supply relationships which are characterised by OCR. Marchington and Vincent (2004, p. 1034) argue that such relationships are characterized at institutional, organisational and interpersonal levels. This distinction is important because it enables us to distinguish between factors that may influence RI relationships from “those that influence initial choices about potential partners by senior managers in the organization, and the day-to-day maintenance of the relationship by boundary-spanning agents”. We argue that HR practices influence obligational aspects of such relationships, and set out to test this hypothesis by means of our first research question: RQ1. How do HR practices influence a RI supply relationship? Adaptation of HR practices in supply relationships Adaptation in supply relationships has different theoretical perspectives. Coordination of the heterogeneous needs of the different supply partners may involve adaptation through altering internal processes to accommodate the other party (Borys and Jemison, 1989; Halle´n et al., 1991; Haokansson and Snehota, 1995; Araujo et al., 2003). In their study of the effect of supply relationships on quality practices and performance, Fynes and Voss (2002, p. 14) measured adaptation in terms of the changes to production processes and tooling that a supplier needed to make in order to match the customer

Open market negotiations Figure 1. Transitions leading to supply chain partnership

Price-based negotiations Adversarial relationships

Cooperation

Fewer suppliers Longer-term contracts

Source: After Spekman et al (1998)

Coordination

Information links e-enablement, EDI

Collaboration

SC Integration Joint SCstrategies

requirements. In other words, these authors focused on technical adaptation. Storey et al. (2005, p. 244) point out that there can be enormous organisational and behavioural barriers to technical adaptation: “technological capability is by no means sufficient, and inter-organisational cooperation can be hard-won”. A parallel perspective is that of “hard” and “soft” technologies. “Hard” technologies are basically products and processes, while “soft” technologies are people dependent. Brannen (2004, p. 597) considers that softer technologies are “more vulnerable to misalignment, because they are more closely embedded in the socio-cultural context of the recipient culture”. In his study of the transfer and adaptation of management systems to Honda subsidiaries, Mair (1998, p. 301) states that the production system is being transferred intact, while adaptation is confined to “softer” aspects of organisation and employment relations to provide a fit between the production system and the local operating environment. Adaptation means making changes to a firm’s internal processes in order to accommodate the needs of supply partners. Gro˝ssler et al. (2006, p. 256) argue that the process of adaptation can take three forms: explicit (changes in a firm’s guiding principles), implicit (autonomous changes that are independent of the management system) and a mixture (of both explicit and implicit). In the case of knowledge-intensive firms, Swart and Kinnie, 2003 found evidence of the exercise of “both strategic choice and tight constraints” as clients sought to influence HRM in their suppliers. Halle´n et al. (1991) expect adaptation to occur more symbiotically in long-term supply relationships where buyer and supplier form significant parts of each other’s business. One would expect that “suppliers adapt to the needs of specific important customers as well as that customers adapt to the capabilities of specific suppliers” (Halle´n et al., 1991, p. 30). Lepak and Snell (1999, p. 41) add that HR systems that “encourage and reward cooperation, collaboration and information sharing” are also likely to be necessary in such relationships. Hunter et al. (1996) take the view that supply partners may be able to develop a common set of control mechanisms to operate the joint relationship by adapting their internal practices. While the initiative for the change is generally customer-led, adaptation may not be unilateral in that customers may also have to alter their existing systems: The two organisations will still be subject to independent governance but will have in common a similar set of governance procedures and mechanisms specific to their joint working relationship, thus replicating in some measure the conditions within an integrated organisation (Hunter et al., 1996, p. 244).

There is mixed evidence as to whether HR practices must be adapted, or whether they remain relatively insulated from supply chain effects (Scarbrough, 2000). Thus, Roper et al. (1997) found that new procedures and training, required within a supply relationship, received only cosmetic attention and were only adopted superficially. Scarbrough also contends that there is a conflict between the external requirements from supply relationships – which aim for tighter integration – and the internal hierarchical work organisation. He argues that some HR practices such as training, team working and socialisation may be more easily adapted than hierarchically-determined conditions of employment, work rules and payment systems, which are linked to the organisation as a whole. Scarbrough (2000) helped to pioneer the issue of HRM within supply chains. However, his study of three companies was limited to changes within suppliers’

The pervasive human resource picture 11

IJOPM 27,1

12

internal organisations – and so did not engage with the mutual adaptation considered to be necessary within reciprocal, long-term relationships. Similarly, Rubery et al. (2004, p. 1218) studied the way that “the external organisation may extend its tentacles inside the organisation to reshape the internal HR practices” again limiting the study to the extent of the supplier’s adjustment to client demands. The unit of analysis in these studies was based on the single firm. This results in lack of an inter-organisational perspective in terms of both partners to the relationship. We sought to address this gap by means of our second research question: RQ2. How do firms adapt their HR practices as a result of a RI supply relationship? Methodology There have been calls for more case studies and field research in OM (for example, Meredith and Samson, 2002). Because of the exploratory nature of this investigation, we selected case study methodology (Yin, 2003) as our research strategy in order to develop an in-depth understanding of HR practices within a supply relationship. The supply dyad we selected provided exceptional opportunity for access by both organisations, one of which was a former employer of one of us. We selected a single case because of the opportunity to provide the “greater depth” (Voss et al., 2002) needed to address the operational issues inherent in our research questions. While this approach is not without risk from the aspect of lack of comparative logic (Eisenhardt, 1991) and the accompanying limits on generalisability of conclusions (Voss et al., 2002), it nevertheless provided the opportunity to develop new theory in an area characterised by messiness and large volumes of qualitative data. The relationship was a closed-loop supply relationship between two industrial partners, who were both customer of, and supplier to, each other. Operationally, this provided a relevant context for the RI supply relationship that was the focus of our enquiry. We used on the logistics process (Furlan et al., 2006) as the focus of our enquiry – as distinct from pricing, knowledge transfer or new product development processes. Details of the case are provided in the next section. A feature of the research design was 48 semi-structured interviews with a wide cross-section of employees and managers within the dyad. The aim was to collect data, which was “pluralist” in nature, hence providing competing versions of reality (Pettigrew, 1990). We sought to collect data on the perspectives of informants at different levels. These levels included operators, process engineers, local and corporate management as shown in Table I. The rationale for the choice of informants aimed to address a broad range of interviewees from each of the units involved in the supply relationship. We also wanted to include a broad functional representation across levels. We began with key managers in charge of the relationship and followed leads from these initial interviews.

Table I. Details of informant sampling (48 interviews in total, with some informants interviewed more than once)

Operators Engineers Management Corporate Total

Wheatco TCS

Chemco

Wheatco Rubber

Total

2 6 4 1 18

4 10 3 1 18

1 2 2

7 18 9 2 36

Especially, revealing were the views of corresponding informants on both “sides” of the supply relationship. The main wave of data collection was complemented by follow-up interviews in order to corroborate data through peripheral sampling, and to gather new data on how the relationship was perceived to be evolving over the course of our study. Interviews with corporate informants provided an alternative perspective on the local situation whilst allowing a view of the case from interconnected levels of analysis (Pettigrew, 1990). Interviews began with an introductory phase, where key objectives of the research were presented. The informant’s role in the relationship was then discussed, together with that of their counterpart in the other firm. The next stage of the interview centred on the informant’s general perception of the supply relationship. The requirements of the relationship in terms of HR were explored towards the end of the interviews. Our interview guide is shown in the Appendix. The length and protocol for conducting interviews evolved over the course of the research. To begin with, interviews tended to be rather informal and lasted from 1 to 3 hours, to provide an in-depth understanding of the research settings. At a later stage, they were more focused and structured, as they were meant to provide specific additional evidence or to verify earlier research findings (Pitman and Maxwell, 1992). The study drew on other sources of evidence, such as documentation and observation. Some documents – such as joint meeting minutes, contract agreements, job descriptions and HR procedures – were directly relevant. Others, such as company brochures and organisation charts, pertained to the context at company or site level. We collected evidence for the influence of HR practices and their adaptation from seven measures: (1) Staffing. Investigated the choice of appropriate persons to work within the supply relationship (Pucik, 1988; Ring and Van de Ven, 1992). (2) Job design. Looked at the process of combining tasks and responsibilities to form jobs and the relationships of jobs within the supply relationship (Cascio and Serapio, 1991; Kanter, 1994). (3) Appraisal. Investigated the formal or informal system to provide information about how well jobs were performed and objectives met within the relationship (Lorange, 1986; Lei et al., 1997). (4) Rewards. Looked at the exchange by organisations for the contributions of their employees and how these pertained to the relationship (Lei et al., 1997; Kanter, 1994). (5) Training. Investigated the attempt to improve current or future employee performance by increasing (through learning) their ability to perform within the relationship (Pucik, 1988; Hunter et al., 1996). (6) Socialisation. Looked at the process by which employees and managers from both partners learnt about and adapted to jobs involving the other organisation’s workplace (Jick, 1979; Lei et al., 1997). (7) Communication. Investigated the provision of information to employees concerning aspects of their employment and wider issues relating to the organisation in which they worked, and the relationship in which they took part (Pucik, 1988; Lei et al., 1997).

The pervasive human resource picture 13

IJOPM 27,1

14

While these measures were not intended to be comprehensive, they nevertheless enabled us to probe the RI relationship across a broad front. The study took place over a nine-month period (August 2000-February 2001), which enabled us to become familiar with the setting whilst not becoming over-involved (Pettigrew, 1990); this time-span also allowed us to capture longitudinal aspects of the relationship. The Wheatco-Chemco relationship Wheatco and Chemco are two US chemical corporations, both leaders in their fields and with similar sales (around $2bn per year). Eight years prior to our study, the two companies formed a partnership with the strategic objective of gaining competitive advantage through mutual access to low-cost raw materials. One outcome was the establishment in the UK of a small Chemco facility (70 employees) on a large Wheatco site (700 employees). The Chemco facility was located next to the Wheatco “Basics” unit, and linked by a bridge. While a fence divided the two plants, selected employees were able to pass between the two by means of swipe card access. A Chemco manager commented: We are symbiotically linked. If you take away the Chemco and Wheatco signs, we’re really one site . . . we have a relationship and it’s an umbilical cord.

Chemco was dedicated to production of a chemical additive used in the production of rubbers, paints and other compositions. The feedstock used in the Chemco process was supplied by the Wheatco “Basics” unit. The manufacturing process of the additive generated a gaseous by-product, which was recycled back into the Wheatco feedstock. Half of the additive made on the Chemco site was sold to Wheatco’s “Rubber” unit, and the rest to other customers in Europe and the USA (Figure 2). The two firms thereby formed a “closed loop” supply chain – whereby they were both customer of, and supplier to, each other. The production processes operated on a round-the-clock basis and there was very little buffer stock within the supply loop: “if we have a problem, then Chemco has a problem 10 seconds later”. This close interdependency of the processes meant that the operating teams were in contact on a 24-hour basis. There was a direct telephone link between Wheatco and Chemco operators to allow easy communication and instant warning of changes in either of the processes, or to inform of production stoppages. The supply relationship was multifaceted, with interactions taking place at many levels. Locally it included plant management, engineers and operators. In the USA, Recycled B3 Gas

Raw material supply

Figure 2. The Wheatco-Chemco supply process

Chemco Chemical additive (A1) manufacture

Basics-TCS building Feedstock 1 & 2 manufacture (Fluid bed reactor) B2 gas Wheatco

Rubber Manufacture, Building 115

Other customers Chemco

an executive contact was appointed by each firm to manage the relationship at a strategic level. This applied in particular to the global contract agreement, which provided the commercial terms for the relationship. A joint Steering Committee determined the local operational strategy for the relationship and provided guidelines to two other joint teams: “quality improvement” and “technical”. We began collecting data in August 2000, which was eight years after the supply relationship had started. At that time, the upstream Wheatco process was recurrently unreliable. There were also quality issues with the chemical additive supplied by Chemco, which impacted rubber production at the downstream Wheatco unit. In the early days of the relationship, operators had been encouraged to socialise through company events and plant visits. This allowed a common language to be developed, through interaction: “we may spend a day there, they spend a day here” and thus “we didn’t need to communicate where if something did go wrong they would automatically take care of it”. More recently, the relationship had become arm’s length. Indeed, both partners were busy implementing internal programs which drew attention away from the local relationship. At shop floor level, less interaction and fewer visits were allowed. This was made worse by employee turnover. As a consequence, operators felt that they could no longer “put a face to a name”. Lack of interaction, together with the recurring technical issues, put a strain on the overall relationship. Recognizing that a blame culture had developed, site management from Wheatco and Chemco decided to organise a “Team Day” to ensure that operators, shift managers and engineers from the three manufacturing units could meet, socialise and be trained on the specificities of the supply loop. However, the “Team Day” was cancelled due to a company-wide workforce reduction plan announced by Wheatco: given the circumstances, such a socialisation event was seen as inappropriate. Findings We review the evidence we collected against each of the seven dimensions listed in the foregoing section in sequence. Staffing Two HR decisions from Wheatco had recently impacted the relationship at shop floor level: (1) The TCS operator team had been moved to shift manager positions, which meant that Chemco had to deal with a whole team of newly appointed Wheatco TCS operators. (2) A cross-training (job rotation) approach was introduced within the Wheatco site, at about the same time. Chemco operators perceived the new Wheatco operators as lacking experience and were also frustrated at having to do with several interfaces during one shift. This translated into a feeling of lack of priority for the relationship: A feeling from our operators that the TCS plant is used as a kind of training ground for operators that then move on to the core Wheatco competence places (Chemco manager). If they could keep their operators at the TCS plant for longer rather than move them on to other plants I think that is the biggest part for improvement (Chemco operator).

The pervasive human resource picture 15

IJOPM 27,1

The effect of people turnover on the supply relationship was summarised by one of the Chemco operators: Until you work with somebody for a while you start to think yes, they’re good or they’re telling you the whole facts, then you get a bit more confident that when they say something’s wrong, as opposed to I don’t know what’s going wrong sort of thing.

16

A direct link was established between the level of quality of the people that were assigned to work within the relationship and its performance: We’ve often been frustrated in the past with the lack of progress and I think it’s mainly been due to the calibre of the people involved (Wheatco management).

Thus, the quality of the people appointed to work on the relationship was linked to the level of priority allocated to it. Staffing was perceived as a practice driven by internal priorities: I don’t think we specifically say he is exceptionally good so we should put him in this sort of relationship (Wheatco). I don’t think we’ve ever selected people, and given any thought to the relationship to be frank (Chemco).

One exception was when Chemco recruited one of Wheatco’s former employees specifically to act as point of contact and manage the relationship with the partner. Job design One difficulty of working within this supply relationship was the lack of understanding of the other plant’s internal work organization: If you don’t understand, you assume that the other company does things the same way as you do and it may not be the case (Wheatco management).

Thus, the same job title could cover different job contents in both companies. Another difference was that, Wheatco being a unionized site, job design there was more structured and less flexible than Chemco’s: Wheatco’s system seems to be so rigid that they have to wait for an instrument man to come in, an electrician to come in, whereas our people would just rip the thing apart and get it done.

The “cross-training” of TCS operators created a tension amongst Chemco operators because of the increased turnover, when in reality this work practice may have enabled TCS operators to have more time to spend on the relationship since backup systems could be organised with other operators. It should also have allowed more robustness on the TCS process as more people were trained to run it. However: In the short and medium term it means that we have operators with less skills (Chemco).

Thus, differences in work practices meant that specific communication needed to take place to avoid misinterpretation from the other firm. Job design was seen as an internal practice and Wheatco resisted any suggestion from Chemco to change, for example, their cross-training practice, which was an HR decision at overall site level: I don’t think we are going to change our organisational structure to something that Chemco think would be better because I think we know internally what is better for us.

And there’s a lot more to take into account overall than just the operation of the TCS process (Wheatco management).

Appraisal In both companies, everybody at every level forms their own opinion of the person they interface with. (. . .) I think between the two of us there is a lot of informal people management (Wheatco management).

There was a lot of evidence across both firms of informal assessment of the counterpart, both in terms of technical competence: I think their process operators are good, as well as relational capabilities: Someone wasn’t as friendly or helpful as he could be.

The shared local operational goals, pertaining to the joint relationship, were translated into individual employee’s objectives. Indeed, there was evidence that the joint operational goals drove, to a greater or lesser extent, the employee’s performance objectives or performance standards. There was a specific instance where negative feedback expressed by Wheatco managers contributed to the decision to move a Chemco Quality manager out of the relationship. Thus, the external input was used as an element of appraisal, but only as an influencing factor. Indeed, decisions about the ultimate outcome of the appraisal took place internally: To some extent the decisions we’re taking have been influenced by him (. . .) The change we made on quality was happening anyway. All his comment did was reinforce that this was the right thing to do (Chemco).

Rewards Bonus schemes were used by both partners as part of their performance management process, including shop floor. However, the two systems were managed differently in that the Chemco bonus was linked to the chemical additive process performance and it was therefore impacted by the downtime caused by Wheatco. On the other hand, the Wheatco bonus was related to goals at overall site level and therefore not directly linked to the Wheatco-Chemco process performance. This meant that Chemco operators could be negatively impacted by downtime caused by Wheatco, whilst their counterparts were not penalized. This generated a feeling of grievance that worsened the relationships between the two operating rooms: They’ve got a terrible year and still got a good recognition. It doesn’t motivate them to succeed does it? (Chemco operator). Chemco explained their bonus scheme, which makes their operator’s bonus dependent to some extent on Wheatco performance. This can lead to frustration and situations where there is a lack of trust in communication with Chemco (Steering team meeting minutes).

The Chemco reward system was perceived as disruptive by Wheatco management, whereas the other party viewed it as necessary in order to drive the Chemco site results. Conversely, Chemco managers wanted Wheatco to adapt their rewards so that TCS people were more directly linked to the way their process performed with respect to Chemco. However, the attempt failed because Wheatco management felt this to be an internal work organisation issue.

The pervasive human resource picture 17

IJOPM 27,1

18

Although there was some discussion around the possibility of organising some joint rewards, these were not implemented. Barriers that prevented such joint practices were mainly at Wheatco who would see a conflict between such rewards and their performance management system at overall plant level: Whatever the others would say: “Why not me?” (Wheatco); The barrier was if we have a joint bonus between our people and their 20 people, the other 980 Wheatco people might go on strike if they think their colleagues have got another bonus that they didn’t get (Chemco).

Training It was recognized that operator visits to the other plant served as useful “on the job” learning mechanism because they allowed better understanding of the mutual effects of the joint process. However, there was no record keeping of operator visits across the manufacturing units, hence this was not considered as being part of a formal training plan: I’ve got no documentation to say somebody’s been over there. It’s just on an ad hoc basis really.

Informants argued that a formal programme should be set up in order to ensure that interaction took place. Such training needed to be part of the formal management of the supply relationship: It would be good to have a programme throughout the year of Chemco-Wheatco getting together, both operators and engineers. That could be just a rolling programme throughout the year. Just as we have technical meetings. It would be good to have something similar going on for operators and everyone as a whole (Wheatco).

The induction of new people was raised as a specific issue. Indeed, new people could disrupt the relationship, because they did not understand the unique features of the intricate reciprocal supply relationship: . . . the new people coming in, that sort of relationship is foreign to them, and you have to get them into that relationship (. . .) until they sort of understand the tie up between the two.

This could result in slowing down decision-making: . . . as newcomers did not understand initially perhaps the overall importance of the relationship and have become blockers (Chemco).

Understanding about the other plant appeared more as an issue of informal, “on the job” learning – a socialisation process, rather than formal training. Most employees expressed regret that the “Team day” was cancelled. The intention was to allow personnel from the three manufacturing units to meet for one day for teambuilding activities and to learn about the specific characteristics of the supply “loop”. Hence, the only joint training that took place was a statistical process control (SPC) course, which was organised by Wheatco and included a number of Chemco participants. Informants from both firms perceived this very positively, in that it created a common language: Because we all understand (SPC) the same way. I come to understand their jargon, and they’ve come to understand mine, so we’re “bilingual” you could say (Wheatco).

Socialisation The ability to meet people from the other organisation “face to face” appeared as a key socialisation practice, in particular at operator level. Indeed, the operator link was the most intense and was critical for the coordination of the joint process, which required 24-hour contact amongst the operating teams. A link was established between the view of the working relationship and the amount of physical contact between the people: The higher the level you go the better it gets, at the top of the ladder, they have more meetings face to face with the people they are dealing with rather than we are sort of on the telephone (Chemco operator).

Lack of work socialisation between the three manufacturing units (TCS, Chemco and Wheatco Rubber) was perceived as a key issue. Indeed, within this RI process, poor understanding of the mutual effects between the three plants was a source of frustration: It has always been the perception across the fence that if we have problems we’re actually lying to them about what’s happened and the other way around. (Wheatco).

A socialisation practice was decided: Monday operator meetings between TCS and Chemco operators. The intent was to provide a regular forum for discussing problems in order to ensure a better understanding of each other’s process. Thus, for the first time, operators were allowed to sit down together and share knowledge: The Chemco operator asked a lot of questions and said why do you do this and that. And our guy explained it but also our operators asked the Chemco operator, this gives us problems why do you do this, and they got some explanations and people have actually gone away and started thinking about these things and questioning whether they have to be done that way or whether they can be changed (Wheatco).

A link was established between socialisation and process, although not all informants agreed about the extent to which relations at operator level could impact the actual running of the process: So I don’t think it’ll impact the technical solution at all, but I think it’ll help when the technical solution isn’t in place or not working, it’ll help the relationship between individuals (Chemco) It’s been quite obvious that there are ways they could operate their plant which can help ours and vice versa (Wheatco).

However, barriers to allowing more socialisation at operator level were also referred to: . Resource constraints, such as having to organise a backup when an operator was sent over to the other firm, and the fact that socialisation events were difficult to justify in terms of cost/benefit ratio. Indeed, “the problem is you can do teambuilding exercises together and that sort of thing but it’s very difficult to justify in terms of payback” (Chemco). . Wheatco was a unionised site, which meant that some operators could refuse to deviate from the standard way of working. Thus, in the case of an exchange for a couple of days, there may be “somebody deciding to talk to their union steward and saying what are the implications of me going to work at Chemco”(Wheatco).

The pervasive human resource picture 19

IJOPM 27,1

20

Communication Internal communication was seen as having an effect on the relationship, in that it supported the multifaceted interaction. Such communication took place locally at the level of the Steering Committee, where the Chemco and Wheatco managers in charge of the three manufacturing units sat. Recently, a Steering Committee “charter” was written to provide clear guidance about the local relationship’s purpose and mode of interaction: To set the direction and tone for the Wheatco-Chemco UK strategic partnership, continuous improvement plans, and working interfaces.

Communication at shop floor level was raised as a specific issue in that operators had to be insulated from upsets at management level. And it is important that the managers in both companies whatever the tone of the relationship from a business perspective may be at a particular time just keep it away, don’t let the operators start not trusting each other. But equally, you don’t always hide it. I’ve sat in the control room with the operators saying Chemco are driving me up the wall! And they say ah you too! (Wheatco management).

Discussion Role of HR practices in influencing a RI relationship HR practices appeared to play an important role in operationalising the supply relationship. This role could be positive – as in the case of socialisation – and so encourage “together” behaviour typical of OCR. The role could alternatively be negative – as in the case of rewards and high people turnover – and encourage “separate” behaviour typical of ACR. Within each firm and across all levels, there was an awareness of the effects of internal HR practices on the supply relationship (Table II). Whilst it appeared easy and obvious to collaborate on the technical (or “hard” side) of the relationship, the HR (or “soft” side) was less tangible and did not come to mind so easily. It appeared that some practices that would be obvious in an intra-firm context HR practice

Together

Separate

Staffing

Good calibre of people assigned to the relationship –

High people turnover

Job design Appraisal Rewards Training Socialisation Table II. Summary of HR practices as “together” and “separate”

Communication

Informal assessment and feedback – Joint SPC training and informal processes Formal program of interaction Monday operator meeting helps develop shared knowledge through deep probing Joint leadership communication

Lack of information on the other firm’s job design – Bonus schemes as a source of tension Cancellation of “Team Day” Poor understanding of the other plan. Absence of socialisation associated with estrangement and contributing to opacity of process. Communication at shop floor level

were not considered important within the inter-firm setting. It took a breakdown or quality problem in the process to heighten the profile of HR, typically in conditions when the supply relationship was under stress. On several occasions, informants expressed the idea that the supply relationship would work better if it were part of an internal operation. The relationship would be easier to coordinate because a single team of operators, shift managers and engineers would oversee the whole process. In a single organisation, information flows would be more efficient as well as the associated knowledge of the process. A Chemco manager observed: Shouldn’t Chemco run that rather than Wheatco? That doesn’t mean we need to pick the assets up and move them over to our side of the fence. We need to view this as one site which is owned by different companies. But we need to be able to run it as if it’s one facility. If there were common operators or it was a common control room the plant would be much more reliable.

Some informants established a clear link between HR practices and inter-firm process performance. For example, a link was made between relations at operator level and mutual understanding of the processes and performance. Thus, when operators had good relations or when each understood the other side’s requirements, it was easier to avoid a “trip” (process shutdown). Socialisation as a key practice, across both firms and across levels, is in line with other findings related to an extension of lateral relations across organisational boundaries (Bresnen, 1996, p. 138). However, it may be contrasted with other views regarding the central role of inter-personal relationships focused on a “relationship manager” (Speckman et al., 1998; Marchington and Vincent, 2004) or other managerial position, whilst overlooking the importance of socialisation between other levels. Our study demonstrates the relevance of considering the effects of a broad set of HR practices rather than considering individual ones. The central role of HR practices in shaping supply relationships has been confirmed. However, effects of individual HR practices are also contingent on other HR practices and therefore require an approach to those practices working as a set or “bundle” rather than independently (Delery, 1998). Adaptation of HR practices as a result of the supply relationship HR practices were in general viewed as focused on the priorities of the internal site work organization rather than on the requirements of the supply relationship. Some evidence of adaptation could be identified across the HR practices we studied, more notably at Chemco than at Wheatco. This could partly be explained by a higher dependency on the Chemco side of the relationship, since the Chemco site was devoted to work with Wheatco. Alternatively, it could be explained by the smaller size of the Chemco site, which meant that people were more flexible – perhaps helped by the absence of unionised employee relations – in adapting their HR practices. Table III shows a series of isolated attempts to adapt HR practices to the relationship, offset by more systemic “insulation” behaviour. Overall, our research supports studies which view HR practices as being only loosely connected to the supply relationship (Roper et al., 1997). We help to explain this by analysing the sources and causes of non-adaptation resulting from “insulation” behaviour of the supply partners. There was also some evidence of efforts from both sides to adjust HR practices to the needs of the supply relationship, although these tended to be piecemeal

The pervasive human resource picture 21

IJOPM 27,1

HR practice

Adaptation

Insulation

Staffing

Chemco hires an ex-Wheatco employee to manage the relationship Dedicated roles (Chemco); broader roles (Wheatco-Chemco)

Main selection criteria are internal

Job design

22 Appraisal Rewards

Training Socialisation Table III. Summary of HR practices as “adaptation” and “insulation”

Communication

Feedback from the other firm as input for appraisal (Wheatco and Chemco) Influences job change at Chemco Chemco’s rewards based on joint process performance Some “on the job” training but not part of formal training plan “Monday operator meetings” are set up; quality issues prompt more interaction Some joint communication at management (Steering Committee) level

Chemco fails to influence Wheatco’s decision to implement cross training practice Internal needs drive job design Ultimate decision is internal Wheatco’s rewards based on overall site goals. Wheatco resists Chemco’s attempt to influence reward system No joint rewards Other types of internal training Little interaction at operator level: evolution: more interaction at later stage Site communication is internal

and not pursued systematically over time. In contrast, to other studies that have highlighted the transfer of management practices from one organization to another (Hunter et al., 1996), this research showed little evidence of influence being exerted from one partner to another, whether coercively or through imitation. Several reasons may be advanced for “insulation” behaviour: . In a context of resource scarcity, the relationship was perceived as one priority amongst many. Therefore, the two firms tended to operate at arm’s length with respect to HR practices which they viewed as an internal affair. The priority was to maintain internal consistency rather than external alignment. . HR was viewed as a secondary concern within the relationship. The key focus was on the high technology manufacturing process – actioned by means of QIT and SPC charts and chasing down problems – rather than on human aspects of the inter-organisational relationship. No clear link was recognised by either firm between HR practices and the performance of the technical process. It was therefore difficult to quantify the benefit that would be incurred by cultivating collaborative HR practices, which were perceived as having intangible benefits. Thus, socialisation was referred to in derogatory terms: it (Team day) came down to finances and neither company wanted to pay overtime for everybody to go on a jolly, if you like, so it all fell apart (Wheatco operator). .

HR managers of both firms were not involved in the supply relationship: the only contact between HR personnel was at the level of the local chemical complex, where pay and benefits of industrial staff were broadly aligned between the various units on the site. Otherwise, HR managers focused on their own intra-firm issues.

These behaviours are surprising in view of the high level of interdependence between the firms. They do seem however to be in harmony with Sorenson’s (2003) view that organisational learning is impeded in tightly-coupled organisations. Specificities of RI supply relationships Wheatco-Chemco was an exemplar case of a RI relationship where reciprocity was rooted in the closely linked production processes. Indeed, there was an instantaneous impact of one manufacturing unit over the other, which was due to the lack of buffer stock. This generated a view of the relationship as being “intense” and “intimate”. Informants drew on organic metaphors to depict this interdependence: “symbiotic link” “umbilical cord”. Indeed, the physical layout of Chemco’s plant, which was built along the Wheatco fence, induced a view of a single plant, with the fence as an artificial barrier. Technical failures led to tension, and HR practices were perceived as contributing either to worsening or strengthening the relationship. The intensity of interaction was also a result of repeated contacts at multiple levels. This highlighted the influence of HR on the nature of the relationship at all layers of interaction. In view of the strong operational link, the influence of HR practices appeared as particularly sensitive at operator level. This supports Marchington and Vincent’s (2004) view that there are differences in the perspective of the relationship, depending on the level (here operator vs management levels). Thus, the role of HR was contingent on the above specificities of the relationship. Had the relationship been less intimate, less pressured and with fewer interaction levels, the influence of HR practices would likely have been less prominent. Conclusion This research builds on earlier work on HR management within supply relationships (Hunter et al., 1996; Scarbrough, 2000) by identifying and analysing the effects of HR practices on the relationship and how far these practices are adapted. We illustrate the importance of investigating HR practices across a broad front, and of engaging with the complex and messy detail of supply relationships in practice. We also show the significance of researching supply relationships at different levels: the operational level exhibits quite different characteristics from the more widely-documented managerial levels. Our exploratory case reveals the importance of including HR issues on the agenda of supplier relationship management, in particular in a RI context. We highlight the constant state of flux of such relationships and the pervasive influence of HR practices in shaping the interactions. What we have labelled “insulation” behaviour with respect to HR practices was grounded in an internal logic and coherence that tended to ignore external requirements. However, we found that informal processes emerged to help compensate for the lack of flexibility of formal HR processes. We have presented findings from a single case study: this obviates the possibility of comparative data. The context is also dependent on the “multinational corporation” context of the two companies involved. Further research is needed to provide a stronger theoretical grounding for our conclusions of the effects of HR practices on the RI supply relationship – possibly using an institutional theory perspective. Further empirical work is also needed to investigate this area of OM that is problematic to research. We are conducting further research to apply a similar research methodology

The pervasive human resource picture 23

IJOPM 27,1

24

to another case involving RI supply relationships in order to address the issue of comparative data. Our study addresses the call for more empirical research on supply relationships from an OM perspective (Ahmad and Schroeder, 2003), and supports the pervasive role of HR in this context. In addition, to other findings, which highlight the role of inter-personal relations within supply relationships, our study provides insights into the role of HR practices across different levels of a supply relationship. Our study contributes to better understanding of the role of HR practices in the OM context of supply relationships: more cross-disciplinary research between OM and HR is called for. References Adler, P. and Borys, B. (1996), “Two types of bureaucracy: enabling and coercive”, Administrative Science Quarterly, Vol. 41 No. 1, pp. 61-89. Ahmad, S. and Schroeder, R. (2003), “The impact of human resource management practices on operational performance: recognising country and industry differences”, Journal of Operations Management, Vol. 21, pp. 19-43. Araujo, L., Dubois, A. and Gadde, L. (2003), “The multiple boundaries of the firm”, Journal of Management Studies, Vol. 40 No. 5, pp. 1255-77. Becker, B.E. and Huselid, M.A. (1998), “Overview: strategic HRM in five leading firms”, Human Resource Management, Vol. 38 No. 4, pp. 287-305. Bendoly, B., Bachrach, D., Wang, H. and Zhang, S. (2006), “ERP in the minds of supervisors”, International Journal of Operations & Production Management, Vol. 26 No. 5, pp. 558-78. Borys, B. and Jemison, D. (1989), “Hybrid arrangements as strategic alliances: theoretical issues in organizational combinations”, Academy of Management Review, Vol. 14 No. 2, pp. 234-49. Brannen, M.Y. (2004), “When Mickey loses face: recontextualisation, semantic fit and the semiotics of foreignness”, Academy of Management Review, Vol. 29 No. 4, pp. 593-616. Bresnen, M. (1996), “An organizational perspective on changing buyer-supplier relations: a critical review of the evidence”, Organization, Vol. 3 No. 1, pp. 121-46. Cascio, W.F. and Serapio, M.G. (1991), “Human resource systems in an international allkiance: the undoing of a done deal”, Organizational Dynamics, Vol. 19 No. 3, pp. 63-74. Delery, J.E. (1998), “Issues of fit in strategic human resource management: implications for research”, Human Resource Management Review, Vol. 8 No. 3, pp. 289-309. Eisenhardt, K.M. (1991), “Better stories and better constructs: the case for rigor and comparative logic”, Academy of Management Review, Vol. 16 No. 3, pp. 620-7. Furlan, A., Romano, P. and Camuffo, A. (2006), “Customer-supplier integration forms in the air-conditioning industry”, Journal of Manufacturing Technology Management, Vol. 17 No. 5, pp. 633-55. Fynes, B. and Voss, C. (2002), “The moderating effect of buyer-supplier relationships on quality practices and performance”, International Journal of Operations & Production Management, Vol. 25 No. 1, pp. 6-19. Gadde, L-K. and Snehota, I. (2000), “Making the most of supplier relationships”, Industrial Marketing Management, Vol. 29 No. 4, pp. 305-17. Gro˝ssler, A., Gru˝bner, A. and Milling, P. (2006), “Organisational adaptation processes to external complexity”, International Journal of Operations & Production Management, Vol. 26 No. 3, pp. 254-81.

Guest, D., Michie, J., Conway, N. and Sheehan, M. (2003), “Human resource management and corporate performance in the UK”, British Journal of Industrial Relations, Vol. 41 No. 2, pp. 291-314. Halle´n, L., Johanson, J. and Seyed-Mohamed, N. (1991), “Interfirm adaptation in business relationships”, Journal of Marketing, Vol. 55 No. 2, pp. 29-37. Haokansson, H. and Snehota, I. (1995), Developing Relationships in Business Networks, Routledge, London. Harrison, A.S. and Storey, J. (1996), “New wave manufacturing strategies: operational, organizational and human dimensions”, International Journal of Operations & Production Management, Vol. 16 No. 2, pp. 63-76. Hayes, R.H. and Wheelwright, S.C. (1984), Restoring our Competitive Edge – Competing through Manufacturing, Wiley, New York, NY. Hunter, L., Beaumont, P. and Sinclair, D. (1996), “A ‘partnership’ route to human resource management?”, Journal of Management Studies, Vol. 33 No. 2, pp. 235-57. Jick, T.D. (1979), “Customer-supplier partnerships: human resources as bridge-builders”, Human Resource Management, Vol. 29 No. 4, pp. 435-54. Kanter, R.M. (1994), “Collaborative advantage: the art of alliances”, Harvard Business Review, Vol. 74 No. 4, pp. 96-108. Lei, D., Slocum, J.W. and Pitts, R. (1997), “Building co-operative advantage: managing strategic alliances to promote organizational learning”, Journal of World Business, Vol. 32 No. 3, pp. 203-23. Lepak, D. and Snell, S. (1999), “The human resource architecture: toward a theory of human capital allocation and development”, The Academy of Management Review, Vol. 24 No. 1, pp. 31-48. Lorange, P. (1986), “Human resource management in multinational co-operative ventures”, Human Resource Management, Vol. 25 No. 1, pp. 133-48. McDuffie, J.P. (1995), “Human resource bundles and manufacturing performance: organisational logic and flexible production systems in the world auto industry”, Industrial and Labour Relations Review, Vol. 48 No. 2, pp. 197-213. Mair, A. (1998), “Internationalisation at Honda: transfer and adaptation of management systems”, Employee Relations, Vol. 20 No. 3, pp. 285-302. Marchington, M. and Vincent, S. (2004), “Analysing the influence of institutional, organizational and interpersonal forces in shaping inter-organizational relations”, Journal of Management Studies, Vol. 41 No. 6, pp. 1029-56. Meredith, J. and Samson, D. (2002), “Introduction to the special issue: case study and field research editorial”, Journal of Operations Management, Vol. 20 No. 5, pp. 415-7. Pettigrew, A. (1990), “Longitudinal field research on change: theory and practice”, Organization Science, Vol. 1 No. 3, pp. 267-92. Pitman, M. and Maxwell, J. (1992), “Qualitative approaches to evaluation: models and methods”, in LeCompte, M., Millroy, W. and Preissle, J. (Eds), The Handbook of Qualitative Research in Education, Academic Press, New York, NY, pp. 729-70. Pucik, V. (1988), “Strategic alliances, organizational learning and competitive advantage: the HRM agenda”, Human Resource Management, Vol. 27 No. 1, pp. 77-93. Richardson, G.B. (1972), “The organization of industry”, The Economic Journal, Vol. 82, pp. 883-96.

The pervasive human resource picture 25

IJOPM 27,1

26

Ring, P.S. and Van de Ven, A.H. (1992), “Structuring cooperative relationships between organizations”, Strategic Management Journal, Vol. 13 No. 7, pp. 483-98. Roper, I., Prabhu, V. and Van Zwanenberg, N. (1997), “(Only) just-in-time: Japanisation and the ‘non-learning’ firm”, Work, Employment & Society, Vol. 11 No. 1, pp. 27-46. Rubery, J., Carroll, M., Cooke, F., Grugulis, I. and Earnshaw, J. (2004), “Human resource management and the permeable organization: the case of the multi-client call center”, Journal of Management Studies, Vol. 41 No. 7, pp. 1199-12222. Sako, M. (1992), Prices, Quality and Trust: Inter-Firm Relations in Britain and Japan, Cambridge University Press, Cambridge. Scarbrough, H. (2000), “The HR implications of supply chain relationships”, Human Resource Management Journal, Vol. 10 No. 1, pp. 5-17. Sorenson, O. (2003), “Interdependence and adaptability: organisational learning and the long-term effects of integration”, Management Science, Vol. 49 No. 4, pp. 446-63. Speckman, R.E., Kamauff, J.W. and Myhr, N. (1998), “An empirical investigation into supply chain management”, International Journal of Physical Distribution & Logistics Management, Vol. 28 No. 8, pp. 630-50. Storey, J., Emberson, C. and Reade, D. (2005), “The barriers to customer responsive supply chain management”, International Journal of Operations & Production Management, Vol. 25 No. 3, pp. 242-60. Swart, J. and Kinnie, N. (2003), “Knowledge-intensive firms: the influence of the client on HR systems”, Human Resource Management Journal, Vol. 3 No. 3, pp. 37-55. Thompson, J.D. (1967), Organizations in Action, McGraw-Hill, New York, NY. Voss, C., Tsikritiktsis, N. and Frohlich, M. (2002), “Case research in operations management”, International Journal of Operations & Production Management, Vol. 22 No. 2, pp. 195-219. Wilson, D.T. (1995), “An integrated model of buyer-seller relationships”, Journal of the Academy of Marketing Sciences, Vol. 23 No. 4, pp. 335-45. Yin, R.K. (2003), Case Study Research: Design and Methods, 3rd ed., Sage, Thousand Oaks, CA. Further reading Dwyer, F., Schurr, P. and Sejo, O. (1987), “Developing buyer-seller relationships”, Journal of Marketing, Vol. 51 No. 2, pp. 11-27. Appendix. Interview guide This interview guide was the result of several iterations. The reason for starting with broad, open-ended questions (questions 1-3) was to encourage the informants to discuss the supply relationship as much as possible without being influenced by the researcher. Questions on the specific HR practices were asked later (question 4). The interview generally started with a brief presentation of the research project, credentials of the researcher and with a reassurance that the anonymity of the informant would be respected. Then the interview proceeded along following lines: (1) What is your position and role? What are your activities and length in the job, what is your reporting line? The aim of this question was to allow the informant to briefly introduce themselves and their position in the organization. (2) How do you see your role in the supply relationship and how do you interface with the partner?

The aim of this question was to locate the informant within the supply relationship and to understand the extent of their interaction with members of the partner organization (Who do you interact with? How often? Why? How?). (3) How would you describe the relationship? (Follow up questions: in what ways is it a partnership vs a non-partnership relationship? What are the elements that reinforce or hinder the relationship? Provide specific examples of key events or issues. How have your seen the relationship evolve over time?) The aim of this set of open-ended question was to allow the informant to describe the relationship in their own words, as much as possible without being influenced by the interviewer. The intent was to allow free emergence of various themes about the relationship without constraint from the framework for analysis. (4) What are the requirements of the supply relationship in terms of HR? In what ways can HR influence the relationship. In what ways can HR practices facilitate or hinder the relationship? Within CH, within WTC? within the hybrid? In what ways does the partnership create different requirements for PM. (5) In what ways do you see an adaptation of internal HR practices to the requirements of the relationship? Here each HR practice was used as a prompt: staffing, job design, appraisal, rewards, training, socialisation, communication. Is there anything else that you would like to add?

Corresponding author Marie Koulikoff-Souviron can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

The pervasive human resource picture 27

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0144-3577.htm

IJOPM 27,1

Greening the automotive supply chain: a relationship perspective Dayna Simpson, Damien Power and Daniel Samson

28

Department of Management, University of Melbourne, Australia Abstract Purpose – This study seeks to explore the moderating impact of relationship conditions existing between a customer and its suppliers on the uptake and effectiveness of the customer’s environmental performance requirements (otherwise known as “green-supply”). Design/methodology/approach – The study assesses the extent to which a supplier’s environmental performance is influenced by its customer’s environmental performance requirements when specific relationship conditions (investment, contracting and monitoring routines) are taken into account. Data were collected through a survey of first and second tier component manufacturers in the Australian automotive industry and analysed using linear regression and MMR. Findings – Suppliers were found to be more responsive to their customers’ environmental performance requirements where increasing levels of relationship-specific investment occurred. As the level of investment in the customer-supplier relationship increased, suppliers become less likely to believe that they would be penalized for non-compliance with the customer’s environmental performance requirements. Research limitations/implications – Survey data were collected in 2004 and are limited to the Australian automotive industry. The sample size available for the regression analysis also precluded the use of more comprehensive analytic techniques. Practical implications – The research offers new insight into the issue of how firms might improve the environmental performance of suppliers and the sustainability of their supply chain. Originality/value – Virtually no research exists on the actual effectiveness of green supply requirements when placed in context with the realities of inter-organizational dynamics. The findings suggest that traditional operations theory on inter-organizational performance improvement is just as relevant to the use of environmental performance requirements. Keywords Environmental management, Supplier relations, Automotive industry, Green marketing Paper type Research paper

Introduction Organizations have become increasingly aware of the propensity for environmental pollution incidents within their supply network to cost them in penalties, cleanup and consumer backlash. As a result, minimum standards of environmental performance have become increasingly prevalent in the purchasing contracts or guidelines of multinational corporations for their local and global suppliers (Bowen et al., 2001a; Zhu and Sarkis, 2004). This relatively new expectation for upstream suppliers goes beyond the more traditional requirements of their customers to reduce costs and improve quality and service (Lambert and Cooper, 2000). Prominent examples of such activity include Starbuck’s Responsible Sourcing Guidelines (Starbucks, 2005), Ford Motor International Journal of Operations & Production Management Vol. 27 No. 1, 2007 pp. 28-48 q Emerald Group Publishing Limited 0144-3577 DOI 10.1108/01443570710714529

The authors would like to acknowledge the support and involvement of the Toyota Motor Corporation Australia, the case study suppliers, the Victorian Department of Innovation, Industry and Regional Development, the Federation of Automotive Parts Suppliers and the Tooling Industry Forum of Australia.

Company’s requirement for all suppliers to certify to the ISO14001 management standard and the inclusion of “supplier activities” in statements of environmental responsibility for Toyota, BMW and Mitsubishi (Young and Kielkiewicz-Young, 2001). This is matched by recent academic research which supports the potential of customer-supplier relationships to influence the environmental performance of manufacturing supply chains (Klassen and Vachon, 2003; Zhu and Sarkis, 2004; Rao and Holt, 2005). At a time when global manufacturing industries face significant constraints on the availability of natural resources and multiple threats to survival, the imposition of environmental performance requirements represents a new and complex pressure for the organization to manage. The practice of extending production goals from customers to their suppliers as a means to improve overall performance in a supply chain has been a growing field of research for the past 15 years (Lamming, 1993; Krause et al., 2000; Liker and Choi, 2004). Organizations have used a range of supplier-relationship management styles to improve production processes or introduce new technologies into the supply chain, such as purchasing power (i.e. Walmart) and/or collaboration (i.e. Toyota) (Dyer and Chu, 2003). Both modes of interaction have been successful in the past for achieving more rapid and often inimitable process or product-based improvements. A small but growing body of research has more recently explored the influence of a customer’s relationship with its suppliers in regard to the extension of sustainability-based goals. This process, known broadly as “green-supply,” is a potentially effective mechanism for supply chain managers to improve the organization’s record on corporate social responsibility, minimize reputational risks, reduce wastes and increase flexibility in response to new environmental regulations (Green et al., 1998; Bowen et al., 2001a; Melnyk et al., 2003). Most examples have included requirements to meet minimum standards of environmental practice, certification to an international management standard (i.e. ISO14001) or a general philosophy of supply chain stewardship. The customer – as a major financial stakeholder – has significant potential to force improvements to its suppliers’ environmental management practices, introduce environmentally sound technologies, and collaborate with suppliers to share knowledge and jointly develop more sustainable products and processes. From the customer’s perspective this may require a more hierarchical approach to the issue of supplier greening – that is, expecting that some suppliers will be more or less responsive than others. From a supplier’s perspective this may present both advantages and difficulties in their attempts to meet a new and possibly under-developed set of environmental performance requirements. From a government perspective this may require a more collaborative approach to working with organizations as the challenge to meet the goals of global sustainability increases. Much research into external pressures on the organization’s derivation of environmental responsibilities is considered from the perspective of public or institutional stakeholders or the consumer. Only a small but growing body of research explores the role of the major customer in a supply chain or procurement context. Even less research considers the implications of the nature of the customer-supplier relationship on the uptake and effectiveness of these environmentally-relevant supply requirements. That is – whether the presence of specific relationship conditions or management styles between the customer and the supplier (i.e. purchasing power, governance mechanisms or collaboration) might moderate the influence of these types of requirements.

Greening the automotive supply chain 29

IJOPM 27,1

30

This research seeks to extend the small but growing body of knowledge surrounding the introduction of environmental performance requirements into supply contracts or statements of supply chain stewardship. The major contribution of this research will be not just to explore the existence and application of such requirements but to explore the influence of specific exchange conditions on their uptake and effectiveness. In particular, the transaction cost framework after Williamson (1985) is employed to describe these exchange conditions. The a-priori expectation of this research is that supply relationship conditions that exhibit traits of high investment and governance will provide for a more effective deployment of a customer’s environment-related performance requirements. Such conditions are proposed to moderate the effectiveness of any program of green supply between the customer and the supplier by increasing its efficacy as the level of investment and governance increases. Literature review Global warming, reductions in air quality, pollution of waterways and widespread loss of biodiversity are but a few examples of the types of environmental impact that can be attributed to the coordinated activity of organizations in a supply chain. Much of this arises from manufacturing organizations that continue to produce large amounts of unnecessary waste or emissions rather than investing in better technologies or practices to prevent its generation at the source (Klassen, 2000; King and Lenox, 2002). The use of more environmentally sustainable products or production processes are often prevented from developing within an organization because of the external pressure applied by customers to achieve the requirements of reduced cost, increasing quality and faster delivery. Production of unnecessary waste or choices toward lower-cost but environmentally unsustainable production alternatives are frequently the responses to such supply chain pressures (Green et al., 1998). In the practical realm, the inclusion of environmental performance standards in supply requirements is marching ahead with only limited theory on the managerial implications of this type of inter-organisational action. An opportunity exists for the development of new theories of inter-organisational exchange using theories of the organisation and supply chain management that would substantially inform this new type of practical activity. Customer-supplier interactions and environmental performance Programs developed by business to “green” supplier activities or include environmental performance requirements in supply guidelines are increasingly evident in practice. Such initiatives are broadly referred to as either green-supply or green-supply-chain in both the academic and practitioner literature. These have largely included activities with suppliers such as: . programs to reduce or eliminate materials used in manufacturing processes or products; . programs focused on the environmental compliance status and practices of supplier operations; . joint development of new materials, processes or other solutions to environmental issues (Sarkis, 2003; Green et al., 2000).

A recent body of academic research generates support for the theory that the customer-supplier or supplier-supplier relationship may generate a range of positive environmental outcomes (Klassen and Vachon, 2003; Zhu and Sarkis, 2004; Rao and Holt, 2005). Communicating goals of sustainability or environmental performance through the supply relationship has resulted in for example, collaborative waste reduction, environmentally sound innovation, cost-effective and environmentally beneficial solutions to production problems, and more rapid development and uptake of environmental technologies. There is growing empirical support for the role of the supply relationship in environmental performance management that extends into the inter-firm setting. In the antecedent work of Lamming and Hampson (1996), customer firms engaged in collaborative dialogue with suppliers were better able to understand the environmental impacts of their supply chains. Florida (1996) looked to customer-supplier relationships that were already characterized by improvement or learning activities: . . . environmental improvements flow from ongoing joint efforts to improve productivity, eliminate defects and reduce costs, rather than from direct offers to transfer pollution prevention technology or organizational strategies designed expressly to eliminate toxins or prevent pollution (Florida, 1996, p. 81).

Hall (2000, 2001) extended Florida’s (1996) work by finding that a collaborative customer-supplier relationship often led to environmental performance improvements in both the customer and the supplier firm. Geffen and Rothenberg (2000) found the involvement of suppliers to be critical in the development and implementation of environmentally sound technologies in automotive paint production. More recently, Klassen and Vachon (2003) investigated the role of supply-chain-level evaluation and collaboration activities on plant level environmental investment. They found that greater customer involvement and scrutiny of suppliers tended to: “. . . capture the attention of plant managers and encouraged greater environmental investment” (Klassen and Vachon, 2003, p. 347). The commonality across this body of research provides support for the hypothesis that customers may be able to directly and indirectly improve a supplier’s environmental performance. Stakeholder management and environmental performance Recognition and development of an organizational requirement to manage its impact on the environment has developed largely in response to increasing levels of government regulation and profile-raising of environmental issues by non-financial stakeholders (politics, community and media) (Aragon-Correa, 1998; Hoffman, 1999; Bansal, 2005). An alternative perspective to that of the organization choosing to adopt a proactive or socially responsible strategy and acting in advance of its stakeholders is to take the perspective of Friedman (1970) and Egri and Pinfield (1996) that the organization’s choice of environmental strategy has more to do with the needs of the organization’s financial stakeholders. This somewhat un-romantic view of an organization’s social and environmental responsibilities, takes the less popular position that organizations exist purely for the sake of economic self-interest rather than to fulfill a set of social or altruistic ideals (Smith, 1776; Friedman, 1970). As Egri and Pinfield (1996, p. 472) describe:

Greening the automotive supply chain 31

IJOPM 27,1

32

From the perspective of organizational theory, environmental degradation becomes relevant only when the performance of a focal organization and the welfare of organizational participants are affected by such concerns.

For the organization to internalize a social or environmental responsibility it is proposed that an initial condition of direct impact to, or influence on, the ecology of its field or system must first occur. Antecedent theory proposes that the organization’s response to the environmental requirements of its external stakeholders will be influenced by its existing level of environmental commitment (Hunt and Auster, 1990; Henriques and Sadorsky, 1999; Aragon-Correa and Sharma, 2003). At the same time and within the context of this research, the magnitude of this environmental commitment can be significantly affected by the existence and persistence of an externally derived set of environmental performance requirements. The form and magnitude of this influence however should depend on which stakeholder claims relevance for its environment-related requirement, and the power and level of involvement of that stakeholder with respect to the receiving organization (Mitchell et al., 1997; Buysse and Verbeke, 2003). Customer-supplier relationships Inter-firm relationships are critical to the successful coordination of supply chains and improvements in the performance of suppliers’ production capabilities (Lamming, 1996; Handfield et al., 2000; Scannell et al., 2000). The supply relationship is an important channel for communicating customer requirements to suppliers and achieving longer term goals of production (Lamming, 1996; Handfield et al., 2000; Scannell et al., 2000). Managed supply chain relationships can often attain the types of performance improvement and superior competitive advantage that are not readily generated by open market transactions (Lamming, 1993; Burt and Doyle, 1993; Dyer and Nobeoka, 2000). In considering attempts to “green” suppliers as an issue of performance management in the supply chain – where a customer has a minimum performance requirement or desires an improvement in performance from its suppliers – such environmental performance goals may be influenced by the same factors which influence other supply chain level performance elements (i.e. quality, cost and lead time reductions). The importance of relationship management styles or the “conditions” of the customer-supplier relationship have been increasingly described as important factors in the process of supplier or supply chain performance improvement (Cousins and Stanwix, 2001; Handfield and Bechtel, 2002; Dyer and Chu, 2003). All supply relationships tend to share the same basic principle of conduct – suppliers are required to meet performance or supply targets specified by a customer. Where the purchasing side of the supply partnership (customer) intends to extract a performance gain in a process, product or service being managed by its suppliers, established and emergent theory suggests a variety of mechanisms or conditions to achieve this end (Cousins and Stanwix, 2001; Handfield and Bechtel, 2002; Dyer and Chu, 2003). These mechanisms or “conditions” that occur between a customer and its suppliers provide a number of critical features – structure for the interaction and remedies for failure or non-compliance; a mechanism for power or influence over the supplier; and a climate that promotes either adversarial behavior or collaboration (Ring and Van de Ven, 1992). The literature remains unclear as to which mechanism or

condition will produce the best results for a customer that desires to improve an element of a supplier’s operations performance. For example, the competitive force of the open market has been shown by some authors to be a successful mechanism to improve performance amongst a pool of suppliers (Scannell et al., 2000). Other authors promote the importance of a highly socialized and involved form of relationship for generating more complicated types of performance gains (Dyer and Nobeoka, 2000; Dyer and Chu, 2003). An important consideration for the customer and supplier amongst these various relationship choices is the transaction-cost framework first described by Williamson (1975). Transaction-cost theory is useful for describing the coordination costs and transaction risks of inter-organizational activities. Transaction cost theory in its original explanatory frame, is primarily concerned with its central claim which is the handling of transactions in a manner which minimizes the associated costs (Grover and Malhotra, 2003; Williamson, 1975; 1985). An associated body of literature has used the transaction cost theoretical frame to explore its implications for inter-organizational performance improvement. The major tenet of this body of work has been the proposition that as transaction costs increase and behavioral uncertainty decreases, the mutual benefits of inter-organizational performance improvement will also increase (Rindfleisch and Heide, 1997; Grover and Malhotra, 2003; David and Han, 2004). Development of hypotheses “Environmental commitment” is defined in this research as the willingness of an organization to determine, articulate and manage it responsibilities toward the natural environment. An underlying assumption of this study has been that an organization’s overall level of environmental commitment will provide positive benefits to the organization’s environmental performance. The relationship between environmental commitment and use by the organization of a set of environmentally conscious practices has previously been explored by others. These prior works have found increasing levels of environmental commitment and greater occurrence within the organization of for example, pollution prevention and incorporation of the environment in product design and innovative practices (Buysse and Verbeke, 2003); being more inclusive of environment-related stakeholders (Aragon-Correa, 1998); incorporating practices that support pollution reduction (Sharma and Vredenburg, 1998; Klassen, 2001); more likely to use environmentally conscious manufacturing practices (Bowen et al., 2001a, b); and have an increasing intention to preserve ecological integrity (Judge and Elenkov, 2005). Assuming, from the work of previous authors, that higher levels of environmental commitment will lead to a greater likelihood of improving environmental performance – for example through pollution reduction or environmentally conscious business practices – this study will assess the impact of an externally derived set of environmental performance requirements on the organization’s level of environmental commitment. During preliminary discussions with six automotive component suppliers as part of this study, organizations were asked where their primary motivation to meet formal or contractual environmental performance requirements came from, such as certification to ISO14001. One response in particular illustrated the majority perspective of those interviewed: I’d say that we had to have it, because of the customer saying you have to have it if you want to supply to us. That was pretty much the directive from Ford, Holden and Mitsubishi.

Greening the automotive supply chain 33

IJOPM 27,1

34

It’s mandatory. Our social responsibility is always there but with these sorts of things unless you have to have them . . . because it does cost us a lot of money (Operations Manager, 2004).

Primary stakeholders – such as the organization’s major customer – are able to influence the strategic choices of the organization (Miles and Snow, 1978; Pfeffer and Salancik, 1978; Mitchell et al., 1997). In this study, the interest has been with the most significant stakeholder for the organization’s survival – its major customer – because of the inherent power of this stakeholder in the context of the organization’s supply relationships. Already established in prior research is that higher levels of environmental commitment is likely to generate higher levels of environmental performance in the organization. The study’s first proposition is that the customer’s environmental performance requirements will bear a relationship to the supplier’s level of environmental commitment: H1. The supplier’s level of environmental commitment is related to the environmental performance requirements of its major customer. To integrate the findings of this study with the work of previous authors, the organization’s level of environmental commitment is assessed in relation to the environmental performance requirements of its major customers. This initial relationship is then assessed in the presence of a series of customer “conditions” that act to inform the organization’s response to these environmental performance requirements. Influence of the customer-supplier relationship Where the influence of one firm over another is required with an aim to improve or ensure a process, product or service, established and emergent theory offers varying explanations for the most important factors for customers to consider when attempting to extract performance gains from suppliers. Much of this literature relates to the structure of the inter-firm relationship and the desired outcome of any improvement initiative. As described in the literature review, all supply relationships tend to share the same basic principle of conduct – suppliers attempt to achieve, to the best of their ability and willingness – the specific performance targets required by the customer. Customers often choose to use a range of measures to “encourage” their suppliers to meet such performance requirements. The choice and success of these measures of “encouragement” are often referred to as relationship “conditions” that determine its power or influence over the supplier and its choices. It is a major proposition of this study that suppliers will not only be influenced by the environmental performance requirements presented to them by their customers, but also importantly the magnitude of this influence will depend on the existing conditions of the supply relationship. These conditions in the relationship will moderate the impact that the customer’s environmental performance requirements has on the environmental commitment of its suppliers. This moderating effect is described through the study’s next three hypotheses. Relationship-specific investment. Relationship conditions that promote relationship-specific investment between the customer and the supplier – particularly in regard to the supplier having dedicated time, future goals, equipment or capacity to the relationship – potentially allow the supplier to be more “aware” of the customer’s environmental reputation and value system. Higher levels of asset specificity between the customer and supplier have been found by other authors to lead to positive

performance gains for both firms involved in the exchange (Dyer, 1997; Handfield and Bechtel, 2002). A remark made by an operations manager during preliminary interviews conducted as part of this study highlights the potential importance of relationship-specific investment to goals of environmental performance: .

We’re fairly close to [our major customer ] so if they take a strong interest in a particular issue it obviously affects us because we’re that closely linked to them we’ll have a look at what we’re doing. A lot of people don’t really know who the hell we are but they know who [our major customer ] is and if [our major customer ] had a big environmental impact somewhere around the world, it could have a big impact on their sales. If we had a major environmental incident, it would certainly affect our customer’s opinion of us . . . ” (Operations Manager, 2004).

Investment as a moderating variable has not yet however been investigated in regard to environmental performance improvement. Asset specificity is the most widely used measure for measuring the role of transaction cost theory in inter-organizational relationships (David and Han, 2004) and it has been used again in this study. In this study, the role of relationship-specific investment in the uptake of the customer’s environmental performance requirements is described by the next hypothesis: H2a. The presence of relationship-specific investments will improve supplier response to the customer’s environmental performance requirements. Governance or contracts. Contracts support asset-specific investments in that they provide a form of governance between the customer and the supplier that not only protects assets sunk into the relationship, but also identifies paths for dispute resolution and outcomes for non-compliance (Williamson, 1975). Contracts are effectively a safeguard against opportunistic behaviour and set clear boundaries for default on contractual specifications between the customer and the supplier. At the same time that contracts provide a mechanism of governance they also increase the ex-ante costs for the relationship and in more recent literature they are described as increasing the adversarial climate in the relationship and reducing trust (Ghoshal and Moran, 1996; Liker and Choi, 2004). In this study, the role of contracts in the uptake of the customer’s environmental performance requirements is described by the next hypothesis: H2b. Contracts signed between the customer and the supplier will improve supplier response to the customer’s environmental performance requirements. Monitoring or assessment routines. Assessment or monitoring performs two important roles in the supply relationship – that of a monitoring mechanism to safeguard asset-specific investments and also a mechanism to reduce information asymmetry (Williamson, 1985; Stump and Heide, 1996). In more recent research that described the role of customer evaluation in the context of environmental performance improvement, Klassen and Vachon (2003, p. 347) found evaluation activities to have a positive impact on the level of investment in environmental management by suppliers: More specifically, customers who implement supplier audits, establish formal evaluation processes, and offer feedback to suppliers are likely to see positive changes in how suppliers regard environmental issues. The greater direct scrutiny by customers likely captured the attention of suppliers’ plant managers and encouraged greater environmental investment.

Greening the automotive supply chain 35

IJOPM 27,1

36

As described by one operations manager during discussions with component suppliers as part of this study, assessment routines allowed their organization to better comprehend their own environmental performance and what their customers required of them: [Our major customer ] made us have a good look at what we’re doing in terms of how does it impact the environment and what are some of the environmental benefits of what we do. Their awards process was about showcasing all of the things that you’re doing – and some of it was going on but none of us realized so it wasn’t shared and communicated. They actually had the framework to ask us “well what are you actually doing for sustainability and green innovations” and it made us realize we’re actually doing a lot. With the other car companies it’s just a matter of compliance and ticking the right boxes (Operations Manager, 2004).

The importance of relationship conditions that include the customer using an Assessment routine – such as a customer-derived list of specific performance criteria for suppliers to follow – is still a highly under-explored factor. The possible importance of assessment to the understanding or evaluation of a customer’s environmental performance requirements is extended in this study through the final hypothesis: H2c. Use by the customer of supplier assessment will improve supplier response to the customer’s environmental performance requirements.

Conceptual model In extending the body of research devoted to green-supply, this research explores specifically the role of the transaction relationship between a customer and its suppliers and its influence on the communication, uptake and delivery of supply chain level environmental performance requirements. It seeks to explore both the realm in which environmental performance requirements are to be effectively met – the organization – and the mechanism through which they are exchanged – the customer-supplier relationship. At a strategic level, a supplier’s environmental commitment will be influenced by the environmental performance requirements of the stakeholder of interest – the major customer. An increasing environmental commitment in the supplier organization is proposed to also have a positive relationship to increasing environmental performance. Assuming from the work of previous authors that higher levels of environmental commitment will lead to a greater likelihood of improving environmental performance – for example through pollution reduction or environmentally beneficial operational practices – this study will assess the impact of an externally derived set of environmental requirements on the organization’s level of environmental commitment. This model is shown in Figure 1. The model suggests that under conditions of an existing supply relationship there will be a primary relationship between a customer’s environmental performance requirements for its suppliers and the supplier’s own commitment to its environmental responsibilities. This relationship will then be moderated by conditions of the customer-supplier relationship. The model is tested using the Australian automotive industry as its case.

CUSTOMER Environmental Performance Requirements

H1 SUPPLIER Environmental Commitment

MODERATORS

Greening the automotive supply chain 37

Customer-Supplier Relationship Conditions Investment (H2a) Contracts (H2b) Assessment (H2c)

The Australian automotive industry The global automotive industry provides a unique case for exploring the influence of supply relationships on environmentally-relevant performance indicators because of the range of approaches to relationship management evident in practice. It is also one of the few global industries that contains numerous examples of customers requiring suppliers to meet minimum standards of environmental performance. Four major motor vehicle assemblers operate manufacturing and assembly hubs in Australia – Ford, Toyota, Holden and Mitsubishi. The Australian automotive industry produces large passenger motor vehicles, light commercial and sports utility vehicles (ABS, 2005). The industry has over 200 individual component manufacturers and around 500 smaller firms providing tooling services and other firms that provide specialist automobile services. The Australian industry supplies mainly to the international markets of South Korea, USA, Canada, Japan and China. Australian-based industry makes up 65 percent of the inputs to domestic automotive production. The remaining 35 percent of inputs are mainly imported parts and components sourced from Japan and the USA. The sector produces around 5 percent of the global motor vehicle market. In 2002-2003 imports of vehicles were $14bn and imports of components and parts were nearly $6bn (ABS, 2005). Australia has become a truly globalised production market through de-regulation of the industry, a stable economy and its position and political/economic relationships with the growing automotive markets of the Asia-Pacific. Methodology Sample and data collection Following Snow and Thomas (1994) and Wacker (1998), this study used a mixed method approach to the development of new theory, to better develop the study’s hypotheses and ground a set of constructs for empirical testing. An initial study was undertaken involving preliminary discussions with operations managers and purchasing managers at six first tier automotive component supplier firms in the Australian manufacturing base. These discussions involved semi-structured interviews and shop-floor tours (after Yin, 2003). This was followed by an industry

Figure 1. Conceptual model

IJOPM 27,1

38

survey mailed out to an up-to-date mailing list of first and second tier suppliers in the Australian automotive industry. Survey items and appropriate measurement scales were developed after DeVellis (2003). Scales and items used in the questionnaire, analysis techniques and final results are described in the following sections. The final method of survey delivery and collection was selected as suggested by Dillman (1999). The research sample was sourced from two known industry databases with up-to-date memberships. Database 1 contained all first and second tier component suppliers in the Australian automotive industry (200 contacts) obtained through the Federation of Automotive Products Manufacture. Database 2 contained all tooling firm suppliers in the Australian automotive industry (200 contacts) obtained through the Tooling Industry Forum of Australia. The instrument was distributed by both mail and e-mail. Of the 400 surveys distributed, 56 usable surveys were returned for a response rate of 15 percent. Other research into green supply involving surveys have used a sample size of similar proportion to this research. Rao (2002) in particular produced significant findings of green supply with a sample of only 52 firms and Bowen et al. (2001a) used a sample of only 24 firms (with two sets of respondents within each firm). To assess the possibility of differences between early respondents, late respondents and non-respondents, a comparison of the demographics of the survey respondents was made to the demographics of the larger population (after Armstrong and Overton, 1977). In Table I a comparison is made of the key demographic features of the survey population and those of the respondents. Of the 55 respondents in the survey sample 82 percent of organizations were either manufacturing or component suppliers. 80 percent of organizations had less than 320 employees in their Australian operations and the average number of employees was 232. According to 2004 figures, three automotive brands (Toyota, Ford and Holden) held over a 50 percent share of the Australian automotive market (AIG, 2005). The export figure of 36 percent shown in Table I is expected to reflect the dominance of these three brands in the volume of exported product shown as a gross amount for the Australian industry. The average figure of 12 percent of exported product for those remaining in the industry is expected to be a fair representation of the average automotive supplier rather than any of the larger employers such as the assembly firms. On the basis of a comparison between the expected demographics of the survey population and the demographics of those that responded, it was concluded that the responding population was representative of the larger and target population. Non-response was attributed to “lack-of-interest” bias only (Armstrong and Overton, 1977).

Demographic variable

Population

a, b

Average number of employees Between 60 and 500 Dominant type of organization Component manufacturer Table I. Population demographics and respondent demographics

Average volume of exported product

36 percent (including main assemblers)

Notes: aILO (2004); bAIG (2005)

Survey respondents 232 Component manufacturer (82 percent) 12 percent (not including main assemblers)

Measures Multi-item scales for the survey were developed from both the case study findings and the work of previous authors. A draft survey was pre-tested with Toyota Motor Corporation Australia purchasing staff, six first tier automotive component suppliers and academics. The final instrument used a 5-point Likert scale (1 ¼ not at all; 3 ¼ to some extent; 5 ¼ to a very large extent). Final measures used for the analysis are summarized in the following. Each measure was extracted using factor analytic techniques after Hair et al. (1998) and Tabachnick and Fidell (2001). Environmental commitment. A measure for environmental commitment was drawn from the findings of previous authors (Banerjee, 2002; Aragon-Correa, 1998; Buysse and Verbeke, 2003) and the evidence presented by the supplier case studies. This was intended to capture an aggregate measure of the supplier’s underlying commitment to its environmental responsibilities using a scale which indicated environmental commitment as expressed through the organization’s policies, values and employee awareness programs. The extracted scale for environmental commitment is shown in Table II. Customer’s environmental performance requirements. A measure for the customer’s environmental performance requirements for its suppliers was derived from three items developed largely during the preliminary study. Firstly, a measure of the customer’s explicit or contract-based environmental performance requirements was used to indicate a minimum expectation from suppliers (i.e. compliance to the ISO14001 management standard). Secondly, a measure of the customer’s more implicit or less tangible environmental performance requirements – such as its own values or policies that the supplier might try to emulate – was used to indicate the customer’s policy or position statement with regard to its environmental responsibilities. Finally, a measure of the customer’s actual commitment to its environmental performance requirements for suppliers was used to indicate the extent to which suppliers expected the customer to penalize them for non-compliance with say ISO14001 or other contract-based requirements. The extracted scale for customer’s environmental performance requirements is shown in Table III. Conditions of the customer-supplier relationship. To explore the effect of supply “conditions” on the relationship between the customer’s environmental performance requirements and the supplier’s environmental commitment the analysis includes three moderator variables. These include variables for Investment (after Handfield and Bechtel, 2002; Krause et al., 2000), Contracts (after Heide and Stump, 1995) and Assessment (after Krause et al., 2000 and Klassen and Vachon, 2003). The individual moderator variables for Investment, Contracts and Assessment are shown in Table IV.

Description 1. Our firm has a clear policy statement urging environmental awareness in every area of the business 2. Protecting the environment is a central corporate value in our firm 3. At our firm, we make a concerted effort to make every employee understand the importance of environmental management

Greening the automotive supply chain 39

Factor loadings Cronbach Bartlett’s test (rotated) MSA a

0.826 0.916 0.851

0.701

0.891

p , 0.001 Table II. Extracted scale for environmental commitment

IJOPM 27,1

40

Data analysis Returned surveys were analysed using linear regression analysis with SPSS V.12. Data were checked first for normality assumptions using normal probability plots and tests for kurtosis and skewness. Items were reduced into relevant scales using factor analysis (principal component with a varimax rotation) and factor loadings of less than 0.65 were excluded (Tables II, III and IV). Tests for multi-collinearity were completed using variance-inflation and tolerance factors. Overall, significance of the regression model was assessed using the test statistics of standardized (b) co-efficients, standard error of the co-efficient, F and adjusted R 2. A final moderated multiple regression (MMR) analysis was used to test H2a-2c after that described in Baron and Kenny (1986) and later in Aguinis (2004). According to Hair et al. (1998) a minimum number of 50 observations is required to conduct a factor and regression analysis. The main problem associated with a small sample size in statistical analyses described by Hair et al. (1998) and other authors (Tabachnick and Fidell, 2001) is the inability to detect statistically significant relationships that only appear in larger samples. Alternatively, larger sample sizes attract heightened sensitivity in that normally weak relationships can appear more significant than they would in a smaller sample. Techniques can be used to add confidence to the statistical findings arising from a regression analysis or a MMR analysis on a small sample size. These include an additional methodology such as qualitative interviews or case studies either pre-survey or post and confining the number of variables in any analysis to a minimum number so as to maximize the degrees of freedom (Hair et al., 1998). Although small sample sizes suffer from problems of generalisability and low statistical power, any highly significant relationships found during an appropriate statistical analysis often can prove more robust than when found in a large and statistically sensitive sample (Hair et al., 1998; Tabachnick and Fidell, 2001).

Factor loadings Cronbach Bartlett’s test (rotated) MSA a

Description

Table III. Extracted scale for customer’s environmental performance requirements

1. Our major customer requires us to achieve ISO14000 certification 2. Our major customer has a clear policy statement on their commitment to the environment 3. Our major customer would withhold our supply contract if we did not meet their environmental performance requirements

Investment Table IV. Individual moderator variables

Contracts Assessment

0.923

0.663

0.762

p , 0.001

0.847 0.677

We dedicate and reserve equipment and capacity specifically to maintain this relationship We have signed an extensive agreement (or contract) with this customer specifying price, quality and lead-time This customer assesses our operations (e.g. questionnaire) from time to time

Results The primary relationship which is proposed between the main construct supplier’s environmental commitment and customer’s environmental focus is articulated through H1. The results of the regression analysis for H1 are shown in Table V. In Table V, standardized co-efficients and adjusted R 2 values are provided to describe the results of the analysis. The results of the regression analysis show no statistically significant relationship between the supplier’s level of environmental commitment and the customer’s environmental performance requirements. At the item-specific level, one variable contributed most of the predicted variance in the relationship between the supplier’s environmental commitment and customer’s environmental performance requirements – supplier certification to ISO14001 – with the remaining two items predicting much less. The second stage of the analysis involved including a series of condition variables as moderators of the primary relationship described by H1. Using MMR analysis, the results of this analysis which describes and tests H2a-2c is shown in Table VI. In Table VI, standardized co-efficients are shown for the regression analysis for each of the moderator variables – Investment (H2a), Contracts (H2b) and Assessment (H2c) – along with adjusted R 2 values, change in adjusted R 2 values between the un-moderated and moderated models and the F statistic.

Customer’s environmental performance requirements Customer requires us to achieve ISO14000 certification Customer has a clear policy statement on their commitment to the environment Customer would withhold our supply contract if we did not meet their environmental performance requirements Notes:

Investment: we dedicate and reserve equipment and capacity specifically to maintain this relationship Contracts: we have signed an extensive agreement (or contract) with this customer specifying price, quality and lead-time Assessment: this customer assesses our operations (e.g. questionnaire) from time to time › a

Notes: b values

0.314

0.063

0.075 2 0.048

*p , 0.05; * *p , 0.01; * * *p , 0.001. Standardised co-efficients are shown as

Moderator variables (Z)

41

Environmental commitment b Adj. R 2

Independent construct



Greening the automotive supply chain

b values

Table V. Testing H1 – regression analysis of environmental commitment and customer’s environmental performance requirements

Environmental commitment (Y) Environmental performance requirements (X) D R2 F statistic 0.104

7.15 * *

0.015

0.87

0.028

1.59

p , 0.1; *p , 0.05; * *p , 0.01; * * *p , 0.001. Standardised co-efficients are shown as

Table VI. Testing H2a, 2b and 2c – MMR analysis of customer’s environmental performance requirements when moderated by relationship conditions

IJOPM 27,1

In the MMR analysis, the primary regression relationship between the environmental commitment construct and the customer’s environmental performance requirement construct becomes statistically significant for the Investment model (change in adjusted R 2 is significant at p , 0.001) but not for the Contracts or Assessment models.

42

Discussion As other authors have described, the environmentally committed organization can be more reliably considered to have well-developed systems and practices of environmental management, have higher levels of environmental performance and be more likely to respond to its customer’s environmental performance requirements. For the customer’s environmental performance requirements, the presence of Investment between the customer and the supplier provided a significant influence on the response of the supplier to its customer’s needs. In the analysis a measure of transaction-specific investment or asset-specificity (termed Investment) was provided by the item – We dedicate and reserve equipment and capacity specifically to maintain this relationship. This represented sunk assets for the supplier that as they were increased, also decreased their value as re-deployable assets. In the relationship between the customer’s environmental performance requirements and the supplier’s environmental commitment, the Investment condition as a moderator proved highly significant as a predictor of variance in the environmental commitment construct (significance of the change in adjusted R 2 value – p , 0.01). The influence of the customer’s environmental performance requirements on the supplier’s level of environmental commitment was not significant unless specific customer-supplier relationship conditions were also present to moderate the supplier’s response. The customer’s environmental performance requirements were not accommodated in a homogenous manner across all customers as might have been expected from the findings described by previous research, but were instead dealt with by the supplier in a more heterogeneous manner according to the importance of the customer-supplier relationship. In the absence of customer-supplier relationship conditions – that is without any “legitimacy” for the customer’s environmental performance requirements, as assessed by the supplier – the relationship described by hypothesis H1 was not significant. The process of supplier performance improvement can often involve a long-term cooperative effort by the customer including new equipment, supplier training and deployment of customer staff into the supplier’s plant. It can also take a low involvement form whereby the customer relies on the competitive force of the market to extract performance improvement (Krause et al., 2000; Scannell et al., 2000; Handfield et al., 2000). Going back to Williamson (1975) and the original transaction cost framework it is the existence of either transaction-specific investments and/or contracts that both minimises supplier opportunism from the customer’s perspective and has positive performance implications for the relationship. As described in the literature review, higher levels of asset specificity between the customer and supplier have been found by other authors to lead to positive performance gains for both firms involved in the exchange (Dyer, 1997; Handfield and Bechtel, 2002).

The findings described in this study support the importance of asset-specific investments in the relationship between a customer’s environmental performance requirements and their potential to improve the supplier’s commitment to its environmental responsibilities. Without Investment the supplier appears much less likely to be responsive to the customer’s requirements for environment-related performance improvements. These findings are supported by observations made during earlier discussions with operations managers in Australian automotive component supplier firms such that suppliers were desirous of emulating or adopting their customer’s environmental practices especially where the supply relationship was described by the supplier as being most critical to their firm’s survival. Conclusions and further research Encapsulating environmentally relevant goals, practices or technologies within supply requirements bears significant potential for large organizations to impart new knowledge to their under-resourced suppliers. It provides a modus for organizations to extend their goals of corporate social responsibility, communicate their commitment to such goals and provide a leadership role to their suppliers. The major finding of this research was that the customer’s environmental performance requirements can have a positive influence on a supplier’s strategic level of commitment toward its environmental responsibilities. The presence of relationship conditions that promote greater levels of financial commitment between the supplier and the customer – relationship-specific investment – is expected to increase this influence. Increasing levels of the supplier’s strategic environmental commitment is in turn expected to have a positive impact on the supplier’s environmental performance. A practical application of these findings may be that customers consider the application of any program of green-supply in a hierarchical manner such that critical or strategic suppliers receive more intensive assistance or development with their environmental performance. Non-strategic suppliers, particularly those that provide more commodity goods or are sourced on a market basis, could be required to certify to a basic level of compliance such as an industry management standard (ISO14001). This provides the customer with a minimum assurance of risk management and environmental performance without the associated transactional investment. At a less tangible level our study suggests that customers should remain conscious of the old adage “Do as I say and not as I do” such that suppliers may become less responsive to the customer’s environmental performance requirements where the customer does not demonstrate a level of commitment toward its environmental performance that exceeds its own requirements for the supplier. Ideally a customer intending to upgrade the environmental performance of its suppliers could consider that its suppliers are initially likely to wait for the customer to determine their environmental responsibilities for them. With time these suppliers will develop these competencies and determine their own strategies or responsibilities. The customer’s goal for those organizations involved in its supply chain should be at a minimum, to reduce their overall waste burden or generation of pollution in both hazardous and non-hazardous forms. This has a dual purpose of both decreasing the organization’s actual or potential impact on the environment and also reducing costs for supply chain stakeholders.

Greening the automotive supply chain 43

IJOPM 27,1

44

The exploratory nature of this study has provided for the development of a more rigorous conceptual model that may be applied to further studies of the implications of supply greening. Of particular interest will be the use of the revised model and hypotheses to a larger sample size or different industry. Many additional questions still remain surrounding the relationship factors that might support, influence or degrade any customer or supplier-driven program of supply greening. This study has been able to provide some empirical support for a number of potential new theories in a large and under-developed field of research. Customer-driven programs of green-supply remain a potentially powerful tool for reducing the environmental impacts of product supply chains in addition to the influence that other non-financial stakeholders (government, community, employees) may have. Limitations The conduct of this research is limited to the Australian automotive industry and its constituents. Although this industry is populated by a number of global assemblers and many organizations supplying internationally, the study’s participants operated and resided within the Australian cultural context. Equally, this industry operates and supplies almost entirely toward the provision of only automotive goods. The analysis and findings of this research should be considered with respect to this context. The quantitative findings describe the results of an analysis based on 55 organizations that responded to an industry-wide survey (out of a possible 397 organizations). Whilst it is still possible to detect significant effects and make reasonable assumptions with respect to hypotheses with a sample of this size, it limits the ability to establish any causality or more complex relationship (i.e. mediation or path models). Equally the sample size is too small to analyze any more complex scenarios such as a path scenario or latent constructs. The techniques used aim to explore basic relationships and maximize confidence in the potential for causality only. References ABS (2005), “Australia’s automotive industry”, Year Book Australia – Manufacturing, Australian Bureau of Statistics, Belconnen, January. AIG (2005), The Victorian Automotive Components Industry, Australian Industry Group, Melbourne. Aguinis, H. (2004), Regression Analysis for Categorical Moderators, Guilford Press, New York, NY. Aragon-Correa, J. (1998), “Strategic proactivity and firm approach to the natural environment”, Academy of Management Journal, Vol. 41 No. 5, pp. 556-67. Aragon-Correa, J. and Sharma, S. (2003), “A contingent resource-based view of proactive corporate environmental strategy”, Academy of Management Review, Vol. 28 No. 1, pp. 71-88. Armstrong, J. and Overton, T. (1977), “Estimating non-response bias in mail surveys”, Journal of Marketing Research, Vol. 14 No. 3, pp. 396-400. Banerjee, S. (2002), “Corporate environmentalism - the construct and its measurement”, Journal of Business Research, Vol. 55 No. 3, pp. 177-91. Bansal, P. (2005), “Evolving sustainably, a longitudinal study of corporate sustainable development”, Strategic Management Journal, Vol. 26 No. 3, pp. 197-218.

Baron, R. and Kenny, D. (1986), “The moderator-mediator variable distinction in social psychological-research: conceptual, strategic and statistical considerations”, Journal of Personality and Social Psychology, Vol. 51 No. 6, pp. 1173-82. Bowen, F., Cousins, P., Lamming, R. and Faruk, A. (2001a), “The role of supply management capabilities in green supply”, Production and Operations Management, Vol. 10 No. 2, pp. 174-89. Bowen, F., Cousins, P., Lamming, R. and Faruk, A. (2001b), “Horses for courses: explaining the gap between the theory and practice of green supply”, Greener Management International, No. 35, pp. 41-60. Burt, D. and Doyle, M. (1993), The American Keiretsu: A Strategic Weapon for Global Competitiveness, Business One Irwin, Homewood, IL. Buysse, K. and Verbeke, A. (2003), “Proactive environmental strategies: a stakeholder management perspective”, Strategic Management Journal, Vol. 24 No. 5, pp. 453-70. Cousins, P. and Stanwix, E. (2001), “It’s only a matter of confidence! A comparison of relationship management between Japanese- and UK non-Japanese-owned vehicle manufacturers”, International Journal of Operations & Production Management, Vol. 21 Nos 9/10, pp. 1160-79. David, R. and Han, S. (2004), “A systematic assessment of the empirical support for transaction cost economics”, Strategic Management Journal, Vol. 25, pp. 39-58. DeVellis, R. (2003), Scale Development: Theory and Applications, Sage, Thousand Oaks, CA. Dillman, D. (1999), Mail and Electronic Surveys: The Tailored Design Method, Wiley, New York, NY. Dyer, J. (1997), “Effective interfirm collaboration: how firms minimize transaction costs and maximize transaction value”, Strategic Management Journal, Vol. 18 No. 7, pp. 535-56. Dyer, J. and Chu, W. (2003), “The role of trustworthiness in reducing transaction costs and improving performance: empirical evidence from the United States, Japan, and Korea”, Organization Science, Vol. 14 No. 1, pp. 57-68. Dyer, J. and Nobeoka, K. (2000), “Creating and managing a high-performance knowledge-sharing network: the Toyota case”, Strategic Management Journal, Vol. 21 No. 3, pp. 345-67. Egri, C. and Pinfield, L. (1996), “Organizations and the biosphere, ecologies and environments”, in Clegg, S., Hardy, C. and Nord, W. (Eds), Handbook of Organization Studies, Sage, London. Florida, R. (1996), “Lean and green: the move to environmentally conscious manufacturing”, California Management Review, Vol. 39 No. 1, pp. 80-105. Friedman, M. (1970), “The social responsibility of business is to increase its profits”, The New York Times Magazine, September 13, New York Times Company. Geffen, C. and Rothenberg, S. (2000), “Suppliers and environmental innovation - the automotive paint process”, International Journal of Operations & Production Management, Vol. 20 No. 2, pp. 166-86. Ghoshal, S. and Moran, P. (1996), “Bad for practice, a critique of the transaction cost theory”, Academy of Management Review, Vol. 21 No. 11, pp. 13-47. Green, K., Morton, B. and New, S. (1998), “Green purchasing and supply policies: do they improve company’s environmental performance?”, Supply Chain Management: An International Journal, Vol. 3 No. 2, pp. 89-95. Green, K., Morton, B. and New, S. (2000), “Greening organizations”, Organization & Environment, Vol. 13 No. 2, pp. 206-28.

Greening the automotive supply chain 45

IJOPM 27,1

46

Grover, V. and Malhotra, M. (2003), “Transaction cost framework in operations and supply chain management research: theory and measurement”, Journal of Operations Management, Vol. 21 No. 4, pp. 457-73. Hair, J., Anderson, R., Tatham, R. and Black, W. (1998), Multivariate Data Analysis, Prentice-Hall International, Englewood Cliffs, NJ. Hall, J. (2000), “Environmental supply chain dynamics”, Journal of Cleaner Production, Vol. 8, pp. 455-71. Hall, J. (2001), “Environmental supply-chain innovation”, Greener Management International, Vol. 35, pp. 105-19. Handfield, R. and Bechtel, C. (2002), “The role of trust and relationship structure in improving supply chain responsiveness”, Industrial Marketing Management, Vol. 31 No. 4, pp. 367-82. Handfield, R., Krause, D., Scannell, T. and Monczka, R. (2000), “Avoid the pitfalls in supplier development”, Sloan Management Review, Vol. 41 No. 2, p. 37. Heide, J. and Stump, R. (1995), “Performance implications of buyer-supplier relationships in industrial-markets - a transaction cost explanation”, Journal of Business Research, Vol. 32 No. 1, pp. 57-66. Henriques, I. and Sadorsky, P. (1999), “The relationship between environmental commitment and managerial perceptions of stakeholder importance”, Academy of Management Journal, Vol. 42 No. 1, pp. 87-99. Hoffman, A. (1999), “Institutional evolution and change: environmentalism and the US chemical industry”, Academy of Management Journal, Vol. 42 No. 4, pp. 351-71. Hunt, C. and Auster, E. (1990), “Proactive environmental management: avoiding the toxic trap”, Sloan Management Review, Vol. 31 No. 2, pp. 7-18. Judge, W. and Elenkov, D. (2005), “Organizational capacity for change and environmental performance, an empirical assessment of Bulgarian firms”, Journal of Business Research, Vol. 58, pp. 893-901. King, A. and Lenox, M. (2002), “Exploring the locus of profitable pollution reduction”, Management Science, Vol. 48 No. 2, pp. 289-99. Klassen, R. (2000), “Exploring the linkage between investment in manufacturing and environmental technologies”, International Journal of Operations & Production Management, Vol. 20 No. 2, pp. 127-47. Klassen, R. (2001), “Plant-level environmental management orientation: the influence of management views and plant characteristics”, Production and Operations Management, Vol. 10 No. 3, pp. 257-75. Klassen, R. and Vachon, S. (2003), “Collaboration and evaluation in the supply chain: the impact on plant-level environmental investment”, Production and Operations Management, Vol. 12 No. 3, pp. 336-52. Krause, D., Scannell, T. and Calantone, R. (2000), “A structural analysis of the effectiveness of buying firms strategies to improve supplier performance”, Decision Sciences, Vol. 31 No. 1, pp. 33-55. ILO (2004), Automotive Industry Trends Affecting Component Suppliers, International Labour Office, International Labour Organization, Geneva. Lambert, D. and Cooper, M. (2000), “Issues in supply chain management”, Industrial Marketing Management, Vol. 29 No. 1, pp. 65-83. Lamming, R. (1993), Beyond Partnership: Strategies for Innovation and Lean Supply, Prentice-Hall, New York, NY.

Lamming, R. (1996), “Squaring lean supply with supply-chain management”, International Journal of Operations and Production Management, Vol. 16 No. 2, pp. 183-96. Lamming, R. and Hampson, J. (1996), “The environment as a supply chain management issue”, British Journal of Management, Vol. 7, pp. S45-S62. Liker, J. and Choi, T. (2004), “Building deep supplier relationships”, Harvard Business Review, December. Melnyk, S., Sroufe, R. and Calantone, R. (2003), “Assessing the impact of environmental management systems on corporate and environmental performance”, Journal of Operations Management, Vol. 21 No. 3, pp. 329-51. Miles, R. and Snow, C. (1978), Organizational Strategy, Structure, and Process, McGraw-Hill, New York, NY. Mitchell, R., Agle, B. and Wood, D. (1997), “Toward a theory of stakeholder identification and salience: defining the principle of who and what really counts”, Academy of Management Review, Vol. 22 No. 4, pp. 853-86. Pfeffer, J. and Salancik, G. (1978), The External Control of Organizations, Harper and Row, New York, NY. Rindfleisch, A. and Heide, J. (1997), “Transaction cost analysis: past, present, and future applications”, Journal of Marketing, Vol. 61 No. 4, pp. 30-54. Rao, P. (2002), “Greening the supply chain: a new initiative in South East Asia”, International Journal of Operations and Production Management, Vol. 22 No. 6, pp. 632-55. Rao, P. and Holt, D. (2005), “Do green supply chains lead to competitiveness and economic performance?”, International Journal of Operations & Production Management, Vol. 25 Nos 9/10, pp. 898-916. Ring, P. and Van De Ven, A. (1992), “Structuring cooperative relationships between organizations”, Strategic Management Journal, Vol. 13 No. 7, pp. 483-98. Sarkis, J. (2003), “A strategic decision framework for green supply chain management”, Journal of Cleaner Production, Vol. 11 No. 4, pp. 397-409. Scannell, T., Vickery, S. and Droge, C. (2000), “Upstream supply chain management and competitive performance in the automotive supply industry”, Journal of Business Logistics, Vol. 21 No. 1, pp. 23-48. Sharma, S. and Vredenburg, H. (1998), “Proactive corporate environmental strategy and the development of competitively valuable organizational capabilities”, Strategic Management Journal, Vol. 19 No. 8, pp. 729-53. Smith, A. (1776), An Inquiry into the Nature and Causes of the Wealth of Nations, Methuen and Company Limited, London. Snow, C. and Thomas, J. (1994), “Field research methods in strategic management: contributions to theory building and testing”, Journal of Management Studies, Vol. 31 No. 4, pp. 457-80. Starbucks (2005), “Supplier code of conduct”, available at: www.starbucks.com/aboutus/ supplier_code.asp (accessed September 20). Stump, R. and Heide, J. (1996), “Controlling supplier opportunism in industrial relationships”, Journal of Marketing Research, Vol. 33 No. 4, pp. 431-41. Tabachnick, B. and Fidell, L. (2001), Using Multivariate Statistics, Allyn and Bacon, Boston, MA. Wacker, J. (1998), “A definition of theory: research guidelines for different theory-building research methods in operations management”, Journal of Operations Management, Vol. 16 No. 4, pp. 361-85.

Greening the automotive supply chain 47

IJOPM 27,1

48

Williamson, O. (1975), Markets and Hierarchies, Analysis and Antitrust Implications, Free Press, New York, NY. Williamson, O. (1985), The Economic Institutions of Capitalism: Firms, Markets, Relational Contracting, Free Press, New York, NY. Yin, R. (2003), Case Study Research: Design and Methods, Sage, Thousand Oaks, CA. Young, A. and Kielkiewicz-Young, G. (2001), “Sustainable supply network management”, Corporate Environmental Strategy, Vol. 8 No. 3, pp. 260-8. Zhu, Q. and Sarkis, J. (2004), “Relationships between operational practices and performance among early adopters of green supply chain management practices in Chinese manufacturing enterprises”, Journal of Operations Management, Vol. 22 No. 3, pp. 265-89. Corresponding author Dayna Simpson can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0144-3577.htm

Managers’ perceptions of learning in new product development

Managers’ perceptions of learning in NPD

Ursula Koners and Keith Goffin Cranfield School of Management, Cranfield, UK

49

Abstract Purpose – Post-project reviews (PPRs) are widely recommended as an appropriate mechanism to stimulate project-to-project learning in new product development (NPD) teams. However, empirical research on their potential is limited and so this study aims to answer two questions: how do NPD professionals perceive PPRs? and is tacit knowledge created during PPRs? Design/methodology/approach – Five in-depth case studies were conducted in German companies. The main sources of data were: company documentation; in-depth interviews with managers; and observation of PPRs. The different data sources enabled extensive data triangulation and a high degree of reliability and validity to be achieved. Findings – The results indicate that NPD personnel perceive PPRs to be a very useful mechanism. They also show that social interactions and tacit knowledge play key roles in NPD learning. Research limitations/implications – Tacit knowledge was investigated using a proxy measure – the usage of metaphors and stories in NPD discussions and documents. More robust ways of studying the generation and transfer of tacit knowledge are needed. The whole area of knowledge and learning in NPD requires significant further study. Practical implications – Management needs to drive the dissemination of the results of PPRs more effectively, in order to make more use of the tacit knowledge generated. Originality/value – The main contributions of this study are the insights it provides on how PPRs are perceived and the exploratory investigation it makes of tacit knowledge – a complex topic that has previously only been discussed at a theoretical level. Keywords New products, Project evaluation, Project management Paper type Viewpoint

Introduction Managers responsible for new product development (NPD) need to constantly improve their processes and strengthen core R&D capabilities (Wheelwright and Clark, 1992). A post-project review (PPR) is, “a formal review of the project which examines the lessons which may be learnt and used to the benefit of future projects” (Lane, 2000) and such reviews are a potentially valuable method to capture the knowledge generated during the course of a NPD project. The importance of PPRs is frequently stressed by practitioners and academics, but rigorous research into how they are typically conducted, are perceived, or how learning occurs, is scarce. Companies rarely conduct PPRs – as shown by a wide body of anecdotal evidence (Tidd et al., 1997) and a few empirical studies (Bowen et al., 1994; Huber, 1996; Saban et al., 2000). An opportunity for learning is lost when PPRs are not used and this can result in companies making very similar mistakes to those made in previous NPD projects (Tidd et al., 1997). This paper describes how NPD personnel perceive PPRs and this is important because only if they are positively perceived, will the necessary motivation for learning be present. The research also investigated the role of social interactions and tacit

International Journal of Operations & Production Management Vol. 27 No. 1, 2007 pp. 49-68 q Emerald Group Publishing Limited 0144-3577 DOI 10.1108/01443570710714538

IJOPM 27,1

50

knowledge in NPD. The scope of the literature that should be considered relevant to the study of PPRs is a salient point. Both the project management and NPD literature give useful pointers. The organizational learning literature is also relevant but, surprisingly, most researchers working in the NPD context have failed to apply this knowledge (McKee, 1992). In this paper we review all three literatures to identify gaps in the extant knowledge and gain insights into suitable methodologies. Next, we explain the choice of methodology and the actions taken to ensure validity. The results are then presented and the implications for practitioners and academics are identified, including ideas for further research. Review of the extant literature Project management literature It was in the 1950s that the need to review the performance of projects was first recognized, at the same time as project management was emerging as a discipline (Weinberg and Freedman, 1984). Later, recommendations appeared on how to conduct reviews at the end of a project (Gulliver, 1987) and the advantages of PPRs became apparent. However, it has been identified that a limitation of the project management literature on PPRs is that most previous studies have not been conducted in a systematic fashion and so the validity of the recommendations is questionable (Koners and Goffin, 2005). The literature identifies three main issues about PPRs. Firstly, much can be learnt from previous projects and this can help prevent similar mistakes being made again (Pitman, 1991). Secondly, disseminating the lessons learned across the organization is of critical importance (Ayas, 1996). This can be achieved by using databases of lessons learned, rotating personnel, and circulating written reports (Balthazor, 1994; Holtshouse, 1999). Thirdly, each and every project needs to contribute to the continuous improvement of the organization’s processes (Prahalad and Hamel, 1990). However, the adoption of PPRs has been slow. There are a number of recommendations in the literature on how best to conduct PPRs (Baird et al., 1999; Busby, 1999; Freedman and Weinberg, 1977; Schindler and Eppler, 2003). However, many of these are vague including “discourage glib categorization” (a recommendation given but not explained by Busby, 1999). Others are more specific. For example, Schindler and Eppler (2003) suggest a “project knowledge broker” should be responsible for transferring the lessons learnt within and between project teams. Such mechanisms are necessary as there are many individual and organizational barriers to learning in NPD. The fact that few organizations conduct PPRs is the result of various factors. The pressure of current projects leads to a lack of time to reflect on past projects (Kotnour, 1999). Managers may be uncertain as to whether they can learn from project experiences (Boudes et al., 1998). Reviews often suffer from the reluctance of participants to critically evaluate performance, although most people genuinely want to review past projects (Gulliver, 1987). Experiences and insights from projects may be hard to share and difficult to capture in reports and databases (Durrance, 1998). NPD literature Knowledge has become recognized as a key source of long-term competitive advantage in R&D (Corso et al., 2001). Consequently, the importance of learning from NPD

projects is made by several authors (Bowen et al., 1994; Leonard-Barton, 1992; Liyanage et al., 1999). However, there has only been sparse application of the principles of learning theory to empirical studies of the NPD process – there is a need for more research into how learning can improve product development (Saban et al., 2000). Such learning can take place at the individual and organizational levels and there are many barriers to successful learning in NPD teams (Lindkvist, 2001). It has been identified that PPRs are seldom used (Bourgault and Sicotte, 1998; Bowen et al., 1994; Wheelwright and Clark, 1992) but there have been few empirical studies of this topic. One study showed that only two out of 33 microelectronic manufacturers use formal PPRs (Boag and Rinholm, 1989). Goffin and Pfeiffer (1999) found that only four of their 16 case study companies used PPRs but failed to give details of how they were used. A survey of 63 NPD managers identified that only 3 percent of their organizations conduct a review after every project but the majority (94 percent) think their organizations should conduct PPRs (von Zedtwitz, 2003). There are a number of papers which give recommendations of how to run PPRs (Duarte and Snyder, 1997; Lilly and Porter, 2003; Smith, 1996; Wheelwright and Clark, 1992). These are often based on small samples, or the personal experience of the authors. Similar to the project management literature, the recommendations given are seldom based on empirical data and so it is unclear how valid they are (Koners and Goffin, 2007). Moreover, the focus of the NPD literature is on knowledge that can be written down, documented and easily shared. In our previous research we collated the findings in the literature and then collected empirical data to further the understanding of the characteristics of PPRs that influence their effectiveness. Based on five case studies (Koners and Goffin, 2007), 14 different characteristics (Table I) were identified as pertinent. For example, the research showed that the timing of PPRs is important, as are the participants and the way in which such a meeting is moderated and the findings disseminated. Although this research gives insights into the way in which PPRs should be organized, there are other important issues. For example, there have been no studies of whether NPD professionals who participate in PPRs perceive them to be valuable and effective. Such studies are needed, as the motivation of individuals is a necessary but not sufficient condition for learning in NPD teams (Lindkvist, 2001). Just as managers’ perceptions of PPRs have been overlooked, so it is that most researchers have ignored learning theory when looking at knowledge creation in NPD (McKee, 1992). The work of Thomke and Fujimoto (2000) recognized the difference between knowledge in R&D that is easy to share and that which is best transferred by close interaction. Thus, their research is in the vanguard of applying concepts from organizational learning to NPD but unfortunately it did not investigate PPRs. Organizational learning There is a vast body of knowledge on organizational learning but little agreement as to what it really is and how it occurs (Chiva and Allegre, 2005). However, there are two concepts in the organizational learning literature that appear particularly relevant to NPD: the understanding of knowledge and the importance of social interactions for learning. The concept of knowledge, and the differences between “tacit” and “explicit” knowledge relate back to Polanyi (1962) and his famous quote that “we can know more than we can tell.” Explicit knowledge is what we can readily explain and document,

Managers’ perceptions of learning in NPD 51

IJOPM 27,1

52

Table I. 14 key characteristics of PPRs

Characteristics of PPRs

Conclusions (based on five case studies and the literature)

1.

Objective of PPRs

2.

Timing of PPRs

3.

Duration of PPRs

4.

PPR participants

5.

Moderation of PPRs

6.

PPR discussion method

7.

Location for PPR

8.

Use of guidelines for PPRs

9.

Preparation of PPR

10.

Atmosphere during PPRs

11.

Documentation of the results of PPRs

12.

Dissemination of PPR results

13.

Creation of action points

14.

Agreement on improvement suggestions

Closure is important Focus is normally on learning from mistakes – there is an opportunity to learn from successes PPRs support knowledge dissemination and this should be an objective Guidelines of about six months after market launch are not usually followed due to timing problems or other priorities Discipline is needed to ensure PPRs happen Length of PPRs varies a lot which is also reflected in the results and depth of discussion Core team always needs to be present The presence of senior management at the presentation of the results both acts as a motivator and a means of disseminating the knowledge gained The responsibility for running and moderating a PPR is often given to the project manager Using a professional moderator (as used by Appliances Co.) appears to be a more effective way to challenge the team and generate more learning The setting, the questions asked by the moderator, and the visual aids used all appear to influence the depth of discussion and the learning generated Separate meeting rooms are always used External rooms are sometimes deliberately chosen to facilitate open discussion and concentration outside of the company Only one company uses its guidelines for PPRs, others do not have them, only use the compulsory part or do not use them at all Preparation is largely based on the individual reflection of the team members in advance The moderator and project leader should prepare a specific agenda for the PPR Degree of openness and formality depends on participants and company culture “atmosphere” Hard to measure but very important PPR discussions should be documented but more effective dissemination is needed A presentation to senior management supports a wider awareness Results are only received by the project team (i.e. the participants) and senior management Limited dissemination outside the project team – a missed opportunity Action points are derived by all companies Follow-up is problematic if responsibility is not allocated to project manager Two companies have a target for the minimum number of improvement suggestions All companies document them in their minutes Only one company presents them to the management

Source: Koners and Goffin (2007)

whereas tacit knowledge is difficult to articulate and exists at a subconscious level. Although it is possible to distinguish theoretically between explicit and tacit knowledge, they are hard to differentiate in practice (Johanessen et al., 2001; Lam, 2000; von Krogh, 1998). It should be noted that there is some controversy about whether tacit knowledge can be converted into explicit knowledge (Cook and Brown, 1999). However, the discussion in the literature has largely been at a theoretical level and has failed to make empirical inroads into the understanding of tacit knowledge. Studying tacit knowledge empirically is a problem area (Wong and Radcliffe, 2000). Consequently, the main problem with attempting to apply the concepts from the organizational learning literature to NPD is the difficulty in operationalizing the constructs. However, metaphors and stories have been recognized in the literature as indicators of the generation and exchange of tacit knowledge (Cook and Brown, 1999; Nonaka, 1994). In fact, individuals use metaphors to help explain their intuition to themselves and share it with others (Crossan et al., 1999). The existence of metaphors and stories can be taken as evidence that tacit learning is taking place (Gherardi, 2000). Despite the recommendations in the literature that metaphors and stories can be used as a measure, there is a notable absence of empirical work. The organizational learning literature identifies the importance of social interactions for the transfer of knowledge. Tacit knowledge can only be transferred through detailed discussions among people from similar backgrounds and with common experiences. Communities of Practice (CoPs) are groups of people who are informally bound to one another by exposure to a common class of problems (Wenger and Snyder, 2000). This exposure leads to a high degree of common knowledge, understanding and language, and experience which supports the efficient transfer of knowledge. Project teams can be considered an embryonic form of a CoP (Sense and Antoni, 2003). CoPs theory views learning as a social phenomenon and claims that knowledge, particularly tacit knowledge, can only be produced and held collectively (Howells, 1996). It should be noted that several of the characteristics of PPRs shown in Table I relate strongly to social interactions (e.g. the way such a meeting is moderated, or held in an informal setting). Conclusions from the literature Much of the literature on PPRs has discussed their importance, identified that few R&D organizations use them, and generated recommendations for how they should be organized. Overall, the key conclusions for the current research are: . Although the key characteristics of PPRs have been identified, there are still many aspects about them that warrant detailed investigation. . Both academics and practitioners stress the importance of PPRs but have not investigated whether managers directly involved in NPD perceive them positively. Such an investigation is important as, without personal motivation of staff, the capacity for learning will be limited. . The tacit dimension of learning in NPD is poorly understood. Research design Learning can occur at the individual, project team and project-to-project level and more needs to be understood about the first two before the latter can be adequately investigated. The gaps identified in the literature led to an in-depth study of PPRs,

Managers’ perceptions of learning in NPD 53

IJOPM 27,1

covering a range of topics, including the characteristics of PPRs (Koners and Goffin, 2007) and the type of learning generated. This paper focuses on perceptions of PPRs, including their ability to generate knowledge. It sets out answers to the following research questions: RQ1. How do NPD professionals perceive PPRs?

54

To answer this question we looked at what NPD professionals think of PPRs, what they want to achieve through them, how PPRs support learning, and perceptions of ideal outcomes and how learning can be disseminated. RQ2. Is tacit knowledge created during PPRs? To answer this we explored for evidence of tacit knowledge. Case study methodology Based on the exploratory nature of the research, in-depth case studies were selected. These are appropriate when researching complex social phenomena in real-life contexts (Eisenhard, 1989; Yin, 1994). Furthermore, they allow researchers to look at a wide array of variables and aspects (Hartley, 1994). However, we recognized the need to design the cases carefully to ensure sufficient rigor, through focusing on construct validity and internal validity. Construct validity was addressed by using operational measures for the different phenomena under study. For example, it was necessary to find appropriate “measures” for the occurrence of tacit knowledge during PPRs. As tacit knowledge is hard to identify, the usage of metaphors and stories was taken as a proxy measure. (The use of metaphors and stories as a proxy measure of tacit knowledge in NPD may be controversial and this will be discussed later.) Based on the way the organizational learning literature stresses the importance of social interaction, the data were also coded for such indications. These included looking for references to the way meetings were held, the atmosphere necessary, the company culture required and the like. Internal validity refers to the reliability of a case study and whether the variables chosen for investigation are sufficient to explain the topic under investigation (Dane, 1990). Therefore, in order to maximize internal validity, multiple sources of data were used. Data sources The collection of data per case typically required five full but non-consecutive days of on-site visits. These were spread over approximately six months and took place during the years 2003-2005. These visits were used for collecting documents, conducting interviews and observing a PPR. Following the completion of each case, an extra visit gave specific feedback to the company on its processes and suggestions for improvements (the feedback component was not part of the formal research design but was important in gaining the cooperation of the companies). The research was conducted by a native German speaker with support from a native English speaker who speaks fluent German. Figure 1 shows the data sources and data collection techniques used for the research. This approach allowed a high degree of data triangulation between the three sources of data: company documents, interviews, and observations of PPRs.

Company documents. Copies of company confidential documents on PPR guidelines and the minutes of specific PPR meetings were obtained. A content analysis was conducted for PPR practices, documented lessons learned as well as metaphors and stories. The data coding recorded in documentation typically reflects explicit knowledge. Interviews with NPD personnel. At each company six interviews were conducted with project participants with different levels of experience in the respective company. All interviewees were managers of NPD projects or were actively involved in NPD. The first part of each interview was based on the repertory grid technique. This technique is especially useful in exploratory research settings (Goffin, 2002) and has been used previously to understand supplier-manufacturer relationships (Lemke et al., 2003). The repertory grid technique stimulated interviewees to contrast different projects on which they had worked. This elicited the lessons that they perceived to have learned from previous NPD projects. The particular advantage of the repertory grid technique is that it forces the respondent to think deeply and probes their tacit knowledge. The semi-structured part of the interviews was based around the following six questions: (1) What do NPD professionals think of PPRs? (2) What do NPD managers want to achieve with a PPR? (3) How do PPRs support the learning from projects? (4) What is the ideal outcome of a PPR? (5) How should PPR results be disseminated? (6) What are possible alternatives to PPRs?

Managers’ perceptions of learning in NPD 55

PPR observation. An actual PPR was observed at four companies (one company refused permission) and this was analyzed using ideas from learning theory. Thus, meeting transcripts were analyzed with a particular focus on lessons learned, metaphors and stories as well as social interactions. Sample The sampling frame chosen was the 50 largest companies in Baden-Wu¨rttemberg, Germany. This is considered to be a leading high-tech region because it accounts for the highest number of patents and R&D investments per capita in Europe (Staatsministerium Baden Wu¨rttemberg, 2001). Companies from different sectors were chosen, so that direct competitors would not be in the sample, which was an important Data Sources

DOCUMENTS

INTERVIEWS

Repertory Grid

Semistructured

OBSERVATIONS

PPR Meeting

Data Collection PPR Guidelines

Minutes from PPRs

Rep. Grid Matrices

Interview Transcripts

Notes on Observations

Transcripts of Meetings

Figure 1. Multi-faceted design of case studies

IJOPM 27,1

56

aspect in gaining the co-operation and trust of the companies. The companies have been given disguised names and Table II summarizes the data collected at each site. Data analysis Case analysis was conducted by the authors in unison. It included cross-checking the data coding and ensuring that salient quotes, which often included slang and dialect, were translated appropriately into English. There were three stages. Within-case analysis. Data from each case were analyzed separately to give a complete picture of each company’s approach to PPRs. The same analysis framework was used for each case and the research design ensured that a high degree of data triangulation could be performed. Part of this analysis was the collation of interviewees’ answers to the semi-structured interview questions, looking for evidence of the importance of social interactions. Another vital part of the analysis was the coding of the minutes of PPRs, repertory grid data, and observation transcripts – in all of these the use of metaphors and stories was the proxy measure chosen. Data reduction. About 2-3 page case descriptions were written on each company. The descriptions were then submitted to informants at the case companies to prevent observer bias (Lincoln and Guba, 1985) and to establish the credibility of the interpretation (Wallendorf and Belk, 1989). Cross-case analysis. Comparisons across the companies were made, to determine where similarities and differences existed and to identify a number of “best practices” (Yin, 1994). These comparisons effectively organized the results into the characteristics of PPRs (reported in Koners and Goffin, 2007), and perceptions of learning (reported in this paper). Further analysis is planned on the typical lessons learned in PPRs and NPD in general. Results How do NPD professionals perceive PPRs? Overall, 30 interviewees expressed their personal views on PPRs in the semi-structured interviews. The insights gained from these interviews are presented in Table III. This includes the frequency of the different perceptions. Note that due to time constraints and the semi-structured character of the interviews, not all 30 interviewees provided answers to every question. Thus, the total number of responses (which is given in brackets) is always lower than 30. Furthermore, as some interviewees mentioned several different issues in response to one question, the sum of responses listed in Table III is, in some cases, higher than the overall total of interviewees given in brackets. To demonstrate the trail of evidence, we will discuss Table III in more detail. What do NPD professionals think of PPRs?. The opinions on PPRs were grouped into positive, negative and neutral statements. PPRs were perceived by many interviewees as an ideal occasion at which to discuss what happened during the project. PPRs were perceived positively, because the presence of the development team enables projects to be considered from different viewpoints and because they can help to avoid similar problems in subsequent projects. PPRs were also appreciated as a source of ideas for future projects and for giving formal closure to a project (thus avoiding projects just “fizzling out,” without a clear end). Other interviewees stated that PPRs provide an ideal opportunity to recognize and reward the performance of the team for the overall outcome of the project.

1. 2. 3. 4. 5.

Engineering Co. Appliances Co. Medcare Co. Machinery Co. Publishing Co.

Case study No.

. 1 billion euro 1.5 billion euro 1.3 billion euro 387 million euro 280 million euro

Turnover 5.000 7.000 10.000 2.600 1000

Employees Yes Yes Yes Yes Yes

PPR guidelines (NPD process documentation) Yes, five sets copied Yes, three sets copied Yes, four sets on-site inspected Yes, four sets on-site inspected Yes, three sets copied 19 minutes

Minutes of specific PPRs

Yes – 6 Yes – 6 Yes – 6 Yes – 6 Yes – 6 30 interviews

Interviews

Yes Yes Yes Yes No (access refused) Four observations

Observation of a PPR

Managers’ perceptions of learning in NPD 57

Table II. Overview of case study data

Judgements of PPRs

Personal objectives of PPRs

PPRs and learning

Outcomes of PPRs

Dissemination of results

Alternatives to PPRs

2.

3.

4.

5.

6.

Table III. Summary of empirical results and recommendations 15 (of 29) suggest social interactions 9 (of 29) mention different forms of meetings as alternatives 7 (of 29) highlight database alternatives

19 (of 28) prefer social interactions 10 (of 28) mention documents as dissemination tools

15 (of 27) highlight documents like minutes or guidelines 10 (of 27) propose action based results like checklists and open action points 8 (of 27) prefer social interactions

13 (of 23) focus on evaluation and reflection 9 (of 23) emphasize learning 7 (of 23) suggest to find improvement suggestions 23 (of 28) view PPR discussion as a trigger for generating personal reflection and shared insights 2 (of 28) highlight that PPRs do not solve problems and are highly dependent from the company culture 3 (of 28) neutral comments that only future projects can show the learning effect

15 (out of 28) positive perceptions that PPRs cause team reflection and learning 9(out of 28) negative comments that PPRs do not take place or are missing top management support 4 (out of 28) neutral comments that PPR value depends on participants, culture, etc.

Conduct PPRs for all projects Communicate PPRs as the final and accepted project milestone. Extend the circle of participants outside the core project team and include all project experts. Support PPR importance with (partial) top management presence and combination of PPR with a social event Confirm PPRs as an important learning event. Enable open atmosphere and foster constructive criticism and learning culture within the R&D organization Prepare, structure and manage the PPR discussion in an optimal way, e.g. with the help of a moderator and visual aids. Allow sufficient time for in-depth brainstorming and reflection. Provide a supporting company culture that supports open discussions PPRs are a clear opportunity to achieve project-to-project learning Make sure that minutes are written for all PPRs Highlight the need to document action items and improvement suggestions. Recognize tacit results from PPRs and their value to the organization Support exchange of tacit knowledge and experience via informal networks, job rotation and internal presentations. Enable general access to minutes of PPRs and organized integration of PPR insights into guidelines and handbooks Ideas for other ways of sharing learning can augment PPRs – they do not have to be strict alternatives Arrange regular meetings of project managers Establish presentations of key experiences across projects and business units. Support close physical proximity between senior and junior staff so that “stories” can be transferred Introduce godfather programmes or coaching processes for junior project managers Allocate clear responsibility for the update of databases and guidelines

Practical recommendations based on responses

58

1.

Empirical results

IJOPM 27,1

A perceived negative aspect of PPRs was the problem of time constraints. As NPD organizations generally face increasing pressure to come up with new innovative products, several interviewees mentioned the high pressure to finish a project and to move quickly to the next rather than spending time on a PPR. Neutral statements included that the effectiveness of a PPR is very dependent on top management support and the culture of the NPD team. If the atmosphere within the team and within the NPD organization is not open to constructive criticism, then the outcome of a PPR will be limited. Thus, it appears to be important that senior management clearly establish PPRs for all NPD projects and for them to be at least partially present for some of the time at PPRs. What do NPD managers want to achieve with a PPR?. The objectives that interviewees want to achieve from PPRs were found to fit into three categories; evaluation, learning, and improvement. The objective most frequently mentioned was that a PPR enables an objective evaluation of a project after the product has been launched. In this evaluation, every aspect of a project can be considered from different viewpoints and with an immediate exchange of ideas. Some interviewees indicated that they want to fully evaluate projects during PPRs. This means not only looking at the project outcome and its quality, but also referring back to the original objectives of the project and analyzing if these were met and to what degree. (These comments indicate that some PPRs are perceived to focus too tightly on project outputs, rather than the process.) It was perceived that project learning can result from either positive or negative experiences, from mistakes or from successful practices which were applied. Often, the objective of NPD professionals is not only to learn but also to achieve improvements by applying the lessons learnt to future projects. How do PPRs support the learning from projects?. Responses regarding how PPRs support learning were grouped into positive and negative statements. The majority – 23 interviewees – were positive about the role of PPRs in supporting learning. PPRs were perceived to facilitate the exchange of experience, because the discussion raises different viewpoints, which individuals may not have recognized on their own. This is achieved in that a PPR usually brings together the whole project team, and not only the core team. Thus, the pool of experience available is higher than in many project meetings where certain sub-teams discuss specific issues. Another positive aspect was that PPRs trigger personal reflections and lead to brainstorming within the team which does not happen during other meetings: There are certain issues that only come up at the end in the review, because only then you have the time and peace of mind to actually think about causes and consequences (Interviewee 2, Appliances Co.).

It was also mentioned that the discussion during a PPR enables more transfer of know-how within the team than anything that might be formally documented and stored in databases. “During the discussion the real important points emerge within the team – you will never find these points in minutes or databases” (Interviewee 7, Appliances Co.). Only two interviewees expressed clearly negative perceptions regarding the role of PPRs in supporting learning. One said that PPRs have limited utility in that they discuss problems and potential improvements but action issues are very seldom followed-up in a professional way. The second negative view was that PPRs are only

Managers’ perceptions of learning in NPD 59

IJOPM 27,1

60

effective if the company culture supports open discussion. Without this, PPRs were perceived as not being able to contribute to a learning organization at all. Neutral comments stressed that only the results of future projects would really show whether PPRs had resulted in lessons learned within the organization. Overall, the majority of interviewees perceive the role of PPRs in supporting learning positively. Yet, it is also important to stress that learning depends to a high degree on the way the discussion is structured and moderated, the time a company invests for in-depth reflection during a PPR and the underlying company culture (this links to many of the points in Table I). What is the ideal outcome of a PPR?. Responses regarding the outcomes of PPRs were found to fit into three categories; documents, action points and social interaction. Overall, 15 interviewees thought that written summaries were an ideal outcome and mentioned minutes of PPRs, internal guidelines and checklists as such outputs. It is interesting to note that three of the 15 interviewees mentioned that only a document-based approach would work in their organization. For example, at MedCare Co. there are no personal relationships with colleagues in other business units and so documents are essential. A project manager at Publishing Co. recommended documents as the preferable outcome of a PPR, because the company culture did not support personal and social interactions as a way to transfer learning. It is interesting to note that the interviewees did not recognize the limitations of documents (for example, tacit knowledge cannot be disseminated by documents). This means that many interviewees do not have an understanding of the different types of knowledge that can be created. Ten interviewees went further than mentioning documents and expressed the view that identifying specific action points is important. Such action points should be allocated to a specific person to be completed by a specific date. The action points can be general improvement suggestions, the most important aspects to be considered in future projects or a checklist of remaining actions – all these are also considered to be a vital outcome of PPRs. In addition, eight interviewees recognized the problems of sharing results and suggested social interactions as the optimal outcome of a PPR. They emphasised that the outcomes of PPRs cannot be documented in an efficient way and that the team discussion during the PPR is most important. “I always prefer to do personal presentations after the PPR took place, because only the interaction between people can really transfer the knowledge gained during the meeting” (Interviewee 4, Machinery Co.). How should PPR results be disseminated?. The responses about the dissemination of the results of PPRs were found to fit into two categories; social interaction and documents. A majority of responses (19 out of 28) stressed that results should ideally be disseminated informally through social interactions, since this is easier and more efficient than reading the minutes of PPRs. It was stressed that the best dissemination of the results of PPRs takes place when the participants apply their learning to subsequent projects. Alternatively, the results could also be given to a steering committee, with responsibility to disseminate the results within the wider organization. Other interviewees proposed presentations where PPR results and “stories” could be verbally presented by the project manager or team members to other members of the NPD organization. Some interviewees suggested the use of documents as part of the dissemination of the results of PPRs and said the project team plus senior managers should receive the minutes of PPRs. One of the perceived critical success

factors for such a document, though, is that it should not be too long, because the problem of information overload was mentioned several times. Finally, interviewees suggested to integrate the results of PPRs into official NPD guidelines and handbooks. Such an approach was perceived to make it more likely that the experiences would be applied to future projects. What are possible alternatives to PPRs?. The mechanism most often mentioned as a potential alternative to PPRs was social interaction, i.e. any kind of personal relationship within a department, project group or company which enables the informal exchange of ideas and experiences. For example, interviewees said that senior employees often gain a reputation for their know-how in a very particular area and are consulted by colleagues if questions in this area arise. One factor that supports these informal networks is the co-location of experienced colleagues with younger members of staff. Informal networks and personal discussion can work well and it was mentioned that they provide the easiest and quickest way to find answers to complex questions. However, it was mentioned that one-to-one discussions, are on the one hand, highly focused, but on the other hand they do not provide the same kind of insights as a PPR, because these combine the perspectives of the whole project team. Another mechanism mentioned was godfathers – experienced project managers assigned to meet weekly with junior colleagues in order to discuss their current issues and answer questions (a formal way to stimulate the creation of networks). An alternative to PPRs which is based on social interaction is meetings of project managers. The frequency of these meetings, however, varied a lot between the case companies. Some have adopted the practice of meeting each week, some every two or three months and some only once or twice a year. Interviewees also perceived coaching meetings to be useful. In these, an experienced project manager discusses example problems with a group of junior members of staff in an informal session. Three out of the five case companies actively use databases to store project information and experiences. However, databases are most often used to store technical data like quality issues and how they were solved rather than the storage of experiences. Overall, the mechanisms mentioned were perceived not as strict alternatives, but as ideas that can augment PPRs. The semi-structured interviews showed that social interactions are perceived as important for the transfer of learning within NPD organizations. Table III includes not only the frequency of comments but also gives practical recommendations based on the interviews. It can be seen that many of these are related to social interactions (as indicated by italic type). However, in addition to stressing the importance of interaction, the organizational learning literature emphasizes tacit knowledge. This was investigated by looking at a range of different data from the case studies. Is tacit knowledge created during PPRs? Since, identification of the use of metaphors and stories has been recognized as providing evidence for the creation and transfer of tacit knowledge, the data were coded accordingly. As shown by Table II, the data consisted of the official minutes of 19 PPRs at the case companies, the results of 30 repertory grid interviews, and the transcripts and notes from direct observation of four PPRs. Table IV shows that a total of 94 metaphors and stories were identified (4 in the minutes, 35 in the repertory grids and 55 were observed in the PPRs). For example,

Managers’ perceptions of learning in NPD 61

IJOPM 27,1

62 Table IV. Identification of metaphors and stories

Case no

Case name

1. 2. 3. 4. 5.

Engineering Co. Appliances Co. MedCare Co. Machinery Co. Publishing Co. Total

Metaphors found in minutes of PPRs 2 1 1 4

Number of minutes inspected

Metaphors used during repertory grids

Metaphors mentioned during PPR observations

PPR length (hours)

5 3 4 4 3 19

5 12 5 3 10 35

14 30 6 5 N/a 55

2.5 7.5 2 3 N/a 15

“We always had clear rules of play in our team” was found in the minutes of a PPR from Appliances Co. in the summary of a discussion about the positive aspects of a project. In a repertory grid interview a respondent said: In the past we had a marketing silo and a technical department silo and we have thrown our not very well defined wishes over the wall to the other silo and what we got in return was not what we wanted” (Repertory grid interviewee 5, MedCare Co.).

Several metaphors and stories observed during PPRs were clear to the participants, but not to the researcher(s) present. In these cases, the discussion before or after a specific metaphor or story was used was hard for the researchers to understand. For example, during the PPR at Appliances Co. one of the participants mentioned that: “He [the project manager] was almost like a sheepdog and kept circling the project team like a herd of sheep.” Some minutes later in the discussion, another participant of the PPR then referred back to this statement and said “. . . yes, this is again the example with the herd of sheep.” Although the metaphor was repeated, the researcher was not clear of the meaning, although it had obviously been understood by the whole team. At that point, one of the participants quietly explained to the researcher that the metaphor referred to the project manager’s keenness to successfully meet the project objectives and how he had monitored progress very closely throughout the project. Another example of the intimate nature of the discussions during PPRs was the use of a metaphor observed at Appliances Co.: “It like was a game with moles. You hit one on the head and somewhere else four or five other ones appear.” This particular metaphor caused a lot of laughter amongst the PPR participants but it was not clear to what it referred. Thus, the researcher asked the project manager after the PPR about the background of this metaphor. He then explained that the moles were metaphor for problems, i.e. as soon as one problem was solved by the team, then several others arose straightaway. This metaphor indicates that project teams may develop their own vocabulary (moles ¼ problems), which is intricately linked with the transfer of tacit knowledge. Metaphors and stories also appeared to enliven and bring humor to discussions about complex technological issues, by making them easier to understand for those without detailed knowledge. The observation of four PPRs enabled the researchers to recognize that the atmosphere at a PPR appears to influence the outcome. At MedCare Co., the PPR took place in a very formal and serious manner and did not include a lot of interaction between the participants. In contrast, the PPR of Machinery Co. took place in a beer

garden and was run in a very informal manner. During the observation, the researcher noted that often a single metaphor or story was enough to trigger a long discussion. Table IV compares different types of data and thus gives a new perspective on learning from PPRs in NPD. The official minutes of PPRs were found to include only four metaphors and stories, whereas in the PPRs that were observed, metaphors and stories were used 55 times – on average every 16 minutes. Therefore, it appears that the social interactions in a PPR lead to the use of metaphors and stories that are not then documented in the minutes. (By omitting the metaphors and stories from minutes, companies may be missing opportunities for disseminating knowledge.) In the semi-structured interviews, many respondents indicated that they found interaction to be very important. However, the importance of tacit knowledge was not mentioned, perhaps indicating that the respondents were unaware of this concept. Obviously, it would have been inappropriate to ask them whether they perceived tacit knowledge to be important and so the repertory grid technique was used to identify whether individuals’ learning from NPD involved the generation of tacit knowledge. The repertory grid technique is known for probing deep into individuals’ knowledge and so the transcripts for this part of the interviews were coded for the use of metaphors and stories. As shown in Table IV, the transcripts of the repertory grid interviews contain a total of 35 metaphors and stories. This indicates that working on NPD projects develops tacit knowledge. Overall, it is evident that metaphors are used by interviewees in PPR discussions (as observed), but are almost entirely absent from the minutes produced. The repertory grid interviews indicate that individuals generate tacit knowledge from working on NPD. However, the social interaction between the experts present at a PPR appears to give the most support to the creation of tacit knowledge. Discussion and conclusions Overall, most NPD personnel expressed a positive personal attitude towards PPRs and their impact on learning. This is interesting when considering that most R&D organizations do not conduct PPRs. Within the limitation of the current exploratory research, it was found that: . NPD professionals who take part in PPRs do perceive them to be a useful mechanism for promoting individual and team learning in NPD (based on evidence from Table III). . Taking part in the discussions at PPRs is perceived to trigger insights that are of more value than reading what is documented in minutes, or stored databases. This can be clearly related to social learning theory. A comparison of the frequency of usage of metaphors and stories in the minutes of PPRs and in actual meetings provides support for NPD professionals’ perceptions. It appears that the social interactions at PPRs do stimulate the creation and exchange of tacit knowledge (based on evidence from Tables III and IV). . Documents like the minutes of PPRs and checklists are an important means of disseminating knowledge but they do not always contain all the aspects from the discussions (based on Table III). . The importance of social interactions for learning is stressed in the literature and this was matched by the perceptions of NPD professionals and observations of

Managers’ perceptions of learning in NPD 63

IJOPM 27,1

64

PPR discussions, particularly those held in informal settings (based on Tables III and IV). . The results indicate that metaphors and stories form an important part of PPR discussions and are used (consciously or subconsciously) to stimulate or summarize key points, for example, related to technical or project management issues (based on evidence from Table IV). Limitations and recommendations for researchers As with all exploratory research, this study has significant limitations. These are linked to the necessity of having to try and identify both the key constructs and suitable measures for them. The approach chosen to investigate tacit knowledge may be controversial to some readers, because one could argue that tacit knowledge cannot be operationalized. However, we argue that tacit knowledge should not only be discussed at a purely theoretical level. This paper intends to provide a first step in that it looked for empirical data on tacit knowledge in NPD. Unfortunately, as comparable data does not exist, it is not possible to judge whether the frequency of usage of metaphors and stories is particularly high, or not. Furthermore, the sample of five German companies is naturally not representative. However, as an exploratory sample, it allows our understanding of PPRs to be advanced. Later research will need to look to wider samples to establish externally valid results. In Table V we summarize our conclusions, showing the approaches we took and how we think they can and should be enhanced in future research. For example, we relied on NPD professionals’ perceptions of the importance of social interactions and think that additional approaches, such as longitudinal (ethnographic) observation will be necessary to gain a real understanding of interactions and learning. Similarly, the subjective analysis that the atmosphere of PPRs influences the results needs refining. Tacit knowledge was investigated using a proxy measure – the frequency of usage of metaphors and stories. A categorization of the types of metaphors and stories used Construct

Table V. Suggested additional approaches

Approaches used in this study

Suggested additional approaches

1. Social interaction

NPD professionals’ perceptions of its importance Subjective assessment of the atmosphere at PPRs by the researchers

2. Tacit knowledge generation

Frequency of use of metaphors and stories in documents, repertory grid interviews and in observed PPRs

Longitudinal observation of how NPD professionals interact: frequency, depth, etc. Need to measure the formality of PPR meetings in a more objective way Semantic analysis of how metaphors and stories are used – to develop an understanding of the types of metaphors and stories used NPD professionals’ perceptions of what they have learnt from the usage of specific metaphors and stories Identification of how metaphors and stories are used outside of PPRs (e.g. to pass learning to other project teams) Comparison of individuals’ perceptions of what they have learnt from a project before and after a PPR

by NPD teams is needed, as is an understanding of what individuals perceive they have learnt from them. Researchers need to rise to the challenge and attempt to understand the role of social interactions and tacit knowledge in NPD better – by moving to develop effective approaches as suggested by Table V. In addition, to focusing on operational measures related to learning, there is an urgent need for more research on PPRs and learning in NPD. Such research needs to fully consider organizational learning theory. In particular, the following topics need investigation: . How many companies are currently using PPRs for stimulating learning in NPD? Here a representative survey is needed. . A large-scale survey is also needed to gather generalizable data of NPD professionals’ perceptions of their PPR, in order to establish whether the findings from this exploratory research are representative. . What are the lessons that NPD professionals learn from working on NPD? . What would be suitable performance measures to determine the effectiveness of PPRs? . How do PPRs support project-to-project learning? For practitioners Based on the research results, a number of recommendations can be made which are all targeted at the improvement of learning in NPD: . The importance of PPRs needs to be supported by top management, in order to ensure that they take place, have the right atmosphere, and company culture encourages constructive discussions. . The time and effort invested in PPRs can bring better returns if the knowledge is disseminated to other project teams. Managers need to actively support this process, for example, by job rotation, the creation of project teams with experienced as well as junior members, via informal networks, and through internal presentations. . Typical lessons learned should be disseminated to other project teams in an easily assimilated form, e.g. by presentations, checklists, and short minutes of PPRs. . Tacit knowledge needs to be addressed. For example, PPRs could be combined with a “social event,” in order to celebrate the team’s achievements. The term social event can cover anything from a visit to a museum (Appliances Co.), or a dinner (Engineering Co.). These events could help in the generation and communication of knowledge. . PPR discussions can be based around the use of metaphors and stories, as these appear to be important vehicles for the transfer of tacit knowledge within NPD teams. Overall, the research gives new insights into learning in NPD teams. It demonstrates the potential of PPR’s to generate knowledge and indicates the importance of interaction between NPD professionals to transfer knowledge. Minutes and databases of lessons learned have their place but as one interviewee said, “You cannot really write down experiences, even if you try. This is almost impossible and it would be a huge book” (Interviewee 2, Publishing Co.).

Managers’ perceptions of learning in NPD 65

IJOPM 27,1

66

References Ayas, K. (1996), “Design for learning and innovation”, Long Range Planning, Vol. 29 No. 6, pp. 898-906. Baird, L., Holland, P. and Deacon, S. (1999), “Learning from action: imbedding more learning into the performance fast enough to make a difference”, Organizational Dynamics, Vol. 27 No. 4, pp. 19-31. Balthazor, L.R. (1994), “Project review – do you really know where you are?”, The Aeronautical Journal, Vol. 98 No. 973. Boag, D.A. and Rinholm, B.L. (1989), “New product management practices of small high technology firms”, Journal of Product Innovation Management, Vol. 6 No. 2, pp. 109-22. Boudes, T., Charne-Dubac, F. and Midler, C. (1998), “Project management learning: a contingent approach”, in Lundin, R.A. and Midler, C. (Eds), Projects as Arenas for Renewal and Learning Processes, Kluwer Academic Publishers, Dordrecht, pp. 61-70. Bourgault, M. and Sicotte, H. (1998), “Learning conditions and performance of development projects: empirical evidence from a research center”, Proceedings of the 29th Annual Project Management Institute 1998 Seminars & Symposium, Long Beach, California, October 9-15. Bowen, H.K., Clark, K.B., Holloway, C.A. and Wheelwright, S.C. (1994), “Make projects the school for leaders”, Harvard Business Review, Vol. 72 No. 5, pp. 131-40. Busby, J.S. (1999), “An assessment of post-project reviews”, Project Management Journal, Vol. 30 No. 3, pp. 23-9. Chiva, R. and Alegre, J. (2005), “Organizational learning and organizational knowledge. Towards the integration of two approaches”, Management Learning, Vol. 36 No. 1, pp. 49-68. Cook, S.D.N. and Brown, J.S. (1999), “Bridging epistemologies: the generative dance between organizational knowledge and organizational knowing”, Organization Science, Vol. 10 No. 4, pp. 381-400. Corso, M., Martini, A., Paolucci, E. and Pellegrini, L. (2001), “Knowledge management in product innovation: an interpretative review”, International Journal of Management Review, Vol. 3 No. 4, pp. 341-52. Crossan, M., Lane, H. and White, R. (1999), “An organizational learning framework: from intuition to institution”, Academy of Management Review, Vol. 24 No. 3, pp. 522-37. Dane, F.C. (1990), Research Methods, Brookes/Cole, Pacific Groove, CA. Duarte, D. and Snyder, N. (1997), “From experience: facilitating global organizational learning in product development at Whirlpool corporation”, Journal of Product Innovation Management, Vol. 14 No. 1, pp. 48-55. Durrance, B. (1998), “Some explicit thoughts on tacit learning”, Training & Development, Vol. 52 No. 12, pp. 24-30. Eisenhard, K.M. (1989), “Building theories from case study research”, Academy of Management Review, Vol. 14 No. 4, pp. 532-50. Freedman, D.P. and Weinberg, G.M. (1977), Handbook of Walkthroughs, Inspections and Technical Reviews, Evaluating Programs, Projects and Products, 3rd ed., Little, Brown and Company, Boston, MA. Gherardi, S. (2000), “Practice – based theorizing on learning and knowing in organizations”, Organization, Vol. 7 No. 2, pp. 211-23. Goffin, K. (2002), “Repertory grid technique”, in Partington, D. (Ed.), Essential Skills for Management Research, Sage, London, pp. 198-225.

Goffin, K. and Pfeiffer, R. (1999), “Innovation management in UK and German manufacturing companies”, Anglo-German Foundation Report Series, ISBN 1-900834-17-0, 70 pp, London. Gulliver, F.R. (1987), “Post-project appraisals pay”, Harvard Business Review, Vol. 87 No. 2, pp. 128-32. Hartley, J.F. (1994), “Case studies in organizational research”, in Cassell, C. (Ed.), Qualitative Methods in Organizational Research: A Practical Guide, Sage, London. Holtshouse, D. (1999), “Ten knowledge domains: model of a knowledge-driven company”, Knowledge and Process Management, Vol. 6 No. 1, pp. 3-8. Howells, J. (1996), “Tacit knowledge, innovation and technology transfer”, Technology Analysis & Strategic Management, Vol. 8 No. 2, pp. 91-106. Huber, G.P. (1996), “Organizational learning: a guide for executives in technology-critical organizations”, International Journal of Technology Management, Vol. 11 Nos 7/8, Special Publication on Unlearning and Learning, pp. 821-32. Johannessen, J-A., Olaisen, J. and Olsen, B. (2001), “Mismanagement of tacit knowledge: the importance of tacit knowledge, the danger of information technology, and what to do about it”, International Journal of Information Management, Vol. 21 No. 1, pp. 3-20. Koners, U. and Goffin, K. (2005), “Learning from new product development projects: an exploratory study”, Creativity and Innovation Management, Vol. 14 No. 4, pp. 334-44. Koners, U. and Goffin, K. (2007), “Learning from post-project reviews: a cross-case analysis”, Journal of Product Innovation, forthcoming. Kotnour, T. (1999), “A learning framework for project management”, Project Management Journal, Vol. 30 No. 2, pp. 32-3. Lam, A. (2000), “Tacit knowledge, organizational learning and societal institutions: an integrated framework”, Organization Studies, Vol. 21 No. 3, pp. 487-513. Lane, K. (Ed.) (2000), “Project management today”, available at: www.projectnet.com (accessed February 2, 2000). Lemke, F., Goffin, K. and Szwejczewski, M. (2003), “Investigating the meaning of supplier-manufacturer partnerships: an exploratory study”, International Journal of Physical Distribution & Logistics Management, Vol. 33 No. 1, pp. 12-35. Leonard-Barton, D. (1992), “The factory as a learning laboratory”, Sloan Management Review, Vol. 34 No. 1, pp. 23-8. Lilly, B. and Porter, T. (2003), “Improvement reviews in new product development”, R&D Management, Vol. 33 No. 3, pp. 285-96. Lincoln, Y.S. and Guba, E. (1985), Naturalistic Enquiry, Sage, Beverly Hills, CA. Lindkvist, B. (2001), “Kunskapso¨verfo¨ring mellan produktutvecklingsprojekt”, PhD thesis, Stockholm School of Economics, Stockholm. Liyanage, S., Greenfield, P.F. and Dan, R. (1999), “Towards a fourth generation R&D management model – research networks in knowledge management”, International Journal of Technology Management, Vol. 18 Nos 3/4, pp. 372-93. McKee, C. (1992), “An organizational learning approach to product innovation”, Journal of Product Innovation Management, Vol. 9 No. 3, pp. 232-45. Nonaka, I. (1994), “A dynamic theory of organizational knowledge creation”, Organization Science, Vol. 5 No. 1, pp. 14-37. Pitman, B. (1991), “A systems analysis approach to reviewing completed projects”, Journal of Systems Management, Vol. 42 No. 6. Polanyi, M. (1962), Personal Knowledge – Towards a Post-critical Philosophy, Routledge, London.

Managers’ perceptions of learning in NPD 67

IJOPM 27,1

68

Prahalad, C.K. and Hamel, G. (1990), “The core competence of the corporation”, Harvard Business Review, Vol. 68 No. 3, p. 79. Saban, K., Lamosa, J., Lackman, C. and Peace, G. (2000), “Organizational learning: a critical component to new product development”, Journal of Product & Brand Management, Vol. 9 No. 2, pp. 99-119. Schindler, M. and Eppler, M.J. (2003), “Harvesting project knowledge: a review of project learning methods and success factors”, International Journal of Project Management, Vol. 21 No. 3, pp. 219-28. Sense, A.J. and Antoni, M. (2003), “Exploring the politics of project learning”, International Journal of Project Management, Vol. 21 No. 7, pp. 487-94. Smith, P.G. (1996), “Your product development process demands ongoing improvement”, Research Technology Management, Vol. 39 No. 2, pp. 37-44. Staatsministerium Baden Wu¨rttemberg (2001), available at: www.baden-wu¨rttemberg.de (accessed May 22, 2001). Thomke, S. and Fujimoto, T. (2000), “The effect of front-loading problem-solving on product development performance”, Journal of Product Innovation Management, Vol. 17 No. 2, pp. 128-42. Tidd, J., Bessant, J. and Pavitt, K. (1997), Managing Innovation: Integrating Technological, Market and Organizational Change, Wiley, Chichester. von Krogh, G. (1998), “Care in knowledge creation”, California Management Review, Vol. 40 No. 3, pp. 133-53. von Zedtwitz, M. (2003), “Post-project reviews in R&D”, Research Technology Management, Vol. 46 No. 5, pp. 4-49. Wallendorf, M. and Belk, R.W. (1989), “Assessing trustworthiness in naturalistic consumer research”, in Hirschman, E.C. (Ed.), Interpretive Consumer Research, Association for Consumer Research, Provo, UT. Weinberg, G.M. and Freedman, D.P. (1984), “Reviews, walkthroughs and inspections”, IEEE Transactions on Software Engineering, Vol. 10 No. 1. Wenger, E. and Snyder, W.M. (2000), “Communities of practice: the organizational frontier”, Harvard Business Review, Vol. 78 No. 1, pp. 139-45. Wheelwright, S.C. and Clark, K.B. (1992), Revolutionizing Product Development: QUANTUM Leaps in Speed, Efficiency and Quality, The Free Press, New York, NY. Wong, W.L.P. and Radcliffe, D.F. (2000), “The tacit nature of design knowledge”, Technology Analysis & Strategic Management, Vol. 12 No. 4, pp. 493-512. Yin, R.K. (1994), Case Study Research: Design and Methods, 2nd ed., Applied Social Research Methods Series,Vol. 5, Sage, London. Further reading Reed, N. (2000), “Personal construct psychology and knowledge management in organizations”, unpublished dissertation, submitted in partial satisfaction of the requirements for the Diploma in Application of Personal Construct Psychology. Corresponding author Keith Goffin can be contacted at: [email protected] To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0144-3577.htm

How do subcontractors evolve? Andrea Furlan, Roberto Grandinetti and Arnaldo Camuffo

How do subcontractors evolve?

Department of Economics, University of Padova, Padova, Italy Abstract

69

Purpose – The purpose of this study is to investigate how small and medium sized subcontracting firms evolve. Design/methodology/approach – A cluster analysis was applied to a sample of 417 North East Italian subcontractors to explore if (and to what extent) Italian subcontractors differ and can be classified on the basis of their design and marketing capabilities. Using this classification as a starting point, multiple case study analysis is conducted on a sample of ten subcontractors and a model developed of how subcontractors’ capabilities evolve over time. Findings – Four profiles of subcontractors are identified as a function of their design and marketing capabilities: developed, developing, question mark and traditional. A model is proposed to understand and predict subcontractors’ evolution. In the model knowledge codification, supply management, design and marketing capabilities mutually reinforce one another and tend to align over time. Research limitations/implications – Firstly, future research should articulate the four clusters identified. Secondly, the framework for subcontractors’ evolution should be tested on large-scale databases. Thirdly, more accurate measures of subcontractors’ capabilities should be conceived and tested. Practical implications – Results of this study are critical for industrial buyers who need to segment their subcontractors and understand how their marketing and design capabilities evolve. Moreover, they are also critical for subcontractors’ managers who wish to avoid cost-based strategies, enlarge their customer base, broaden their international scope and engage in durable relationships with their customers. Originality/value – This study proposes an original model of subcontractors’ classification and evolution and suggests good practices to design and manage supply networks. Keywords Subcontracting, Design management, Marketing management Paper type Research paper

Introduction The role subcontractors play in supply systems is a research topic located at the crossroads of a variety of managerial and economic disciplines such as strategic management, operations and supply chain management, organizational design and industrial organization. Such a role, and its dynamics, is especially important in the Italian economy, known worldwide for its high fragmentation, its organization around geographically coupled supply systems (industrial districts) and the prevalence of small and medium sized firms, vertically specialized in one or few phases of a supply chain (Sabel and Piore, 1984; Porter, 1990). Challenged by globalization and new technologies, in recent years these supply systems have often lost their competitive edge and undergone major structural and strategic changes, partly losing their historical peculiarities (Berger and Locke, 2001; Grandinetti and Bortoluzzi, 2004). Firstly, increasing competition from producers located in low cost countries and ever new, more powerful information and communication technologies have reduced the importance of geographical proximity as a competitive advantage factor. Secondly, globalization and the related risks ask for financial structures and managerial capabilities not easily accessible and adoptable by district firms, which are mostly

International Journal of Operations & Production Management Vol. 27 No. 1, 2007 pp. 69-89 q Emerald Group Publishing Limited 0144-3577 DOI 10.1108/01443570710714547

IJOPM 27,1

70

undercapitalized, family-owned and run businesses. Finally, while manufacturing, built on a heritage of craftsmanship and skilled labor, has historically been district firms’ core competence, marketing and design capabilities have been neglected and are underdeveloped. Now that innovation and internationalization are key success factors, competencies other than manufacturing efficiency and flexibility have become critical. To address these new challenges, subcontractors have to change their strategies and adapt their structures. Above all, they need to develop appropriate design and marketing capabilities (Esposito and Lo Storto, 1991; Nassimbeni et al., 1996). As concerns design, subcontractors need to take on more responsibilities, abandon a mere productive role and undertake proactive behaviors towards customers. Subcontractors are required to develop specific product and/or process technologies, so that they can be incorporated in the customers’ products and/or processes and become integrative part of their value propositions. As concerns marketing, subcontractors have to be able to offer innovative products, processes or services, to navigate autonomously the global market, to position in new, more defendable niches, to find out new segments and opportunities, to serve and manage more demanding and differentiated customers, and to develop their own brands. Marketing capabilities (MC) can alleviate subcontractors’ dependence from local demand and facilitate diversification and internationalization. Design capabilities (DC) are the pre-requisite to establish more valuable and balanced relationships with a wider array of customers. Within this complex and worrisome situation, common to other European countries than Italy, this paper asks the following research question is: RQ1. How are small and medium sized subcontracting firms changing as globalization and new technologies challenge traditional supply systems in mature industries? Such a question is critical for industrial buyers who, in order to improve supply chain performance and match sourcing with their business strategies, need to segment their subcontractors and understand how, on top of manufacturing excellence and beyond cost efficiency, their marketing and design capabilities evolve. But it is also critical for subcontractors who, in order to survive, need to avoid strategies based on mere cost competition, enlarge their customer base, offer unique, higher quality products, serve their customers with ever higher degrees of flexibility and engage them in durable relationships. We address this question studying the case of Italian subcontractors. Applying cluster analysis to a sample of 417 North East Italian subcontractors, we explore if (and to what extent) Italian subcontractors differ and can be classified on the basis of their design and marketing capabilities. We identify four profiles of subcontractors as a function of their design and marketing capabilities: “developed” “developing” “question mark” and “traditional”. Using this segmentation as a starting point, we also use data from ten in-depth subcontractors’ case studies and propose a model of how subcontractors’ capabilities develop over time. We identify some typical subcontractors’ evolutionary patterns, showing how the transition from one profile to another can take place.

The paper is organized as follows. The second section presents a review of the literature on subcontractors’ segmentation and evolution. Third section proposes a subcontractors’ classification based on the results of the cluster analysis. Fourth section illustrates the findings of the multiple case study analysis and proposes a model of how subcontractors’ capabilities develop over time. Fifth section discusses our findings’ implications in terms of the resource-based and knowledge-based views of the firm. It also suggests managerial guidelines about how industrial buyers should approach subcontractors’ strategic segmentation and how subcontractors could develop design and marketing capabilities. Theoretical background Subcontractors’ strategies, structures and role within supply system have been approached from two different perspectives. The first perspective takes a static approach and classifies subcontractors on the basis of industry-wide, company-specific or relational characteristics. The second perspective takes a dynamic approach and proposes evolutionary models of subcontractors and of the industries they operate in. This section provides a literature review for each of the two perspectives. The static perspective: subcontractors’ segmentation In purchasing and supply management literature, authors often build on the concept of supplier segmentation based on industry, suppliers’ or relational characteristics. Generally speaking, the segmentation of suppliers into different tiers of sourcing help to prioritize and differentiate supply management practices among different suppliers (Dyer et al., 1998; Lamming et al., 2000; Petroni and Panciroli, 2002; Vokurka et al., 1996). Kraljic’s (1983) seminal work uses two classification criteria to develop a portfolio model of suppliers. One is the importance of the purchased item, the other is the complexity of the supply market. Out of these two dimensions, he identifies four categories of materials and components: (1) non-critical; (2) leverage; (3) bottleneck; and (4) strategic. Since, Kraljic’s (1983) groundbreaking work, a number of studies have developed models of supplier segmentation. For example, Van Weele (2000) builds on Kraljic’s (1983) and Olsen and Ellram’s (1997) models and proposes two dimensions to segment suppliers: (1) supplier’s impacts on buyer’s financial results; and (2) how risky the supplier relation is. He derives four types of suppliers (strategic, leverage, bottleneck and routine suppliers) and associates to them four different sourcing strategies: (1) partnership with strategic suppliers (i.e. market leaders, specific know-how, and balance of power may differ among buyers-suppliers); (2) competitive bidding – leverage suppliers (i.e. many competitors, commodity products, and a buyer dominated segment);

How do subcontractors evolve? 71

IJOPM 27,1

72

(3) securing continuity of supply – bottleneck suppliers (i.e. technology leaders, few – if any – alternative suppliers); and (4) systems contracting – routine suppliers (i.e. large supply, many suppliers with dependent position, and a reduction in the number of suppliers). Studying relational contracts in the Japanese auto industry, Asanuma (1989) identifies three types of suppliers/subcontractors on the base of their involvement in the product development process of the core firm: drawings supplied vendors, drawings approved vendors and marketed goods vendors. Dyer et al. (1998) comparative analysis of supplier-automaker relationships in the USA, Japan and Korea suggests that firms have to think strategically about supplier management and abandon “one-size-fits-all” approaches. Rather, firms should strategically segment suppliers and assess how they contribute to the core competence and competitive advantage of the buying firm. Bensaou (1999) builds on relational contracting theory and develops a typology based on the nature of the supplier relationship. He uses the specific investments made by buyers and suppliers in the relationship as segmentation criteria, identifying four relationship categories: (1) market exchange; (2) captive buyer; (3) captive supplier; and (4) strategic partnership. Within this segmentation he considers suppliers’ technological capabilities as one of the three constituents (along with product characteristics and market characteristics) of the contextual profiles for the management of supplier relations. Masella and Rangone (2000), building on Nydick and Hill (1992) and Narasimhan (1983), develop a supplier segmentation system based on two dimensions: the time frame (i.e. short-term versus long-term) and the content (i.e. logistic versus strategic) of the buyer-supplier relationship. They identify four different types of suppliers, grounding this typology on an analytic hierarchy process framework, on different sets of measures, and on a resource-based view of suppliers’ strategy. These studies share the belief that, in order to be meaningful and predictive, suppliers’ segmentation ought to rest on criteria related to suppliers’ capabilities (rather than to other suppliers’ characteristics) and should specify the nature of the relationships (rather than of the purchase) to be developed. The dynamic perspective: subcontractors’ evolution Though the static perspective gives some conceptual and empirical tools that can be used for segmenting subcontractors, it fails to provide an understanding of the dynamics within and between different subcontractors’ segments. Indeed, comparatively fewer studies have focused on the topic of subcontractors’ evolution. Lamming’s (1993) groundbreaking work on suppliers’ evolution identifies nine factors (e.g. competitive pressure, adoption of information sharing and negotiation practices, supplier integration in new product development and logistics) that need to be considered in the analysis of buyer-supplier relationships. On the basis of these

factors, he proposes evolutionary patterns that mark the transition from traditional subcontracting to partnership relationships. Several Italian studies analyze subcontractors’ evolution, too. Some of them model the evolution of subcontractors analyzing the characteristics of the relations between them and their main contractors. For example, Zanoni (1991) identifies an evolutionary pattern (from traditional subcontracting into partnership) contingent on the evolution of some characteristics of the supply relationship (i.e. information sharing, technological integration, early supply involvement in new product development). Some others emphasize knowledge transfer as the main determinant of subcontractors’ evolution. For example, Esposito and Lo Storto (1991) first identify four components in the technological knowledge of a subcontractor: machines (e.g. computerized numerical control (CNC) tools), individual skills (e.g. design, engineering, testing), formal documentation (e.g. drawings, specs) and organizational routines (quality manuals and procedures); then they argue that changes in one or more of these components entail the evolution of the subcontractor. Esposito and Raffa (1994) subsequently develop this point, suggesting that several knowledge transfer mechanisms (technical help, on-site support, resident engineers, training, exchange of document/equipment/machines) can induce changes in the equilibrium among the four components of technological knowledge, thus triggering subcontractors’ evolution. Their findings provide evidence that Italian supply systems are moving towards supplier relations characterized by more intense technological knowledge transfer, stronger inter-firm collaboration and increasing subcontractors’ involvement and responsibilities in design. Moreover, they find that, although subcontractors still focus on few customers (on average the first two customers account for 70 percent of the total sales of the subcontractors), they tend to diversify their customer portfolios leveraging on their growing technological knowledge. Almost all these studies illustrate how subcontractors evolve and how their relationships with customers change, but do not explain the dynamics underlying these evolutionary processes. They describe the elements involved in the evolution process (the “what” of subcontractors’ evolution) but they do not explain the reason why these elements evolve and what are the interactions occurring among these elements over time (the “why” of subcontractors’ evolution). Despite their descriptive nature, these studies provide a fundamental insight: subcontractors’ evolution is driven by: . subcontractors’ capability development (i.e. how subcontractors learn to do better different activities); and . the nature of the buyer-supplier relationships in which subcontractors are engaged (i.e. what kind of customers, what they require, how they manage the relationship). Providing a synthesis Our study blends the static and the dynamic perspectives in an attempt to overcome the limits of a mere descriptive approach. Following Olsen and Ellram (1997), we build on the fundamental intuitions provided by the two streams of research briefly reviewed in the previous sections, i.e. that subcontractors’ diversity and evolution derive from

How do subcontractors evolve? 73

IJOPM 27,1

74

the capabilities they are able to develop as well as from the type of buyer-supplier relationships they are into. We also refer to Hakansson and Snehota’s (1995) activities-actors-resources model, which describes how business relationships can be analyzed through their individual substance layers (activity links, resource ties and actors bonds). Activity links are the links between the activities performed by each actor involved in the relationship. Resource ties are the links used by the actors to exchange existing resources, access to complementary resources and create new ones. Actors bonds describe the bonds between the actors through their perceptions, their trust and their understanding of each other. It is the interplay of bonds, ties and links that is at the origin of change and development in relationships. From a methodological standpoint, our study puts together the two perspectives, proceeding in two steps: (1) We first propose a classification of Italian subcontractors, identifying four different profiles as a function of their marketing and design capabilities. These capabilities are appropriate segmentation criteria for several reasons. First of all, it is demonstrated that they are involved in subcontractors’ evolution. For example, the studies on “lean suppliers” or “voice” practices in supply management (Cusumano and Takeishi, 1991; Helper and MacDuffie, 1997; Helper and Sako, 1998; Helper and Kiehl, 2004) provide empirical evidence about increasing supplier involvement in design activities. They also suggest that suppliers need to develop valuable design and marketing capabilities to handle the complex interdependencies arising from these new types of relationships. Secondly, industrial marketing literature and the studies on business markets structure and dynamics (Ford et al., 1986; Hakansson and Snehota, 1995) emphasize that subcontractors leverage on the interface capabilities (e.g. design and marketing capabilities) to manage their relationships portfolios. Thirdly, as highlighted in the previous sections, design and marketing capabilities are particularly critical for Italian subcontractors. (2) We conduct multiple case study analysis on a sample of ten subcontractors, and propose, using the segmentation resulting from the cluster analysis as a starting point, a model of how subcontractors’ capabilities evolve. This general model highlights: . the capabilities, in addition to design and marketing, that are involved in subcontractors’ evolution; and . the interactions over time among these capabilities. Clustering Italian subcontractors Data and variables As part of a broader research project on the evolution of Italian subcontracting relationships conducted in 2003 and 2004, we use an online questionnaire survey to gather a wide array of data regarding, among other aspects, the marketing and design capabilities of subcontractors located in Friuli-Venezia Giulia, in the North East of Italy. The data source is the Regional Center for Subcontracting of Friuli Venezia Giulia, an agency within the Pordenone Chamber of Commerce.

The data set provides information on a total of 431 firms operating in several industries. We eliminate 14 firms due to incomplete data which yields a complete set of information for 417 firms. Even if the analyzed subcontractors are located only in the Region of Friuli Venezia Giulia, the dataset is representative of the Italian subcontracting system as a whole. Indeed, a recent research based on a database of 5,262 Italian subcontractors located in several Italian Regions shows that our sample is representative (in terms of size, industry and basic skills distributions) of the whole Italian subcontracting system (Grandinetti and Bortoluzzi, 2004). Since, our intent is to classify subcontractors on the basis of their design and marketing capabilities, i.e. to identify clusters of subcontractors that share similar basic marketing and design capabilities profiles, we start by identifying proxies for these two variables. Our database offers several possible measures of design and marketing capabilities. Among them, we select which to be subjected to cluster analysis following the criteria suggested by Chiu et al. (2001). First, we delimitate the range of measures to only few and accurate ones. Second, because of the requirements posed by the aggregation algorithm, we choose to exclude the measures with a high correlation. Construct 1: marketing capabilities. These capabilities relate to subcontractors’ ability to monitor the market, to seek and identify new opportunities and market niches, to develop interactive relationships with its customers and to adopt customer relationship management practices. This definition overlaps with that of “relational capabilities” in the industrial marketing literature (Hakansson and Snehota, 1995). MC positively affect the subcontractors’ ability to: . diversify their customers’ portfolio; and . internationalize sales. Consequently, we use the degree of diversification of the customers’ portfolio and the degree of internationalization of sales as proxies for MC. More specifically, we use two variables to capture subcontractors’ MC: the degree of diversification of the customers’ portfolio (customer portfolio diversification – CPD) and degree of the sales internationalization of the subcontractor (sales internationalization – SI). We use the number of the customers served by each subcontractor during the last fiscal year (the dataset provides the following ranges of customers: 1-2; 3-5; 6-9; 10-19; 20-49; . 50) as the measure for the first variable. We use only the non-weighed number of customers (rather than the proportion of the main customers on the subcontractors’ total sales or other measures like the number of industries served) for two reasons. Firstly, small, young subcontractors typically have few customers each accounting for a high proportion of total sales (e.g. in the cases of spin-offs, subcontractors may have only one customer, the originating firm). As subcontractors evolve, the number of customers increases and the proportion of each customers’ sales on subcontractors’ total sales usually decreases. Secondly, since each customer represents a source of knowledge regardless the proportion of its sales on subcontractors’ total sales, the larger the number of customers the more opportunities for subcontractors’ learning (Hakansson and Snehota, 1995). We use the percentage of export sales on the total sales as the measure for the second variable, SI. Construct 2: design capabilities. DC relate to the ability to autonomously develop products/services that meet the client requirements. We use the adoption of

How do subcontractors evolve? 75

IJOPM 27,1

76

computer-aided design (CAD) systems as a proxy to measure subcontractors’ DC. We choose this variable for two reasons strictly connected with the peculiarities of the Italian subcontracting system. Firstly, other measures of DC suggested in the literature – such as the presence of appropriate product development systems (Liu and White, 1997; Liu et al., 2000; Wynstra et al., 1999; Petroni and Panciroli, 2002) – are not appropriate in the case of small and medium sized business like those included in the analyzed sample (35 percent of the firms has less than ten employees and 88.5 percent of the firms has less than 50 employees). Secondly, the implementation of CAD systems seems to be a clear and unequivocal signal of the presence, in the subcontractor’s organization, of specific DC. Interestingly, we observe that: . of 173 subcontractors that claim to have autonomous DC, 162 (94 percent) have implemented a CAD system; and . of 244 subcontractors that claim to have not autonomous DC, 14 subcontractors (6 percent) have implemented a CAD system. Given the fact that firms that claim to have or have not autonomous DC can have overestimated or underestimated their DC, we considered the adoption of CAD as a more conservative and accurate measure of autonomous DC. Cluster analysis Scholars have used cluster analysis techniques to study phenomena of interest in the management of supplier relations (Bensaou and Venkatraman, 1995). Using SPSS, we apply two-step cluster analysis, based on the aggregation algorithm proposed by Chiu et al. (2001), to our data (i.e. the three variables – CPD, SI, DC – used as measures for the two constructs defining subcontractors’ design and marketing capabilities). We choose this procedure for two reasons. Firstly, it allows dealing with variables that have different nature (in our case CPD is a ordinal variable, SI is a continuous variable and DC is a dichotomous variable). Secondly, it provides an effective and efficient way for identifying outliers (Chiu et al., 2001). The two-step procedure proposes the number of clusters on the basis of a sub-procedure that uses the Bayes information criterion and the variation of the distance between the clusters (Ketchen and Shook, 1996). The procedure uses a distance measure based on the log-likelihood function to assign each firm to its cluster. It automatically excludes 33 firms (7.9 percent of the sample) as outliers that cannot be aggregated to any of the clusters. Table I provides a summary of the sample characteristics by industry and size. The procedure identifies four clusters. Table II summarizes the profile of each cluster, providing some descriptive statistics. We define the clusters as follows: (1) A-type of subcontractors – developed subcontractors (n ¼ 93, 24.2 percent of the sample). They show the highest percentage of export on total sales (23.7 percent) with a diversified customers’ portfolio (all the firms have more than 50 customers). Almost all subcontractors (94 percent) adopt a CAD system. (2) B-type of subcontractors – developing subcontractors (n ¼ 77, 20.1 percent of the sample). They have a lower propensity to export than the developed subcontractors (12.2 percent is the average percentage of export to total sales)

Subcontractor size (number of employees) 1-9 10-19 20-49 50-99 .100

Industry Electronics Mechanics Rubber and plastics Furniture Apparel Textiles Total

Clusters A (n ¼ 93) B (n ¼ 77) C (n ¼ 123) D (n ¼ 91)

Number Percent Number Percent Number Percent Number Percent Number Percent Number Percent Number Percent

6 35.3 68 34.2 13 35.1 33 30.6 15 75.0 5 38.4 140 35.5

4 23.5 56 28.1 9 24.3 42 38.9 4 20.0 3 23.1 118 29.9

Percentage of export on total sales (SI) 23.7 12.2 6.7 6.0

4 23.5 49 24.6 10 27.1 25 23.1 0 0.0 3 23.1 91 23.1

1 5.9 16 8.0 1 2.7 5 4.6 1 5.0 1 7.7 25 6.4

Number of customers (CPD) (mode and percent) 50 or more 20-49 20-49 1-9

(100.0) (54.7) (38.9) (40.7)

2 11.8 10 5.0 4 10.8 3 2.8 0 0.0 1 7.7 20 5.1

Total 17 100.0 199 100.0 37 100.0 108 100.0 20 100.0 13 100.0 394 100.0

How do subcontractors evolve? 77

Table I. The sample composition: breakdown by industry and firm size

CAD system (DC) (percent) Yes (94.9) Yes (100.0) No (100.0) No (100.0)

and a less diversified customers’ portfolio (the customer range with the highest frequency is 20-49 with almost 55 percent of the subcontractors). However, all the developing subcontractors have adopted a CAD system. (3) C-type of subcontractors – question mark subcontractors (n ¼ 123, 32 percent of the sample). These subcontractors do not have DC (nobody has adopted a CAD system), present a low propensity to export (6.7 percent is the average percentage of export on total sales) but show a certain degree of diversification of the customer portfolio (20-49 is the customer range with the highest frequency with about 39 percent of the subcontractors falling into this range). (4) D-type of subcontractors – traditional subcontractors (n ¼ 91, 23.7 percent of the sample). They have a low propensity to export (6 percent is the average percentage of export on total sales), a limited customer portfolio (40.9 percent of the subcontractors have 1 to 9 customers) and no DC (all the subcontractors do not employ a CAD system). Since, cluster analysis is static, this segmentation does not provide any data on subcontractors’ evolution. However, we can infer that the different clusters are the result of some kind of evolution. Indeed, the literature quoted in the previous sections refers to changes occurred in subcontractors’ strategies and structures during the last decade. Moreover, several studies show that Italian subcontractors usually start as

Table II. Cluster analysis summary

IJOPM 27,1

78

“traditional” subcontractors (Becattini, 1999; Dei Ottati, 2003). For example, a large number of subcontractors originate from spin offs and begin doing business as outsourced operations of the originating firm. The “traditional” type can therefore be considered a sort of baseline for subcontractors. Consequently, since our cluster analysis identifies subcontractor types other than “traditional” we argue that some change has occurred and that at least some subcontractors have evolved to a different configuration as concerns design and/or marketing capabilities. In order to provide a crisper characterization of the clusters, we also calculate, for each of them, further descriptive statistics on other variables/subcontractors’ characteristics of related interest. These variables and data, presented in Table III, also come from the online questionnaire survey. They confirm the results of the cluster analysis, providing additional evidence about the diversity between the four types of subcontractors. First of all, the first two types of subcontractors are, on average, larger than the others. This would suggest that there are some scale effects involved due either to technological indivisibilities and threshold effects in marketing or, alternatively, to complementarities among these activities (Carlaw, 2004). However, a closer look to the data reveals that size is unevenly distributed across the four clusters. For example, 23.8 percent of “developed” subcontractors have less then nine employees while 22.7 percent of “traditional” subcontractors fall within the 20-49 size range. More generally, our cluster variables are not collinear with firm size and data do not show correlation between firm size and the clustering variables. For example, there is a significant fraction of small subcontractors (25 of those with less then nine employees, 17.8 percent), that have more than 50 customers. Similarly, approximately 30 percent of the larger subcontractors (those with more than 100 employees) have more than 50 customers while this fraction rises to 44 percent for subcontractors within the 50-99 employees size range. Furthermore, 35 percent of small subcontractors (less than nine employees) adopt CAD while 33 percent of subcontractors within the 10-19 employees size range adopt CAD. Finally, linear correlation coefficients with firm size (measured by the number of employees) are negligible in the cases of both MC (for export sales r ¼ 0,18) and DC (for the adoption of CAD r ¼ 0.19). The average proportion of sales coming from exclusive, customer-dedicated shop floor activities to total sales is smaller for developed and developing subcontractors. For example, the average for this variable is 26 percent for developed subcontractors, 51 percent for traditional ones. The presence of exclusive, customer-dedicated shop floor activities is a typical feature of the traditional subcontracting relationships, with subcontractors treated as buffers against demand fluctuations, as pure contractual extensions of the customer’s operations. Developed subcontractors seem to have moved away from this kind of attachment to their customers. Conversely, “developed” and “developing” subcontractors adopt and use advanced production technologies (APT) – like flexible automation, more than “traditional” and “question mark” subcontractors. Most of the literature maintains that, for its technological and managerial implications, as well as for its typical connection with CAD adoption and the ability to customize production, the adoption of APTs is positively correlated with the development of marketing and design capabilities (Petroni and Panciroli, 2002). Thus, APTs adoption and use, as well as patents, proprietary technologies, and even quality certification, signal the presence of innovation capabilities.

26 36 47 51

20-49 (26) 10-19 (41) 1-9 (39)

Percentage of sales from shop floor activities

20-49 (36)

Modal size (percent)

Note: Figures in parentheses represent percent

Developed subcontractors Developing subcontractors Question mark subcontractors Traditional subcontractors

Clusters

10 (11)

20 (16)

23 (30)

30 (32)

APT (percent)

5 (5)

5 (4)

10 (13)

14 (15)

Patents

15 (16)

16 (13)

22 (29)

36 (39)

Proprietary technology

9 (10) 35 (38)

24 (31) 17 (13)

28 (30)

20 (22)

31 (33)

7 (8)

Industrial district

Certified quality

How do subcontractors evolve? 79

Table III. Cluster’s profiles

IJOPM 27,1

80

All these variables confirm that the four clusters are significantly diverse, and suggest that developed and developing subcontractors have developed a wider set of capabilities, valuable in relationships other than traditional subcontracting. As suggested below, these capabilities are grounded on the ability to codify tacit knowledge (Nonaka and Takeuchi, 1995). Finally, the geographic location of subcontractors and, consequently, their embeddedness in industrial districts is not a feature specific to any of the subcontractors’ clusters. This makes sense since industrial districts are extremely diverse, in nature and structures, even in the same region. For example, in Friuli Venezia-Giulia, some industrial districts (e.g. the wood chair district of Manzano) are characterized by a prevalence of traditional subcontractors while in other industrial districts (e.g. the knife district of Maniago) developed subcontractors prevail. Summarizing, our cluster analysis allow us to classify, contingent on design and marketing capabilities, the analyzed subcontractors into four different types. At the same time, however, each individual subcontractor has worked out a trajectory leading to its current status. Along that trajectory, each subcontractor has developed, to a different extent, design and marketing capabilities. The next research question, we wish to address is: RQ2. Is there some kind of dynamic relationship between design, marketing, or other capabilities and what are the determinants of their development? Italian subcontractors’ evolution: an exploratory study Methodology and data collection Using our cluster analysis as a starting point, we wish to explore how and why subcontractors capabilities change and, consequently, understand some of the ongoing dynamics of Italian supply systems. In order to do so, we conducted ten case studies of subcontractors chosen from those included in the sample subjected to the cluster analysis. As usually happens in case study research (Yin, 1994; Eisenhardt, 1989), the choice of the research sample was not random (random or statistical sampling) but theoretical (theoretical sampling). Indeed, we chose subcontractors from different industries that belong to different clusters (i.e. seven developed subcontractors, two developing subcontractors, one question mark subcontractor). As suggested by Ellram (1996) and Voss et al. (2002), we collected information through semi-structured interviews, guided by a common case study protocol built on a review of the relevant literature, and on discussions with several operations managers from the firms involved in the study. For each case study, we interviewed CEOs and top managers. The case studies were first of all case histories (from foundation/inception to present time), with the interviews focused on subcontractors’ changes and transformation over time. On the whole, we interviewed 18 persons and each interview took about two hours. The interviews were taped and written up. Transcripts were used for the subsequent within-case and cross-case analysis. Each interview comprises three sections. The first addresses if and how the subcontractor has developed design and marketing capabilities. This part contains specific questions related to explicit research constructs (e.g. when did you introduce a CAD system? When did you start to use the CAD for designing new products? How have your customers changed, i.e. their number, identity, location, during time?). However, this section also includes more general questions aimed at investigating if

there were other capabilities involved in the above process (i.e. are there any other capabilities, beyond design and marketing capabilities, that have been relevant in the transformation of your firm?). The second section investigates what triggered the development of the capabilities, with questions analyzing both external and internal factors (e.g. what factors did facilitate and/or hurdle the development of the capabilities identified in the previous section?). The third section analyzes what were the conditions permitting the development of capabilities (e.g. what investment in tangible, intangible and organizational resources has supported the development of your capabilities?). Table IV reports a summary of key information about the ten subcontractors included in the sample. A model of subcontracting evolution From the cross-analysis of the case studies and after a series of iterative cycles of data gathering (Westbrook, 1995) we built an interpretative framework that explains subcontractors’ evolution (due to space constraints we do not report the longitudinal within-case analysis of each of the subcontractors). Firstly, the framework defines what are the capabilities involved in the evolution of a subcontractor. Interestingly, the case studies confirm that design and marketing capabilities are crucial in shaping subcontractors’ development, but also suggest that two other capabilities play an important role: (1) Supply management capabilities. The ability to design its upstream supply chain, to select suppliers with the required capabilities and to develop satisfactory relationships (in terms of purchased goods costs and quality, service level, delivery lead times, and flexibility) with them (Fine, 1998). (2) Knowledge codification capabilities. The ability to codify a firm’s understandings of the performance implications of internal routines in written tools such as manuals, blueprints, procedures, spreadsheets, decision support teams, and software (Zollo and Winter, 2002). Secondly, the framework defines the dynamic relationships among these capabilities and illustrates how their evolution affects one another (Figure 1). Knowledge codification capabilities are exogenous to our model. In fact, our cross-case analysis shows that they: . derive from investments in people and technologies needed to develop written tools such as manuals, blueprints, procedures, spreadsheets, decision support systems, and software; and . are usually a consequence of organizational discontinuities (new customers’ requirements, managerial turnover, entrepreneurial succession, financial crisis, change in governance or stakeholders, etc.). We did not observe any significant causal feedback relationships between design, marketing and supply management capabilities and knowledge codification capabilities. Our case analysis highlights that the main driver of subcontractors’ evolution is the codification of at least part of their tacit technical and commercial knowledge. In other words, it is knowledge codification that drives subcontractors’ development, enlarging and enriching their capabilities’ portfolio. Knowledge codification widens subcontractors’ relational capabilities because it allows them to “establish

How do subcontractors evolve? 81

Stage of evolution

Developed

Question mark Developed Developed Developed Developd Developed Developing Developed Developed

I1

I2 I3 I4 I5 I6 I7 I8 I9 I10

Table IV. The case studies

Firm

1956 1972 1990 1978 1985 1952 1980 1950 1969

Furniture Mec. Mec. Mec. Mec. Mec. Plastics Rubber Plastic

Furniture

Industry

7.105 40.762 4.573 2.861 4.132 4 1.70 19.192 44.802

4.994

Sales (mil. e)

88 351 50 80 37 35 40 440 130

150

Emp.

No Yes Yes Yes Yes Yes Yes Yes Yes

Yes

CAD

.150 20 .200 300 .100 .500 100 .100 .50

. 100

No. cust.

5 70 10 75 15 50 5 55 25

20

Percentage of export

No ISO 9001 No ISO 9001 No ISO 9001 ISO9001 ISO9001 ISO9001

ISO 9001 FSC

Cert. of quality

Automated handling (AH); FMS; CNC; CAM FMS; Automated handling; CNC FMS; EDI; ERP FMS; CNC; CAM EDI; AH; FMS; DNC; CAM EDI; AH; FMS; CNC; CAM EDI; CNC EDI; CAM; Robots; Moldflow FMS; CNC; CAM FMS; CNC; CAM

APT

82

1979

Funded

IJOPM 27,1

Relation 3

Supply management capabilities

How do subcontractors evolve?

Relation 4 Knowledge codification capabilities

Relation 1

Design capabilities

83

Relation 5 Relational capabilities Relation 2

Searching and monitoring capabilities Marketing capabilities

conversations” with non-local actors. Also, it allows developing self-awareness about the knowledge, routines and relationships the firm has. This, in turn, makes such capital replicable or, at least applicable to different business situations at negligible costs. Finally, being able to codify knowledge represents a necessary condition to structure an autonomous R&D function and introduce total quality management (TQM) and quality certification (Be´ne´zech et al., 2001; Romano and Vinelli, 2001). Knowledge codification capabilities constitute the ground on which DC are developed, too (relation 1). For example, the possibility to retrieve drawings or design elements from electronic archives facilitates autonomous new product development shortening lead times and reducing design costs. Moreover, being able to codify knowledge means to set up marketing information systems which support market searching and monitoring capabilities (relation 2). Knowledge codification capabilities also entail the establishment of non passive relationships with key customers (for example, enabling such cross-firm activities as root cause problem solving, just in time, target costing, TQM), thus enhancing subcontractors’ relational capabilities (relation 2). Finally, knowledge codification capabilities positively impact on supply management capabilities (relation 3). For example, the possibility to access and update systematically market data on the cost of raw materials and components supports such processes as vendor rating and suppliers’ development. The evidence provided by our case studies is also consistent with supply chain management theory (Fine, 1998), which maintains that design and supply management capabilities need to be aligned over time (relation 4). On the one hand, DC improve the subcontractors’ ability to efficiently allocate design tasks within and outside its boundaries. They also support the design process of its upstream partners. On the other hand, supply management capabilities improve the subcontractors’ ability to absorb new technical knowledge from its suppliers (Kotabe et al., 2003) thus enhancing their DC.

Figure 1. A model of subcontractors’ evolution

IJOPM 27,1

84

The case analysis highlights a bi-directional dynamic relationship also between DC and MC (relation 5). On the one hand, better design capabilities improve the subcontractor’s ability to approach the market proactively and represent the prerequisite for co-development practices. For example, shorter development times and lower design costs, as resulting from CAD adoption and use, widen subcontractors’ market horizons, allowing to reaching more potential customers, thus overcoming the local approach that typically characterizes traditional subcontractors. On the other hand, the development of relational capabilities grants access to technological knowledge from the customer (Dyer and Singh, 1998), thus enhancing DC. For example, fulfilling customers’ requirements in terms of quality, safety, and legal standards enriches the subcontractors’ ability to autonomously develop products/services that meet the requirements of other potential customers. The development of market and design capabilities positively impact on the diversification of customers’ portfolio and the diversification of products’/processes’ portfolio. As for the first point, the enhancement of monitoring and searching capabilities favors an enlargement of market horizons, reaching an increasing number of potential customers and overcoming a local market situation that typically characterized a traditional subcontractor. Leveraging on these capabilities, subcontractors can diversify their customer portfolios, geographical markets and served industries. As for the second point, DC are the basis of product and process innovation (Petroni and Panciroli, 2002). Interestingly, our case studies do not show any specific relation between subcontractors’ age (date of foundation) and their status or evolutionary trajectory. For example, the youngest and the oldest companies analyzed are both classified as “developed” while the third oldest is a “question mark”. Since, our sample for the case study analysis includes only subcontractors that have developed some design and marketing capabilities, it is interesting to note that there is not a unique and predefined evolutionary trajectory, and that the pace and the type of subcontractors’ transformation and capabilities development are not rigidly determined. Discussion and conclusions Our analysis has two main empirical results. Firstly, it provides a classification of Italian subcontractors on the basis of their design and marketing capabilities. Secondly, it proposes a model of how subcontractors’ capabilities develop over time. From these results we can derive some important theoretical and managerial implications. As for the first point, results of the cluster analysis support theory predictions and confirm our understanding of Italian supply systems: subcontractors are diverse and can be meaningfully classified on the base of their design and marketing capabilities. As for the second point, from a cross-case analysis of ten case studies, and after a series of iterative cycles of data gathering, we develop a general model to understand subcontractors’ change as a function of capabilities development. The model highlights several important aspects of this process of change. Firstly, the model shows that subcontractors’ transformation involves the dynamic over time of a wide range of capabilities. More specifically, the model poses that knowledge codification, supply management, design and marketing capabilities mutually reinforce one another and tend to align over time. For example, subcontractors do not develop marketing activities in spite of or without developing

over time DC. On the whole, this result suggests that, in order to understand firm’s competitive advantage, scholars, particularly those that draw on RBV (resource-based view) of the firm, should carry out longitudinal studies to capture how firm’s capabilities interact over time (Teece et al., 1997; Lorenzoni and Lipparini, 1999), thus creating new knowledge and capabilities, rather than consider them as independent entities (Combs and Ketchen, 1999; Kotabe et al., 2003). Secondly, supply management capabilities emerge as distinctive in the model. This is somewhat surprising, given the fact that most of the literature on supply networks focuses on the ability of focal or lead firms to deliberately manage their supply networks to create architecture of capabilities located both internally and externally (Frohlich and Westbrook, 2001; Furlan et al., 2006; Lorenzoni and Baden-Fuller, 1995; Lorenzoni and Lipparini, 1999; Dyer and Nobeoka, 2000). We claim that this ability is important also upstream the supply chain, for subcontractors. Thirdly, knowledge codification capabilities constitute a sort of logical antecedent and organizational prerequisite for subcontractors’ development to occur. This finding makes an original contribution to the literature on the strategy for knowledge creation in firms, especially medium and small ones. Indeed, in the knowledge-based view of the firm (Grant, 1996; Kogut and Zander, 1995; Nonaka and Takeuchi, 1995), the relationship between knowledge-codification capabilities and the creation of new tacit or explicit knowledge has not been sufficiently explored. As our results show, knowledge codification capabilities contribute to the creation of new knowledge in the form on new design, marketing and supply management capabilities. Knowledge codification capabilities foster the creation of new knowledge by enhancing both willingness to share knowledge and understanding among individuals (Un and Cuervo-Cazurra, 2004). Knowledge codification capabilities develop new organizational-level communication routines between individuals at all levels of the organizations thus promoting the transfer of knowledge within and across organizational boundaries. Furthermore, they contribute to create a common and available base of knowledge for all the organization thus creating a common code (Arrow, 1974) or a common knowledge (Grant, 1996) that make possible for individuals to understand each other. From this standpoint our study supports the view that knowledge codification is intimately related to the concept of dynamic capabilities in that it is a deliberate process of learning and one of the mechanisms involved in the creation and evolution of dynamic capabilities (Zollo and Winter, 2002). Beyond its managerial and theoretical implications, this study can also be useful in generating further conceptual and empirical research. Firstly, future research may refine subcontractors’ classification, articulating the four clusters we identified. Secondly, our framework for subcontractors’ development should be tested on the field through quantitative research. Thirdly, more accurate measures of subcontractors’ capabilities should be conceived and tested. Finally, given that knowledge codification is a sort of fundamental dynamic capability, it is worth conducting more in-depth research about how this capability originates. References Arrow, K. (1974), The Limits of Organization, Norton, New York, NY. Asanuma, B. (1989), “Manufacturer-supplier relationships in Japan and the concept of relation-specific skill”, Journal of the Japanese and International Economies, Vol. 3 No. 1, pp. 1-30.

How do subcontractors evolve? 85

IJOPM 27,1

86

Becattini, G. (1999), “La fioritura della piccola impresa e il ritorno dei distretti industriali”, Economia e Politica Industriale, Vol. 26 No. 103, pp. 5-16. Be´ne´zech, D., Lambert, G., Lanoux, B., Lerch, C. and Loss-Baroin, J. (2001), “Completion of knowledge codification: an illustration through the ISO 9000 standards implementation process”, Research Policy, Vol. 30 No. 9, pp. 1395-407. Bensaou, B.M. (1999), “Portfolios of buyer-supplier relationships”, Sloan Management Review, Summer, pp. 35-44. Bensaou, M. and Venkatraman, N. (1995), “Configurations of interorganizational relationships: a comparison between US and Japanese automakers”, Management Science, Vol. 41 No. 9, pp. 1471-92. Berger, S. and Locke, R. (2001), “‘Caso Italiano’ and Globalization”, Daedalus, Summer, pp. 85-104. Carlaw, K.I. (2004), “Returns to scale generated from uncertainty and complementarity”, Journal of Economic Behavior and Organization, Vol. 53, pp. 261-82. Chiu, T., Fang, D., Chen, J., Wang, Y. and Jeris, C. (2001), “A robust and scalable clustering algorithm for mixed type attributes in large database environment”, Proceedings of the 7th ACM SIKDD International Conference on Knowledge Discovery and Data Mining, pp. 263-8. Combs, J.G. and Ketchen, D.J. (1999), “Explaining interfirm cooperation and performance: toward a reconciliation of predictions from the RBV and organizational economics”, Strategic Management Journal, Vol. 20 No. 9, pp. 867-88. Cusumano, M.A. and Takeishi, A. (1991), “Supplier relations and management: a survey of Japanese, Japanese-transplants, and US auto plants”, Strategic Management Journal, Vol. 12, pp. 563-88. Dei Ottati, G. (2003), “Local governance and industrial districts’ competitive advantage”, in Becattini, G., Bellandi, M., Dei Ottati, G. and Sforzi, F. (Eds), From Industrial District to Local Development. An Itinerary of Research, Edward Elgar, Cheltenham, pp. 184-209. Dyer, H.J. and Nobeoka, K. (2000), “Creating and managing a high-performance knowledge sharing network. The Toyota case”, Strategic Management Journal, Vol. 21 No. 3, pp. 345-67. Dyer, J.H. and Singh, H. (1998), “The relational view: cooperative strategy and sources of interorganizational competitive advantage”, Academy of Management Review, Vol. 23 No. 4, pp. 660-79. Dyer, J.H., Cho, D.S. and Chu, W. (1998), “Strategic supplier segmentation: the next ‘best practice’ in supply chain management”, California Management Review, Vol. 40 No. 2, pp. 57-77. Eisenhardt, K.M. (1989), “Building theories from case study research”, Academy of Management Review, Vol. 14 No. 4, pp. 532-50. Ellram, L.M. (1996), “An application of the case study method in logistics research”, Journal of Business Logistics, Vol. 17 No. 2, pp. 93-138. Esposito, E. and Lo Storto, C. (1991), “Il sistema di subfornitura”, Sviluppo & Organizzazione, No. 130. Esposito, E. and Raffa, M. (1994), “The evolution of Italian subcontracting firms: empirical evidence”, European Journal of Purchasing & Supply Management, Vol. 1 No. 2, pp. 67-76. Fine, C. (1998), ClockSpeed. Winning Industry Control in the Age of Temporary Advantage, Perseus Books, Boston, MA. Ford, D., Hakansson, H. and Johanson, J. (1986), “How do companies interact?”, Industrial Marketing and Purchasing, Vol. 1 No. 1, pp. 26-41.

Frohlich, M.T. and Westbrook, R. (2001), “Arcs of integration: an international study of supply chain strategies”, Journal of Operations Management, Vol. 19, pp. 185-200. Furlan, A., Romano, P. and Camuffo, A. (2006), “Customer-supplier integration forms in the air-conditioning industry”, Journal of Manufacturing Technology Management, Vol. 17 No. 5, pp. 633-55. Grandinetti, R. and Bortoluzzi, G. (2004), L’evoluzione delle impese e dei sistemi di subfornitura (The Evolution of the Firms and of the Subcontracting Systems), Franco Angeli, Milano. Grant, R. (1996), “Toward a knowledge-based theory of the firm”, Strategic Management Journal, Vol. 16 No. 7, pp. 519-33. Hakansson, H. and Snehota, I. (1995), Developing Relations in Business Networks, Routledge, London. Helper, S. and Kiehl, J. (2004), “Developing supplier capabilities: market and non-market approaches”, Industry & Innovation, Vol. 11 Nos 1/2, pp. 89-107. Helper, S. and MacDuffie, J.P. (1997), “Creating lean suppliers: diffusing lean production through the supply chain”, California Management Review, Vol. 39 No. 4, pp. 118-51. Helper, S. and Sako, M. (1998), “Determinants of trust in supplier relations: evidence from the automotive industry in Japan and the United States”, Journal of Economic Behavior & Organization, Vol. 34, pp. 387-417. Ketchen, D. and Shook, C.L. (1996), “The application of cluster analysis in strategic management research: an analysis and critique”, Strategic Management Journal, Vol. 17, pp. 441-58. Kogut, B. and Zander, U. (1995), “Knowledge and the speed of transfer and imitation of organizational capabilities: an empirical test”, Organization Science, Vol. 6 No. 1, pp. 76-91. Kotabe, M., Martin, X. and Domoto, H. (2003), “Gaining from vertical partnerships: knowledge transfer, relationship duration, and supplier performance improvement in the US and Japanese automotive industries”, Strategic Management Journal, Vol. 24 No. 4, pp. 293-316. Kraljic, P. (1983), “Purchasing must become supply management”, Harvard Business Review, September/October, pp. 109-17. Lamming, R. (1993), Beyond Partnership, Prentice Hall, London. Lamming, R., Johnsen, T., Zheng, J. and Harland, C. (2000), “An initial classification of supply networks”, International Journal of Operations & Production Management, Vol. 20 No. 6, pp. 675-91. Liu, X.R. and White, S. (1997), “The relative contribution of foreign technology and domestic inputs to innovation in Chinese manufacturing industries”, Technovation, Vol. 17 No. 3, pp. 119-25. Liu, J., Ding, F-Y. and Lall, V. (2000), “Using data envelopment analysis to compare suppliers for supplier selection and performance improvement”, Supply Chain Management: An International Journal, Vol. 5 No. 3, pp. 143-50. Lorenzoni, G. and Baden-Fuller, C. (1995), “Creating a strategic center to manage a web of partners”, California Management Review, Vol. 37 No. 3, pp. 146-63. Lorenzoni, G. and Lipparini, A. (1999), “The leveraging of interfirm relationships as a distinctive organizational capability: a longitudinal study”, Strategic Management Journal, Vol. 20, pp. 317-38. Masella, C. and Rangone, A. (2000), “A contingent approach to the design of vendor selection systems for different types of co-operative customer/supplier relationships”, International Journal of Operations & Production Management, Vol. 20 No. 1, pp. 70-84.

How do subcontractors evolve? 87

IJOPM 27,1

88

Narasimhan, R. (1983), “An analytical approach to supplier selection”, Journal of Purchasing and Materials Management, Vol. 19 No. 4, pp. 27-32. Nassimbeni, G., De Toni, A. and Tonchia, S. (1996), “L’evoluzione dei rapporti di subfornitura”, Sviluppo & Organizzazione, Vol. 137, pp. 96-107. Nonaka, I. and Takeuchi, H. (1995), The Knowledge-creating Company, Oxford University Press, New York, NY. Nydick, R.L. and Hill, R.P. (1992), “Using the analytic hierarchy process to structure the suppliers selection procedure”, International Journal of Purchasing and Materials Management, Vol. 28 No. 2, pp. 31-6. Olsen, R.F. and Ellram, L.M. (1997), “A portfolio approach to supplier relationships”, Industrial Marketing Management, Vol. 26 No. 2, pp. 101-13. Petroni, A. and Panciroli, B. (2002), “Innovation as a determinant of suppliers’ roles and performances; an empirical study in the food machinery industry”, European Journal of Purchasing and Supply Management, Vol. 8, pp. 135-45. Porter, M. (1990), The Competitive Advantage of Nations, The Free Press, New York, NY. Romano, P. and Vinelli, A. (2001), “Quality Management in a supply chain perspective. Strategic and operative choices in a textile-apparel network”, International Journal of Operations & Production Management, Vol. 21 No. 4, pp. 446-60. Sabel, C.M. and Piore, M. (1984), The Second Industrial Divide. Possibilities for Prosperity, Basic Books, New York, NY. Teece, D.J., Pisano, G. and Shuen, A. (1997), “Dynamic capabilities and strategic management”, Strategic Management Journal, Vol. 8 No. 7, pp. 509-33. Un, C.A. and Cuervo-Cazurra, A. (2004), “Strategies for knowledge creation in firms”, British Journal of Management, Vol. 15 No. 1, pp. 27-41. Van Weele, A.J. (2000), Purchasing and Supply Chain Management, Business Press/Thomson Learning, London. Vokurka, R., Choobineh, J. and Vadi, L. (1996), “A prototype expert system for the evaluation and selection of potential suppliers”, International Journal of Operations & Production Management, Vol. 16 No. 12, pp. 106-27. Voss, C., Tsikriktsis, N. and Frohlich, M. (2002), “Case research in operations management”, International Journal of Operations and Production Management, Vol. 22 No. 2, pp. 195-219. Westbrook, R. (1995), “Action research: a new paradigm for research in production and operations management”, International Journal of Operations & Production Management, Vol. 15 No. 12, pp. 6-20. Wynstra, F., van Weele, A. and Axelsson, B. (1999), “Purchasing involvement in product development: a framework”, European Journal of Purchasing & Supply Management, Vol. 5 Nos 3/4, pp. 129-41. Yin, R. (1994), Case Study Research, Sage, Beverly Hills, CA. Zanoni, A. (1991), “La gestione strategica degli approvvigionamenti”, in Filippini, R., Pagliarani, G. and Petroni, G. (Eds), Progettare e Gestire la complessita`, Conference Proceedings, AiIG, Vicenza. Zollo, M. and Winter, S.J. (2002), “Deliberate learning and the evolution of dynamic capabilities”, Organization Science, Vol. 13 No. 3, pp. 339-51.

Further reading Ellram, L. (1997), “The supplier selection decision in strategic partnerships”, Journal of Purchasing and Materials Management, Vol. 26 No. 3, pp. 8-14. About the authors Andrea Furlan is an Assistant Professor of Operations and Supply Chain Management at the University of Padova, Italy, Department of Economics. He graduated in business economics at University of Padova where he completed his PhD in Economics and Management with a doctoral dissertation on supply chain management. He works within the Business and Operations Management group at the Department of Business and Economics (University of Padova). His research has appeared in the Journal of Manufacturing Technology Management, Journal of Purchasing and Supply Management and International Journal of Operations & Production Management. Andrea Furlan is the corresponding author and can be contacted at: [email protected] Roberto Grandinetti is a Professor of Management at the University of Padova, Italy, Department of Economics. His main areas of research are the international opening of Italian industrial districts, the basic concepts of knowledge management, and the evolution of marketing in a relational sense. He has published books and articles in journals like the Revue d’Economie Industrielle, Human Systems Management, International Journal of Operations & Production Management, Piccola Impresa/Small Business. Arnaldo Camuffo is a Professor of Management at the University of Padova, Italy, Department of Economics. He holds a degree and a PhD in management from the University of Venice and a Master of Science in Management from the Sloan School of Management at MIT. He has published books (the most recent is Automation in Automotive Industries. Recent Developments, Berlin, Springer, 1999 with A. Comacchio and G. Volpato) and several articles in journals like the MIT Sloan Management Review, Industrial and Corporate Change, Industrial Relations, Industry and Innovation, International Journal of Human Resource Management, International Journal of Training and Development, Journal of Management and Governance, International Journal of Innovation and Learning, Human Resource Development, Journal of Manufacturing Technology Management, International Journal of Operations & Production Management, Journal of Purchasing and Supply Management and Korean Journal of Political Economy.

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

How do subcontractors evolve? 89

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0144-3577.htm

IJOPM 27,1

The impact of forecast information quality on supply chain performance

90

Helena Forslund Department of Logistics and Supply Chain Management, School of Management and Economics, Va¨xjo¨ University, Va¨xjo¨, Sweden, and

Patrik Jonsson Division of Logistics and Transportation, Chalmers University of Technology, Go¨teborg, Sweden Abstract Purpose – This paper aims to describe the extent of supplier access to customer forecast information and the perceived quality of such information and also to explain the impact of forecast information access and forecast information quality (FIQ) on supply chain performance. Design/methodology/approach – FIQ is defined, and a measurement instrument is developed from theory. The analysis is based on a survey of the most important suppliers of 136 Swedish companies. Findings – Findings show that a large proportion of the suppliers receive customer forecasts, but that the FIQ is lower further upstream in the supply chain and, in some variables, lower for make-to-order suppliers. The greatest information quality deficiency of the forecast was that it was considered unreliable. The only significant difference in supply chain performance found between make-to-stock suppliers with and without access to forecast was related to the use of safety stock in finished goods inventory. Research limitations/implications – The study contains two types of conclusions: those developed from the conceptual discussion in the theoretical framework and those of the empirical study. In the theoretical framework, measurement instruments for FIQ and supply chain performance (corrective actions, preventive actions and customer service performance) were developed. The study identified several empirical relationships, but it was conducted on a sample with a lot of variation. Practical implications – The understanding of the performance impact of FIQ. FIQ shows quality deficiencies on all variables, which indicates room for improvement. Originality/value – Research on supply chain information quality as well as dyadic research approaches are rare. Keywords Forecasting, Information, Quality, Supply chain management Paper type Research paper

International Journal of Operations & Production Management Vol. 27 No. 1, 2007 pp. 90-107 q Emerald Group Publishing Limited 0144-3577 DOI 10.1108/01443570710714556

Introduction Several studies and authors have emphasized the importance of sharing information between customers and suppliers in supply chains, especially point-of-sale (POS) and forecast data (Stank et al., 1996; Kelle and Akbulut, 2005; Christopher and Towill, 2000; Cachon and Fisher, 2000; Lee et al., 1997). However, while most studies discuss the general importance of having access to forecast information along the supply chain, they do not explain the impact of the available forecast information on supply chain performance, how available forecast information may have quality deficiencies, or how different information quality deficiencies may impact the usefulness of forecasts.

Supply chain performance is typically related to metrics reflecting cost, tied-up capital and customer service (Brewer and Speh, 2000). The supplier might need to use internal actions to compensate for poor customer service. Corrective actions, such as rush orders and overtime, are mainly related to costs for the supplier. Preventive actions, for example, safety stocks and extra capacity, are mainly related to tied-up capital, but also to costs. The use of corrective and preventive actions could allow for good customer service performance even if the demand is uncertain, for example, as a result of lack of access to customer forecasts. The importance of forecast data also depends on how it is used in the manufacturing planning and control processes. Long-term capacity forecasts may, for example, be considered most important for firms applying make-to-order (MTO) strategies, while short-term material forecasts could be expected to be more important when applying make-to-stock (MTS) strategies. As a supplier, it may, however, not suffice to only have access to customers’ forecasts. The interpretation and possible use of the forecast also depends on the quality of the forecast information, i.e. to what extent the supplier perceives the customer’s forecast information as fulfilling expectations. The forecast could, for example, be available too late to be used in the planning process, be changed so often that the supplier does not trust it, or exchanged in an inappropriate format, for example, as a faxed document that needs much further processing before the supplier can make use of it, etc. More rigorous empirical research on the performance effects of inter-organizational forecast information exchange is scarce. Several authors have emphasized the importance of conducting empirical studies on the performance impact of collaboration and the sharing of forecasts (McCarthy and Golicic, 2002). Consequently, we lack empirical studies on the performance impact of access to customer forecast and forecast information quality (hereafter denoted FIQ) discussed above. Empirical studies describing the present extent of forecast information sharing in supply chains are lacking. For example, to what extent is forecast information exchanged between companies? Does the exchange of forecast information and the quality of the forecast information differ further upstream in the supply chain (e.g. between 1st and 2nd tier suppliers and between 2nd and 3rd tier suppliers) compared to down-stream (e.g. between retailers and OEMs and between OEMs and 1st tier suppliers)? Does the exchange and perceived FIQ differ between companies with different supply strategies, for example, between suppliers using MTS compared to MTO strategies? This paper deals with these research issues. It contains two sub-objectives. The first is to describe the extent of supplier access to customer forecast information and its perceived information quality. The second is to explain the performance impact of forecast information access and FIQ on supply chain performance. Not all types of forecast information are studied; only demand information important for supply chain performance. This means future demand expressed as product content, order quantity and delivery time. Firstly, the theoretical framework is described and hypotheses related to the second explanatory sub-objective are developed. Then, the data collection conducted through a mailed survey is discussed, and findings of the empirical analysis are presented and discussed. Theoretical framework and hypotheses generation The issue of sharing forecasts in the supply chain has been studied from some different perspectives, for example, the collaborative planning forecasting and replenishment

The impact of FIQ

91

IJOPM 27,1

(CPFR) approach and modeling-based approaches. No identified study has looked at the FIQ, but the concept of information quality has been studied in other disciplines. Here, a framework for describing and analyzing the FIQ is defined and developed. The definitions and measures of supply chain performance are also presented, along with a generation of hypotheses.

92

Sharing forecast information A recent survey with 177 responses from Swedish manufacturing companies in different industries (response rate of 38 percent) with more than 100 employees, measuring the extent of forecast information sharing (Sandberg, 2005) shows that 95 percent of the companies exchange forecast information with at least a monthly frequency. Aside from this study, no other broad descriptive study on forecast information sharing was identified. The literature on CPFR discusses the issue of interand intra-organizational collaboration in the forecasting process. One important objective of forecasting collaboration is to develop a common plan for the single company and for the supply chain as a whole (Helms et al., 2000; McCarthy and Golicic, 2002). Most of the CPFR studies adopt a retailer perspective. Holmstro¨m et al. (2002) showed that the suppliers are more interested in sharing information and collaborating in the forecasting process than are the retailers. Their findings also indicated that suppliers gain most of the benefits of increased information sharing. A study by Sma˚ros (2003) on the forecasting collaboration in the grocery industry indicates that CPFR-style intensive forecasting collaboration is not often feasible. Instead, the focus of the supplier-retailer collaboration should be on making useful demand information, such as POS and forecasts, available for the suppliers. This would allow suppliers to gain access to timely and accurate demand information from the retailers. This means that the retailers have the opportunity to delegate some of the planning and forecasting responsibility to motivated suppliers, and the suppliers acquire high information quality input to their planning processes. Even though CPFR shows promise for improving supply chain performance, the adoption rate has been slower than expected, especially in Europe. The single most important obstacle was claimed to be the customer’s lack of forecasting processes and resources (Sma˚ros, 2003). The CPFR related studies emphasize the importance and performance potential of forecast exchange, especially for the supplier in a dyadic supplier – customer relationship. However, the correlations between forecast exchange and performance are not studied. There are also some modeling based studies on forecasting exchange. Cachon and Fisher (2000) recognized that the performance effects of sharing forecast information are low when demand is predictable, as compared to situations where demand is unpredictable. Zhao (2002) showed that the supplier capacity constraints impact the possibility of the supplier successfully using the customer forecast. Aviv (2001) found through a simulation study that collaborative forecasting results in lower costs as compared to local forecasting, i.e. a situation where each individual actor develops its own forecast. A number of studies indicate the importance of sharing forecast and POS data in order to decrease the bullwhip effect (Lee et al., 1997). Consequently, studies show that a forecast received from a customer company could result in positive results but that the effects are dependent on the conditions on hand, i.e. how the forecast is used in the supplier’s planning processes. It could, for example, be expected that forecast exchange is more important if using MTS strategies compared to MTO.

Previous studies have, however, not compared the effects of forecast exchange between different types of companies or manufacturing strategies, but rather focused on specific situations when studying the performance effects of forecasting. Forecast information quality It does not suffice to only have access to customers’ forecasts. The interpretation and possible use of the forecast data depend on the quality of forecast information. Information quality cannot be measured objectively; instead, it must be judged by the receiving supplier (Forslund, 2004). As defined here, FIQ is not equal to forecast error, measured as the deviation between actual demand and forecast and expressed as mean absolute deviation or mean error. FIQ is discussed and defined in the two following sections. More exactly, some features of FIQ are critical. Although no study of measuring FIQ was found, there are studies on information quality in general (English, 1999; Lee et al., 2002) and some on order information quality. Petersen (1999) measured information quality in terms of if it was current, accurate, complete, compatible or convenient to access. Whipple et al. (2002) studied operational information exchange in alliances. They did not use the construct information quality, but measured the value of information deployed as timely and accurate. English (1999) and Lindau and Lumsden (1993) stated that the information quality dimensions could be derived from the seven rights of logistics, i.e. the right place, time, quantity, quality, price, condition and customer (Stock and Lambert, 1992). The seven rights origin from Weld’s (1916) five rights of a logistics system to supply the right product, at the right place, at the right time, in the right condition for the right cost to the customers consuming the product. Lindau and Lumsden (1993) focused on three information quality dimensions derived from the rights. With correct information they meant the right information in the right condition. With timely information they meant information received at the right time, to the right receiver and to the right place. Complete information was related to the right quantity. Moberg et al. (2002) stated that information has little value if it has poor reliability and validity. They further discussed accuracy, timeliness and proper formatting as important qualities of information, but without defining the dimensions. Closs et al. (1997) evaluated information systems with dimensions regarding both information in itself and the information system. The dimensions regarding information were timeliness, accuracy, availability and formatting to facilitate usage. Clikeman (1999) says that companies that have reengineered their production processes have identified four dimensions of quality: relevance, reliability, timeliness and cost. The same dimensions apply to information. Most information quality related studies do not focus on forecast information used in manufacturing or supply chain planning. Therefore, a “transition” from these general studies towards FIQ variables is necessary. We here define and describe FIQ with the four information quality variables: in time, accurate, convenient to access, and reliable. Forecast information being in time means that it is in the agreed time, when the information customer wants it. It is also concurrent to the situation, i.e. that the state of information used for decision making corresponds to the situation, so the data represented is not time-phased with regard to when it was registered and presented. Accuracy concerns the degree of obvious mistakes in the information. Forecast information coming from a customer might be impaired by obvious mistakes, which must be corrected before being entered into the supplier’s planning system.

The impact of FIQ

93

IJOPM 27,1

Convenient to access, deals with the ease of accessing the data without further processing. Processing could mean adapting an item code or entering it manually into the supplier’s planning system. The fourth variable, reliability, refers to the probability that a forecast will remain unchanged. Unreliable information means uncertainty to the supplying company, which has to be absorbed by safety mechanisms.

94

Supply chain performance Some of the modeling-based studies on sharing forecasts in supply chains link forecast exchange to supply chain performance. Lee et al. (1997) and several others have shown that the demand variability can be amplified upstream in the supply chain when not sharing accurate forecasts with the suppliers. Zhao et al. (2002) concluded that the value of information sharing is significantly influenced by the demand pattern, forecasting model used and the supplier’s capacity tightness, i.e. its total production capacity in relationship to the total demand to be satisfied, but that the suppliers usually can improve their total costs and customer service dramatically through information sharing under all conditions. Aviv (2001) compared local forecasting with exchange of collaborative forecasts and concluded that the supply chain costs were reduced when exchanging forecast information. Also several CPFR related pilot studies (Fliedner, 2003) include performance improvement figures, such as higher order fill rates, lower product inventories, obsolescence and deterioration for the supplier. McCarthy and Golicic (2002) made an exploratory study of collaborative forecasting, which was defined as a long-term relationship among organizations actively working together with forecasting (Mentzer et al., 2000), and identified substantial impact on supply chain performance. Improvement in customer service performance, such as shorter lead times, improved inventory availability and better response to fluctuations in demand, was found. Furthermore, improvements in cost and capital were found which could be related to reductions in safety stock. A survey by Frohlich and Westbrook (2001) showed that firms with higher levels of supplier and customer integration also demonstrated the highest supply chain performance. Frohlich and Westbrook did not explicitly study sharing forecasts, but sharing production plans was one of the integrative activities studied. Another study (Thonemann, 2002) that focuses on supply chain exchange of information also identified significant cost savings for the supplier. Consequently, numerous studies emphasize the positive impact of forecast information exchange, but there is a lack of studies that empirically explain the performance impact of forecast exchange. Generation of the conceptual model and hypotheses Figure 1 shows the variables studied in the two sub-objectives of the paper. The first sub-objective describes the state-of-the-art access of customer’s demand forecast and Sub-objective 1

Supplier

Figure 1. Conceptual model of the study

Customer’s demand forecast Forecast information quality

- MTS/MTO - 2nd tier/1st tier /OEM

Supply chain performance Sub-objective 2 - Corrective actions (H1-H2) - Preventive actions - Customer service

its perceived information quality at suppliers. The access and perceived information quality is compared between companies applying MTS and MTO strategies, and between suppliers in different supply chain tiers. Studies show that the level of supply chain integration is more developed further down-stream, i.e. closer to the end customer (Mattsson, 2002). Here, access to customer forecasts and perceived FIQ are compared between 2nd tier suppliers, 1st tier suppliers and OEMs. To analyze sub-objective two about the performance impact of customer forecast access and perceived FIQ, two hypotheses could be derived from the above discussion in the theoretical framework. The first hypothesis deals with the performance impact of forecast access. Supply chain performance could be related to customer service, corrective and preventive actions. Customer service goals could be fulfilled by using corrective actions. These are the supplier’s unplanned actions to correct deficiencies occurring in customer service related to a specific customer order. These actions are such that result in “increased” costs. Based on Lindau and Lumsden (1993) and Mattsson (2002) the following corrective actions are defined: subcontracting, expediting, part delivery, re-scheduling, reservation breaking, overtime and express transports. Using preventive actions or safety mechanisms can also deal with planning uncertainty. Preventive actions are of a character that results in “increased” tied-up capital and costs. Based on Lindau and Lumsden (1993) and Fahle´n (1997) the following preventive actions are defined: safety stocks in raw material and finished goods inventories, safety capacity, safety lead times and over-planning. These are the supplier’s planned actions to prevent future deficiencies in customer service. The use of corrective and preventive actions allows for good customer service performance even though the planning environment is uncertain, for example, as a result of lacking or quality deficient forecast information. Consequently, the combined effect of costs and tied-up capital, related to corrective and preventive actions, and customer service make up the supply chain performance. It can be expected that a history of forecasts makes it possible for the supplier to act correctively and preventively when deficiencies in supply chain performance may occur. However, if the environment is very uncertain or the costs and cash-flow effects of the corrective and preventive actions are very high, the customer service performance can still be expected to be lower in firms without access to customer forecasts compared to firms with access to customer forecasts. Obviously, it is not possible to explain the impact of forecast exchange on any of the individual performance variables corrective actions, preventive actions and customer service. In accordance with the above discussion about the expected supply chain performance impact of forecast exchange (Fliedner, 2003; Thonemann, 2002; Aviv, 2001; Zhao et al., 2002; McCarthy and Golicic, 2002) we use the combined effect of corrective actions, preventive actions and customer service when defining supply chain performance, and formulate the first hypothesis in the following way: H1. Supply chain performance is higher for suppliers with access to customer forecasts compared to suppliers without access to forecasts. Hypothesis two deals with the performance impact of the FIQ. Several studies have emphasized the expected supply chain performance impact of exchanging forecasts as discussed when deriving H1. No identified study has, however, studied the

The impact of FIQ

95

IJOPM 27,1

96

performance impact of FIQ. Keebler et al. (1999) and McCarthy and Golicic (2002) concluded that accurate forecast information significantly improves efficiency but they did not define accuracy and did not study any other FIQ dimension. The only FIQ-related dimension identified in previous research is the forecast error, defined as the deviation between forecast and actual demand. It has a direct impact on the customer service and safety stock levels (Vollmann et al., 2005). In this paper, the forecast error is, however, not defined as a FIQ dimension. Instead, we use the four variables: in time, accurate, convenient to access and reliable. A common reasoning is that delayed, inaccurate and unreliable forecasts create demand uncertainty upstream in the supply chain, and therefore, impact the supply chain performance in negative ways (Lee et al., 1997; van der Vorst and Beulens, 2002). A forecast could arrive in time, be formally correct and reliable but still not useful for the user because of its inconvenience to access. The convenience to access is especially important when the forecast information is processed manually, for example, stored on paper print outs or excel sheets that are not automatically readable by the receiver’s ERP system. In such situations, low-levels of convenient access could be expected to be both time consuming and result in data registration errors (Lindau, 1995). Thus, the convenience to access forecast information could also be expected to impact the performance. Consequently, we could expect that improved FIQ impacts the use of corrective actions, preventive actions and customer service in a way that has a positive impact on the combined supply chain performance. This results in the following formulation of hypothesis two: H2. Supply chain performance is positively correlated with FIQ. Methodology A dyadic research approach was necessary to describe the cooperation between suppliers and certain customers. Surveys were selected as the empirical data generating method, as recommended by McCarthy and Golicic (2002). Surveys fit into a desire to conduct objective, deductive research with little interviewer bias. A larger number of respondents can be addressed with surveys, which makes it possible to use statistical analysis techniques such as factor analysis (Hair et al., 1998). Survey instrument Tables I-IV show the definitions and operationalisations of the variables related to FIQ, corrective actions, preventive actions and customer service. The FIQ variables are derived from the theoretical framework on FIQ (English, 1999; Petersen, 1999; Moberg et al., 2002). The corrective and preventive action variables are based on the works of Lindau and Lumsden (1993), Ericsson (1997), Fahle´n (1997), and Mattsson (2002). The customer service variables are based on Stock and Lambert (1992) and Mattsson (2002). Likert scales from 1 to 7 were used for all these variables, measured on ordinal scales. The questions asked and definitions of scales for the respective variable are included in Tables I-IV. The average of the four information quality variables was defined and used as an overall FIQ index (SIQFOR). This scale was considered to possess both content and construct validity. Content validity was assured as the scale was based on extensive literature reviews and by conducting pretests with academic colleagues and practitioners (Flynn et al., 1990). Construct validity was ensured, as all eigenvalues

Definition

In time

Arriving in the agreed time – within the Lindau (1995), Clikeman (1999), and supplier’s planning horizon English (1999) Free from obvious mistakes Petersen (1999), Whipple et al. (2002), and English (1999) Easy access without further processing Petersen (1999), English (1999), Closs et al. (1997), Moberg et al. (2002), and Keebler et al. (1999) The probability that a forecast remains Mattsson (2002) and Moberg et al. (2002) unchanged

97

Notes: Question: forecasts received from the customer are: (a) in time; (b) accurate; (c) convenient to access; and (d) reliable; scale: seven point Likert scale from 1 (strongly disagree) to 7 (strongly agree)

Table I. Variables of FIQ

Accurate Convenient to access Reliable

Corrective action variable Subcontracting Expediting Part delivery Re-scheduling Reservation breaking Overtime Express transports

Source (order information quality)

The impact of FIQ

Variable

Definition

Source

Short-term, as a result of unforeseen overload Finding and rushing “hot” jobs through production Smaller batches in production or delivery Re-plan

Lindau and Lumsden (1993)

Already reserved material (for another customer) is used earlier Short-term A faster and more expensive means of transportation is used to speed up a delivery

Lindau and Lumsden (1993) Lindau and Lumsden (1993) Lindau and Lumsden (1993) and Fahle´n (1997) Lindau and Lumsden (1993) Ericsson (1997) and Mattsson (2002) Mattsson (2002)

Notes: Question: to perform the promised customer service we use: (a) subcontracting; (b) expediting; (c) part delivery; (d) re-scheduling; (e) reservation breaking; (f) overtime; and (g) express transports; scale: seven point Likert scale from 1 (to very low extent) to 7 (to very high extent)

were larger than one and individual factor loadings exceed 0.40, (in fact they exceed 0.80). The scale was further considered reliable according to the Cronbach’s a value of 0.87. An a value of 0.70 is considered acceptable for a scale, and for new scales 0.60 is acceptable (Hair et al., 1998). Data collection One customer and one supplier survey questionnaire were developed. To obtain the view of customers, purchasing managers were addressed. Swedish companies with 100 or more employees were selected in four industries (fabricated metal industry, machinery, electronic equipment and automotive), and 432 purchasing managers were found. It was then decided to address the entire population. Customers answered questions related to their most important Swedish supplier. Some 171 usable responses were received, corresponding to a response rate of 39.6 percent. About 149 customers provided contact information for the addressed supplier.

Table II. Variables of corrective actions

IJOPM 27,1

98

Table III. Variables of preventive actions

Table IV. Variables of customer service performance

Preventive action variable

Definition

Safety stock in raw material Stock kept as a reserve to guard inventory against material shortage because of uncertainties in supply, demand and lead time Safety stock in finished Stock kept as a reserve to guard goods inventory against material shortage because of uncertainties Safety capacity The reservation of extra capacity, i.e. plan with under-capacity utilization to protect against unforeseen events Safety lead time The order starts earlier to be finished before its due date Over-planning (demand Instead of safety stock or safety lead hedges) time, a larger quantity than known demand is planned

Source Lindau and Lumsden (1993) and Fahle´n (1997) Lindau and Lumsden (1993) and Fahle´n (1997) Lindau and Lumsden (1993) and Fahle´n (1997) Lindau and Lumsden (1993) and Fahle´n (1997) Lindau and Lumsden (1993)

Notes: Question: to perform the promised customer service we use: (a) safety stock in raw material inventory; (b) safety stock in finished goods inventory; (c) safety capacity; (d) safety lead time; and (e) over-planning; scale: seven point Likert scale from 1 (to very low extent) to 7 (to very high extent)

Customer service performance variable

Source

Promised lead time (the time between placing and receiving an order) On-time delivery (orders are delivered at agreed time) Rush orders when needed Promised inventory availability (to what degree orders can be delivered from inventory) Accurate orders (the right number of items ordered arrives) Availability of delay information

Stock and Lambert (1992) Stock and Lambert (1992) Stock and Lambert (1992) Stock and Lambert (1992) Stock and Lambert (1992) Stock and Lambert (1992) and Mattsson (2002)

Notes: Question: for our most important customer we perform perfect: (a) promised lead time; (b) on-time delivery; (c) use rush orders when needed; (d) promised inventory availability; (e) accurate orders; and (f) availability of delay information; scale: seven point Likert scale from 1 (strongly disagree) to 7 (strongly agree)

Thus, 149 supplier questionnaires were mailed to those suppliers, covering questions related to customer service performance, the use of preventive and corrective actions, and perceived FIQ from that customer. Some 136 supplier responses were returned. This corresponds to a response rate in the supplier survey of 91.3 percent (136/149). The distribution by industry and company size varied. Altogether, ten different industries were represented among the supplier responses, compared to four among the customers. Among these 136 responses, suppliers representing wholesaling companies were subtracted, leaving 121 manufacturing suppliers. Hence, the study covers 121 matched customer-supplier dyads.

Reliability and validity Reliability was ensured by using a standardized, structured survey instrument with instructions for the respondents. Testing non-response bias is a part of a reliability assessment. In the customer survey, x 2 statistics revealed no significant differences at the p , 0.05 level between respondents and non-respondents regarding industry and company size. Owing to the fact that the population in the supplier survey was hard to define, and because of the prevailing high response rate, no analysis of non-response bias was undertaken in the supplier survey. Inter-item reliability for a scale is another part of a reliability analysis. This analysis was performed in connection with the survey instrument. As the study’s reliability was acceptable, an important prerequisite for validity was obtained. Validity was increased as scales were drawn from directly from existing sources and as pretests were conducted, as recommended by Flynn et al. (1990). Altogether, the study’s reliability and validity were assessed as acceptable.

The impact of FIQ

99

Findings State-of-the-art description of forecast access and its perceived information quality In 105 of the 121 customer supplier dyads studied (87 percent), suppliers were receiving forecast information. About 69 of the 80 (86 percent) MTS suppliers received forecast information, as compared to 35 of the 41 (85 percent) MTO suppliers. No significant difference in forecast access could be identified when MTS and MTO suppliers were separated. Forecast access was also studied between 2nd tier, 1st tier and OEMs. Out of 20 OEMs 19 (95 percent) received customer forecasts. This is a larger proportion than for the 2nd tier suppliers, where 20 of 24 suppliers (83 percent) received forecast information, and for the 1st tier suppliers, where 65 of 77 suppliers (84 percent) received forecast information. It was not possible to conduct significance tests (x 2) on the difference in forecast access between the groups due to the fact that some cells contained too few observations. The average perceived FIQ by suppliers for each variable is shown in Table V, together with the results of an ANOVA test (with Bonferroni’s pair-wise comparisons) for differences between the variables. The index SIQFOR is also shown in Table V. The reliable variable is significantly lower (i.e. the quality deficiency is higher) than in time and accurate. The accurate variable is significantly higher than the other variables. In Tables VI and VII the perceived FIQ is compared between MTS and MTO firms and between 2nd tier and 1st tier suppliers and OEMs. No significant difference was identified between MTS and MTO firms except for in time ( p , 0.10). Most quality variables were lower (i.e. the quality deficiency was higher) for the 2nd tier suppliers as compared to the 1st tier suppliers and the OEMs, but not significantly. FIQ variable In time Accurate Convenient to access Reliable SIQFOR (index)

N 105 104 102 103

Mean (SD)

Significant difference with variablea

5.43 5.86 5.37 4.77 5.34

Accurate reliable In time convenient to access reliable Accurate In time accurate

(1.71) (1.52) (1.75) (1.84) (1.47)

Notes: ANOVA with F-statistics: 15.02; ashows variables with significant different mean using Bonferroni’s tests

Table V. Perceived FIQ by all suppliers

IJOPM 27,1

100

The performance impact of forecast information access and forecast information quality In order to test H1 and H2 in a group of companies with somewhat homogeneous manufacturing strategies, they were only tested among the MTS suppliers. T-tests were used to analyze the significant differences in supply chain performance between suppliers with access to customer forecasts and suppliers without access to forecasts (H1). Tables VIII and IX present the findings. The findings of H1 (Table VIII) indicate that suppliers without access to forecasts use less corrective actions than suppliers with access to forecasts. However, the differences are not significant for any variable. There are also indications that suppliers use preventive actions to a larger extent than corrective actions. Especially, extensive is the use of safety stock in raw material and finished goods inventory. Suppliers without access to a forecast use safety stock more than suppliers with access to a forecast. The difference is significant ( p , 0.05) for the use of safety stock in finished goods inventory and significant ( p , 0.1) for the use of safety stock in raw material inventory. For the remaining preventive actions there are no significant differences. No significant difference in customer service performance between suppliers with and without access to forecasts was found on any variable. An explanation for the finding that only preventive actions are used to a significantly larger extent and that the extent of using corrective actions and customer service are not significantly different between suppliers receiving forecast information and those not receiving forecast information may be the high overall customer service requirements. To test H2, Pearson correlation was used to analyze the correlation between supply chain performance and FIQ. The hypothesis could not be verified. Table IX shows that the only significant correlation existed for the customer service variable “rush orders when needed” and SIQFOR. Discussion A large proportion of the suppliers were receiving customer forecasts. One reason for this may be that the suppliers are all rated as the customers’ most important supplier, FIQ variable

Table VI. Perceived FIQ by MTS/MTO suppliers

In time Accurate Convenient to access Reliable SIQFOR (index)

69 69 69 68

Mean (SD) 5.65 5.97 5.42 4.79 5.46

n MTO

(1.54) (1.37) (1.78) (1.78) (1.36)

35 34 33 34

Mean (SD) 4.97 5.65 5.30 4.79 5.12

(1.96) (1.79) (1.74) (1.95) (1.68)

Mean diff

T-value

0.68 0.32 0.12 0.00 0.34

1.94 * 1.02 0.31 0.00 1.12

Note: *significance at the p , 0.10 level

FIQ variable Table VII. Perceived FIQ by 2nd tier/1st tier/OEM suppliers

n MTS

In time Accurate Convenient to access Reliable SIQFOR (index)

n 2nd tier Mean (SD) n 1st tier Mean (SD) n OEM Mean (SD) F-stats 20 19 19 19

5.25 5.58 4.89 4.26 4.89

(1.45) (1.71) (1.82) (1.85) (1.50)

65 65 64 64

5.45 5.98 5.42 4.88 5.44

(1.85) (1.45) (1.76) (1.83) (1.86)

19 19 19 19

5.53 5.74 5.74 5.05 5.51

(1.58) (1.59) (1.66) (1.78) (1.58)

0.13 0.54 1.11 1.17

Corrective action variable Subcontracting Expediting Part delivery Re-scheduling Reservation breaking Overtime Express transports Preventive action variable Safety stock in raw material inventory Safety stock in finished goods inventory Safety capacity Safety lead time Over-planning Customer service performance variable Promised lead time On-time delivery Rush orders when needed Promised inventory availability Accurate orders Availability of delay information

Mean no forecast2 forecast ¼ mean difference

T-value

1.71 2 1.98 ¼ 2 0.27 3.63 2 4.26 ¼ 2 0.63 2.90 2 3.17 ¼ 2 0.27 3.10 2 4.06 ¼ 2 0.96 2.00 2 2.35 ¼ 2 0.35 3.20 2 3.65 ¼ 2 0.45 2.60 2 2.74 ¼ 2 0.14

2 0.47 2 0.97 2 0.45 2 1.76 2 0.74 2 0.73 2 0.24

6.10 2 4.96 ¼ 1.14 5.50 2 3.97 ¼ 1.53 2.50 2 3.40 ¼ 2 0.9 3.00 2 3.37 ¼ 2 0.37 2.00 2 3.03 ¼ 2 1.03

1.86 2.21 * 2 1.44 2 0.55 2 1.54

6.36 2 6.03 ¼ 0.33 6.36 2 5.99 ¼ 0.38 6.18 2 6.11 ¼ 0.07 6.00 2 5.69 ¼ 0.31 6.27 2 6.46 ¼ 2 0.18 5.91 2 5.88 ¼ 0.03

1.09 1.25 0.21 0.69 2 0.80 0.07

Note: *significance at the p , 0.05 level

Corrective action variable Subcontracting Expediting Part delivery Re-scheduling Reservation breaking Overtime Express transports Preventive action variable Safety stock in raw material inventory Safety stock in finished goods inventory Safety capacity Safety lead time Over-planning Customer service performance variable Promised lead time On-time delivery Rush orders when needed Promised inventory availability Accurate orders Availability of delay information Note: *significance at the p , 0.01 level

Pearson correlation with SIQFOR

Sig (two-tailed)

20.05 0.11 20.12 0.03 0.21 20.02 20.05

0.72 0.35 0.30 0.83 0.10 0.87 0.67

0.11 0.02 20.04 20.12 20.04

0.34 0.88 0.78 0.33 0.75

0.13 0.05 0.31 0.12 0.03 20.11

0.28 0.67 0.01 * 0.32 0.79 0.33

The impact of FIQ

101

Table VIII. Result from testing H1

Table IX. Result from testing H2

IJOPM 27,1

102

and a deeper cooperation than the average can be expected. If random suppliers had been chosen the proportion might have been lower. Still, the extent of exchange identified in this study is similar to other studies (Sandberg, 2005). The forecasting type or how frequently forecasts are exchanged were not studied either. Quality deficiencies were found for all FIQ variables. The accurate variable is, however, significantly higher than the other variables, which implies that suppliers perceive that forecasts are rather free from obvious mistakes. This could be expected since important and established customer/supplier dyads were studied, and accuracy should have been discussed and dealt with. Reliability was rated significantly lower. The reliability of FIQ, i.e. the probability that a forecast will remain unchanged, could be interpreted as the forecast error. Convenient to access and in time were also rated significantly lower. No identified study measured the information quality of forecasts. However, Closs et al. (1997) found convenient to access to be the variable with the lowest quality when measuring order information quality. This result might be related to how forecasts are communicated between customer and supplier. Forecasts might arrive on fax or in an excel spread sheet, which implies that the customer needs to do a considerable amount of work to be able to use it. The aggregation level might also be unadapted to the customer’s needs, for example, be on product group level where a product level would have been preferred. The study indicated that there are large quality deficiencies among 2nd tier suppliers as compared to 1st tier suppliers and OEMs. However, the differences are not significant, partly due to the low number of OEMs and 2nd tier suppliers in the study. The quality variable in time was considered significantly lower for MTO compared to MTS suppliers. Forecasts are an important input to capacity planning in MTO companies and to priority planning and execution in MTS companies. The lack of timely forecasts in MTO companies may, consequently, impact their capacity planning processes. When testing H1, the only significant difference in supply chain performance between suppliers with and without access to forecasts, was related to the use of safety stock in finished goods inventory. The customer service did not differ between the groups. Consequently, it seems to be necessary for all suppliers to provide high customer service. Here, they were all selected as the customers’ most important supplier, which may further explain this focus. The conclusion that exchange of forecast information has a positive performance impact is logical and in line with previous research (Fliedner, 2003; Thonemann, 2002; Aviv, 2001; Zhao et al., 2002; McCarthy and Golicic, 2002). H2 was not verified. Nonetheless, it is important to ask whether or not FIQ matters in this context. This study was characterized by a large spread in the supplier sample; it consisted of both large and small suppliers in ten industries. It is possible that customer demand is predictable, an aspect that was not measured. Information quality is perhaps not so important when demand is predictable, as claimed by Cachon and Fisher (2000). Sma˚ros (2003) found that the single most important obstacle for collaborative forecasting was the customer’s lack of forecasting processes. Research on the managerial side of forecasting, i.e. how forecasts are handled, was called for by Wacker and Lummus (2002). Our study has not revealed information about the customer’s own forecasting process. It is also worth repeating that this study was not focused on collaborative

forecasting, but merely asked if forecast information was transferred or not. This study was a first attempt to measure the performance impact of FIQ but it was limited, especially in terms of the number of surveyed companies of similar types. The study reveals the need for more studies in the area of FIQ. One potential area of research should deal with explaining the causes of high or low perceived FIQ. Another should be about the performance impact of FIQ. More detailed research questions in these two areas could be related to the conditions of the actual demand pattern and the processes related to actual forecasting at the customer, transmission of the information from the customer to the supplier, registration of the data at the supplier and the characteristics of the processes and actors using the forecast data. The following are issues for such further studies: . Customer’s demand pattern. The predictability of the demand influences the need for forecasts and, consequently, impacts the need for exchange and the potential performance impact of forecast exchange and its information quality. Forecast information could thus be expected to have different performance impacts in different demand situations. . Customer’s forecast design. The customer’s decisions about how to present the forecasts should have an information quality impact (e.g. the aggregation level and the exchange frequency). Different FIQ could, for example, be expected between situations where the supplier is involved in the design or if a CPFR-style collaboration is established compared to when that is not the case. . Customer’s forecasting process. The way the process of generating forecasts is performed and how well it is functioning should impact the exchange of forecasts and the perceived information quality. Differences could, for example, be expected between situations where the supplier, in a consensus forecasting manner, is involved in the data collection and manual evaluation and adjustment of forecasts or if a CPFR-style collaboration is established compared to when that is not the case. . Customer’s forecast transferring method. The interface between customer and supplier, for example, the use of information and communication technology, could impact the exchange frequency and the information quality. It could, for example, be expected that using EDI or web-EDI results in the exchange of more timely but also more accurate forecasts. . Registration of customer forecasts at the suppliers. The registration of received forecasts could impact the information quality, for example, if it is automatically registered through EDI, electronically transferred from an Excel sheet or manually typed into the ERP system. . Supplier’s manufacturing process where the forecast is used. The forecast could be used in situations with different supply strategies (e.g. MTS or MTO), in different planning processes (e.g. with a priority or capacity planning perspective), on different levels of detail (e.g. master planning or more detailed material planning based on material requirements planning or re-order point systems), and on different planning horizons. The different users place different requirements on the forecast and could consequently interpret the same forecast as having different information quality.

The impact of FIQ

103

IJOPM 27,1

.

104 .

Who is the forecast information customer? Not only the specific process at the supplier where the forecast is used as input judges the information quality of the forecast, but also the actor operating the process and his/her support. Issues such as user position, knowledge, experience and software support impact the perceived information quality, how well the supplier may utilize the forecast information received from the customer, and consequently also the supply chain performance effects related to forecast information. The performance impact. Here, supply chain performance impact was measured as the perceived customer service and the extent of using preventive and corrective actions. The performance impact could be more tightly linked to the supplying company’s materials planning or manufacturing process and be measured as the inventory turnover rate, inventory days on hand or the perceived overall performance of manufacturing. Also the user satisfaction among the supplier’s planning personnel could be a relevant and interesting measure.

Conclusions The study contains two types of conclusions, those developed from the conceptual discussion in the theoretical framework and those of the empirical study. In the theoretical framework, measurement instruments for FIQ and supply chain performance – the combined effect of corrective actions, preventive actions and customer service performance were developed. FIQ was measured as the four individual variables: in time, reliable, accurate and convenient to access, and as an index of the four variables. The measures were tested and used with reliable results in the empirical study. The empirical findings indicated that a large proportion of the suppliers receive customer forecasts, but that the FIQ is lower further upstream in the supply chain. It also showed that the quality variable in time was considered significantly lower for MTO suppliers compared to MTS suppliers, which emphasized the importance for MTO companies to receive good forecasts for supply and capacity planning. The only significant difference in supply chain performance found between suppliers with and without access to forecast information was related to the use of safety stock in finished goods inventory. The hypothesis testing only considered MTS companies, which may explain why the finished good inventory was the most important safety mechanisms against forecast uncertainty. Research on supply chain information quality is very scarce. Further research is needed to develop the understanding of the performance impact of supply chain information quality, especially FIQ. This study identified several relationships, but it was conducted on a sample with a large spread and in a limited number of companies. Only one of the two hypotheses generated was verified. Therefore, it would be valuable to replicate the study in other empirical settings and for companies with other supply strategies than MTS. Case studies focusing on the actual demand pattern and the processes related to actual forecasting at the customer, transmission of the information from the customer to the supplier, registration of the data at the supplier and the characteristics of the processes and actors using the forecast data could further develop the understanding of the usability of customer forecasts and the perceived performance impact of FIQ.

References Aviv, Y. (2001), “The effect of collaborative forecasting on supply chain performance”, Management Science, Vol. 47 No. 10, pp. 1326-43. Brewer, P.C. and Speh, T.W. (2000), “Using the balanced scorecard to measure supply chain performance”, Journal of Business Logistics, Vol. 21 No. 1, pp. 75-93. Cachon, G.P. and Fisher, M. (2000), “Supply chain inventory management and the value of shared information”, Management Science, Vol. 46 No. 8, pp. 1032-48. Christopher, M. and Towill, D. (2000), “Supply chain migration from lean and functional to agile and customized”, Supply Chain Management: An International Journal, Vol. 5 No. 4, pp. 206-13. Clikeman, P.M. (1999), “Improving information quality”, The Internal Auditor, Vol. 56 No. 3, pp. 32-3. Closs, D.J., Goldsby, T.J. and Clinton, S.R. (1997), “Information technology influences on world class logistics capability”, International Journal of Physical Distribution & Logistics Management, Vol. 27 No. 1, pp. 4-17. English, L.P. (1999), Improving Data Warehouse and Business Information Quality, Wiley, New York, NY. Ericsson, J. (1997), “Sto¨rningsanalys av tillverkningssystem – ett viktigt verktyg inom lean production”, doctoral thesis, Department of production and materials engineering, Lund University, Lund. ´ Fahlen, K. (1997), Sto¨rningars konsekvenser fo¨r tillverkande fo¨retags effektivitet – identifiering, analys och hantering av sto¨rningar, Handelsho¨gskolan, Go¨teborgs universitet, Go¨teborg (in Swedish). Forslund, H. (2004), “The existence of logistics quality deficiencies and the impact of information quality in the dyadic order fulfillment process”, Dissertation No. 62, Linko¨ping University, Linko¨ping, EKI. Fliedner, G. (2003), “CPFR: an emerging supply chain tool”, Industrial Management & Data Systems, Vol. 103 No. 1, pp. 14-21. Flynn, B.B., Sakakibara, S., Schroeder, R.G., Bates, K.A. and Flynn, E.J. (1990), “Empirical research methods in operations management”, Journal of Operations Management, Vol. 9 No. 2, pp. 250-84. Frohlich, M.T. and Westbrook, R. (2001), “Arcs of integration: an international study of supply chain strategies”, Journal of Operations Management, Vol. 19, pp. 185-200. Hair, J.F., Anderson, R.E., Tatham, R.L. and Black, W.C. (1998), Multivariate Data Analysis, Prentice-Hall, Upper Saddle River, NJ. Helms, M., Ettkin, L. and Chapman, S. (2000), “Supply chain forecasting: collaborative forecasting supports supply chain management”, Business Process Management Journal, Vol. 6 No. 5, pp. 392-407. Holmstro¨m, J., Fra¨mling, K., Kaipia, R. and Saranen, J. (2002), “Collaborative planning forecasting and replenishment: new solutions needed for mass collaboration”, Supply Chain Management: An International Journal, Vol. 7 No. 3, pp. 136-45. Keebler, J.S., Manrodt, K.B., Durtsche, D.A. and Ledyard, D.M. (1999), Keeping Score – Measuring the Business Value of Logistics in the Supply Chain, CLM, Oak Brook, IL. Kelle, P. and Akbulut, A. (2005), “The role of ERP tools in supply chain information sharing, cooperation and cost optimizing”, International Journal of Production Economics, Vol. 93-94, pp. 41-52.

The impact of FIQ

105

IJOPM 27,1

106

Lee, H., Padmanabhan, V. and Whang, S. (1997), “Information distortion in a supply chain: the bullwhip effect”, Management Science, Vol. 46 No. 4, pp. 546-58. Lee, Y.W., Strong, D.M., Kahn, B.K. and Wang, R.Y. (2002), “AIMQ: a methodology for information quality assessment”, Information & Management, Vol. 40, pp. 133-46. Lindau, R.A. (1995), “The impact of high-quality information on performance in manufacturing”, doctoral thesis, Chalmers University of Technology, Go¨teborg. Lindau, R. and Lumsden, K. (1993), Disturbance Absorption Actions Used in Material Flow Systems – A Pilot Study, Department of Transportation and Logistics, Chalmers University of Technology, Go¨teborg. McCarthy, T. and Golicic, S. (2002), “Implementing collaborative forecasting to improve supply chain performance”, International Journal of Physical Distribution & Logistics Management, Vol. 32 No. 6, pp. 431-54. Mattsson, S-A. (2002), Logistik i fo¨rso¨rjningskedjor, Studentlitteratur, Lund (in Swedish). Mentzer, J.T., Min, S. and Zacharia, Z.G. (2000), “The nature of interfirm partnering in supply chain management”, Journal of Retailing, Vol. 76 No. 24, pp. 549-68. Moberg, C.R., Cutler, B.D., Gross, A. and Speh, T.W. (2002), “Identifying antecedents of information exchange within supply chains”, International Journal of Physical Distribution & Logistics Management, Vol. 32 No. 9, pp. 755-70. Petersen, K.J. (1999), “The effect of information quality on supply chain performance: an interorganizational information system perspective”, dissertation, Michigan State University, East Lansing, MI. Sandberg, E. (2005), “Logistics collaboration in supply chains – a survey of Swedish manufacturing companies”, Thesis No. 1180, Linko¨ping Studies in Science and Technology, Linko¨ping University, Linko¨ping. Sma˚ros, J. (2003), “Collaborative forecasting: a selection of practical approaches”, International Journal of Logistics: Research and Applications, Vol. 6 No. 4, pp. 245-59. Stank, T.P., Emmelhainz, M.A. and Daugherty, P.J. (1996), “The impact of information on supplier performance”, Journal of Marketing Theory and Practice, Fall, pp. 94-105. Stock, J. and Lambert, D. (1992), Strategic Logistics Management, Irwin Professional, New York, NY. Thonemann, U. (2002), “Improving supply-chain performance by sharing advanced demand information”, European Journal of Operational Research, Vol. 142, pp. 81-107. van der Vorst, J. and Beulens, A. (2002), “Identifying sources of uncertainty to generate supply chain redesign strategies”, International Journal of Physical Distribution & Logistics Management, Vol. 32 No. 6, pp. 409-30. Vollmann, T., Berry, W., Whybark, C. and Jacobs, R. (2005), Manufacturing Planning and Control for Supply Chain Management, McGraw-Hill, New York, NY. Wacker, J.G. and Lummus, R.R. (2002), “Sales forecasting for strategic resource planning”, International Journal of Operations & Production Management, Vol. 22 No. 9, pp. 1014-31. Weld, L. (1916), The Marketing of Farm Products, Macmillan, New York, NY. Whipple, J.M., Frankel, R. and Daugherty, P.J. (2002), “Information support for alliances: performance implications”, Journal of Business Logistics, Vol. 23 No. 2, pp. 67-82. Zhao, Y. (2002), “The value of information sharing in a two-stage supply chain with production capacity constraints: the infinite horizon case”, Manufacturing & Service Operations Management, Vol. 4 No. 1, pp. 21-4.

Zhao, X., Xie, J. and Leung, J. (2002), “The impact of forecasting model selection on the value of information sharing in a supply chain”, International Journal of Production Economics, Vol. 142, pp. 321-44.

The impact of FIQ

Further reading Mason-Jones, R., Naylor, B. and Towill, D. (2000), “Engineering the leagile supply chain”, International Journal of Agile Management Systems, Vol. 2 No. 1, pp. 54-61.

107

Corresponding author Patrik Jonsson can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0144-3577.htm

IJOPM 27,1

108

Structuring front office and back office work in service delivery systems An empirical study of three design decisions Leonieke G. Zomerdijk Operations and Technology Management Department, London Business School, London, UK, and

Jan de Vries Production Management Department, Faculty of Management and Organisation, University of Groningen, Groningen, The Netherlands Abstract Purpose – The aim of this paper is to investigate how the distinction between contact and non-contact activities influences the design of service delivery systems and to identify key design decisions for structuring front office and back office work. Design/methodology/approach – Building on current literature, the paper identifies three design decisions and associated performance trade-offs. The design decisions regard the degree of customer contact in the process, the decoupling of activities and the grouping of employees. The design decisions and the trade-offs are empirically validated in five case studies of 15 service delivery systems in the financial services sector. Findings – Distinguishing between the three design decisions is more suitable for describing today’s practices than traditional front office – back office thinking. For each design decision a trade-off was observed consisting of several design considerations. However, the trade-offs do not involve the weighing of one set of performance objectives against another, as the design choices contribute to the same objectives, yet in different ways. Research limitations/implications – This study concentrated on a limited number of cases in the financial services sector. The contents of the trade-offs should be tested on a larger scale and in different industries. In order to develop design guidelines, future research should also examine the impact of contingency factors, such as the service being delivered and strategic priorities. Originality/value – The three design decisions and the trade-offs improve understanding of the impact of customer contact on a service organisation and provide support for designing service delivery systems in practice. Keywords Service operations, Process planning, Customer relations, Service delivery systems, Case studies Paper type Research paper

International Journal of Operations & Production Management Vol. 27 No. 1, 2007 pp. 108-131 q Emerald Group Publishing Limited 0144-3577 DOI 10.1108/01443570710714565

Introduction Compared to delivering goods, one of the most distinguishing features of delivering services is the amount of customer contact involved in the service delivery system, following from the simultaneity of production and delivery of services (Nie and Kellogg, 1999; Sasser et al., 1978; Parasuraman et al., 1985). Many services cannot be delivered without the customer being present, interacting or participating in the service

delivery system. This customer influence provides a source of complexity that is not generally found in manufacturing operations (Nie and Kellogg, 1999). In general, customer contact introduces uncertainties and variation in the service delivery system and makes demands on the design of facilities, staff and technology in the production system (Chase, 1978; Safizadeh et al., 2003). On the other hand, customer contact provides valuable opportunities for responding to a customer’s needs and cross-selling other products (Chase et al., 1984; Chase and Hayes, 1991). Several authors have addressed the design issues following from the impact of customer contact on a service delivery system. According to the customer contact approach (Chase, 1978, 1981; Chase and Tansik, 1983), customer contact activities should be decoupled from non-contact activities to do justice to their different design requirements and maximise the efficiency of the service delivery system. Consequently, service organisations consist of a front office and a back office. Recently, Metters and Vargas (2000) added that in several situations coupling front office and back office jobs can be a viable strategy, meaning that the same employees carry out both front office activities (FO) and back office activities (BO). In addition, they argue that both coupled and decoupled designs can support low cost and high service strategies. However, service system design, in particular front office and back office issues, requires further attention (Chase, 1996; Johnston, 1999, 2005; Hill et al., 2002). First, the current contributions often take a dichotomous perspective, assuming that service organisations focus on either efficiency or cross-selling, either low costs or high service (Chase and Bowen, 1989; Metters and Vargas, 2000). In today’s competitive environment, however, most organisations cannot afford to concentrate on just one set of performance objectives. Instead, they are forced to perform well at multiple criteria simultaneously (Ferdows and De Meyer, 1990; Roth and Van der Velde, 1991; Boyer and Lewis, 2002). Research shows that it is possible to overcome several performance trade-offs, largely due to developments in information technology (Skinner, 1992; McDermott et al., 1997; Menor et al., 2001). Investigating the impact of customer contact in more detail to see which performance trade-offs still hold and which ones can be overcome, is required. Second, given the developments in technology, the nature of customer contact is changing (Voss, 2003). Whereas Chase (1978) concentrated on physical customer contact, i.e. the customer being present in the service delivery system, today multiple forms of customer contact can be identified, such as on the phone or mediated by technology, e.g. the internet. It is clear that these kinds of contact make different demands on a service delivery system than physical customer presence, so it is important to explore their implications for the design decisions regarding front office and back office work. Third, although the literature is quite clear on the division of service organisations into front office and back office parts, in practice the situation is not so straightforward. Service organisations, particularly financial institutions, are setting up so-called “mid-offices” to act as a platform between front office and back office parts and operate different kinds of back offices, e.g. closely linked to a local front office or centralised shared service centres. Together with the other two developments, this calls for more research on structuring front office and back office work in service delivery systems. The aim of this paper is to investigate how the distinction between contact and non-contact activities influences the design of service delivery systems and to identify key design decisions for structuring front office and back office work. Design decisions

Structuring front office and back office work 109

IJOPM 27,1

110

are decisions that have to be made when designing a service delivery system, for example, about the number of process steps and the division of labour. In this paper, three design decisions that play a part in structuring front office and back office work are proposed: (1) deciding where customer contact should occur in a service delivery process; (2) deciding which activities in the process should be decoupled from each other; and (3) deciding how the employees that are involved in the process should be grouped together. Using empirical evidence from five in-depth case studies in financial service organisations, these design decisions and their underlying trade-offs are elaborated in more detail. The remainder of this paper is structured as follows. To begin with, we review current literature addressing front office and back office design decisions and the impact of customer contact on service delivery systems. This leads to the proposition of three design decisions and a definition of customer contact. Next, the design of the empirical study is explained and the main case study results are presented. The subsequent section presents the data analysis, concentrating on validating the design decisions and elaborating the trade-offs. Finally, the paper ends with a number of conclusions regarding the structuring of front office and back office work in service delivery systems and suggestions for future research. Front office and back office design decisions The front office – back office model of service organisations is probably the most common way of conceptualising the impact of customer contact on a service delivery system. Although different terms are in use, such as “on stage” “front stage” “back stage” “front line” and “back room” “front office” and “back office” seem to be most common. The front office is the part where activities that require customer contact take place and as such is directly experienced by customers, whereas the back office contains processes that are carried out remotely from customers and hence cannot be seen or experienced by customers (Shostack, 1982, 1984; Grove and Fisk, 1992; Johnston and Clark, 2001). The front office – back office model is based on the general notion that FO are carried out by front office employees in front office departments and vice versa for BO. In this paper, however, we propose that FO and BO do not necessarily have to be allocated to different employees and that front office and back office employees do not necessarily have to be grouped in different groups. Instead, it is suggested three separate design decisions are involved in structuring front office and back office work. They can be derived from the current body of knowledge and are described below. The customer contact decision The first design decision we propose goes back to the customer contact approach put forward by Chase (1978, 1981) and Chase and Tansik (1983). According to this approach, not all services are equal in terms of what they can achieve in the way of efficiency, because of differences in the extent of customer contact in the creation of the service. Customer contact refers to the physical presence of the customer in the service delivery system. The basic premise of the customer contact approach is that

high-contact activities are more difficult to control and to rationalise than low-contact activities due to all kinds of disturbances caused by the customer. The low-contact part or “technical core” or a service process can be sealed off from the environment, thus generating a higher degree of efficiency (Thompson, 1967). High-contact activities also require interpersonal skills and need to be located close to the customer, while non-contact activities often require technical skills and can be located close to resources. In order to maximise the efficiency of the system, Chase and Tansik (1983) recommend reducing the ratio of high-contact to low-contact work. Following the customer contact approach, the main design decision involved in structuring front office and back office work is deciding how much customer contact is required in a particular service delivery process, or deciding which activities are carried out with customer contact and which ones without. This decision is not necessarily straightforward and involves a trade-off between the efficiency potential of BO and the sales opportunities related to FO (Chase et al., 1984; Chase and Bowen, 1989). Furthermore, the contact between customer and service provider in FO offers opportunities for tailoring a service to a customer’s needs (Chase and Hayes, 1991). This trade-off is summarised in Table I.

Structuring front office and back office work 111

The decoupling decision The second design decision we propose regards the subject of coupling or decoupling front office and back office work. Following the distinct characteristics of high-contact and low-contact activities, the customer contact approach recommends decoupling them from each other and placing them under separate supervisions, resulting in a front office and a back office (Chase and Tansik, 1983). An important aspect is separation of functions: all high-contact activities should be performed by one group of people, all low-contact activities by another (Chase, 1978, p. 140). In this way, front office and back office parts can be run as “plants within a plant” (Skinner, 1974), with their own objectives, operating procedures and recruitment policies. Larsson and Bowen (1989) and Metters and Vargas (2000) argue there are more ways to structure FO and BO than the decoupling recommended by the customer contact approach. Larsson and Bowen (1989) present a typology with different constellations of customers, front office Customer contact decision: determining which activities are carried out with customer contact and which ones without Front office activities Back office activities Cross-selling Efficiency potential Customising or personalising a service Decoupling decision: organisationally separating activities by allocating them to different employees Coupled processes Decoupled processes Maximum flexibility and responsiveness Employing experts High degree of front office utilisation Sealing off BO from uncertainties Broad tasks Specialisation effects Grouping decision: grouping employees that are consecutively involved in a service delivery system in market or functional groups Market groups Functional groups Workflow coordination Economies of scale Cross-fertilisation More specialisation and uniformity

Table I. Three design decisions and the trade-offs involved

IJOPM 27,1

112

employees and back office employees. Metters and Vargas (2000) show that the customer contact approach is only one of the possible strategies for structuring front office and back office work. They explain that under different strategic conditions, different decoupling strategies for FO and BO are appropriate. They define decoupling as breaking a process into its component back office and FO, segregating those activities into distinct back and front office jobs, and, usually, geographically separating the back and front offices. Their typology differentiates between decoupling for cost reasons, decoupling to improve non-cost objectives (e.g. quality or speed), coupled processes to decrease costs and coupled processes to achieve non-cost objectives. For example, FO and BO can be coupled, i.e. one employee handles both contact and non-contact activities, to achieve high-levels of flexibility and responsiveness. In that case, there are no back office departments, nor back office employees. Thus, high-contact activities and non-contact activities need not necessarily be separated from each other. The second design decision in the structuring of front office and back office work, therefore, regards decoupling FO and BO. In this paper, decoupling is defined as organisationally separating activities by allocating them to different employees. This design decision can also be characterised as a trade-off. On the one hand, decoupling brings about the benefits of employing experts for particular tasks, matching personality types with task requirements and sealing off BO from customer-induced uncertainties. Coupled processes, on the other hand, often have more flexibility and responsiveness and lead to a better utilisation of front office capacity, as front office employees can use the BO to fill their idle time (Metters and Vargas, 2000). In addition to these factors, the decoupling decision can be considered somewhat similar to the issue of division of labour. From this perspective, decoupled processes lead to narrow tasks and coupled processes to more broad tasks. Each approach has its own costs and benefits, for example, in terms of training time, mental (over-) load, utilisation levels and absenteeism (Campion and Thayer, 1987). This trade-off is summarised in Table I. The grouping decision The reasoning of Chase and Tansik (1983) and Metters and Vargas (2000) gives rise to a third design decision that is not yet part of the front office – back office debate: the grouping of front office and back office employees. Both Chase and Tansik (1983) and Metters and Vargas (2000) presuppose that when high-contact activities and non-contact activities are decoupled from each other, this leads to the establishment of separate front office and back office departments. Yet, we argue this does not have to be the case, as the employees can also be grouped in one department and work closely together. Following developments in practice, we propose that the grouping of front office and back office employees requires a separate design decision that determines whether the employees are grouped in separate or common groups. Put differently, when FO and BO are decoupled from each other, one still has a choice between operating a centralised back office department in some remote location far away from the front office (“the way-back, back office”) or assigning one back office employee and one front office employee to small teams and have them share the same office. Therefore, the grouping of employees that are consecutively involved in a service delivery system is proposed as the third design decision. A number of bases for clustering employees can be identified, such as grouping by knowledge or skill, by work process and function, by time, by output, by client or by place.

Mintzberg (1979) compresses these bases to two essential ones: by market (covering output, client and place) and by function (incorporating knowledge, skill, work process and function). Functional groups consist of employees that perform the same function, whereas market groups consist of employees that perform different functions, but work for the same output, client or area. The composition of a group or team is one of the factors that influences team performance and outcomes (Mintzberg, 1979; Ancona and Caldwell, 1992; Jackson et al., 2003). For example, the intra-group coordination in functional groups of employees often leads to cross-fertilisation among those employees and emphasises further specialisation and uniformity of the way in which the work is carried out. Furthermore, individual activities can often be carried out more efficiently in functional groups than in other groups, given the benefits of scale. Yet, functional grouping almost automatically implies that coordination with other functional groups is more difficult, meaning workflow interdependencies often suffer. In market groups, workflow interdependencies are the main reason for grouping, as coordination regarding the workflow is facilitated which often contributes to the speed and quality of a process (Mintzberg, 1979). Hence, there is a trade-off involved between workflow coordination on the one hand, and scale economies, cross-fertilisation and further specialisation and uniformity on the other hand. This is summarised in Table I. Front office – back office configurations As argued above, structuring front office and back office work in a service organisation involves three individual design decisions, that do not necessarily lead to the front office – back office model of service organisations. In order to capture this cluster of design decisions and the large number of design options that follow from different combinations of these decisions, we introduce the term “front office – back office configurations”. A front office – back office configuration is a particular way of structuring front office and back office work in a service delivery system, resulting from the customer contact decisions, the decoupling decisions and the grouping decisions. Together, these decisions determine the basic shape or structure of a service delivery system and also indicate the nature of the performance to be expected, i.e. which performance objectives will be easy to achieve and which ones will be more difficult. We propose that this way of viewing service organisations is more suitable for capturing today’s practices than the front office – back office model. A definition of customer contact Before investigating front office – back office configurations empirically, a definition of customer contact is required. Reviewing the literature reveals there are several definitions or interpretations of what constitutes customer contact. According to Chase (1978) it is physical presence of the customer in the system, while according to Schmenner (1986) it is the interaction between the customer and service provider and the customisation of the service being delivered that make demands on the design of service delivery systems. Mersha (1990) combines customer presence, either face-to-face or mediated through technology, with low- and high-levels of interaction. From the perspective of designing service delivery systems, we define customer contact as: a direct encounter between a customer and a service provider that takes place in the same time but not necessarily in the same place, and has the opportunity for interaction.

Structuring front office and back office work 113

IJOPM 27,1

114

First, this definition implies that physical presence is no longer a prerequisite. Following the basic logic of the customer contact approach, the bottom-line is that non-contact activities have more efficiency potential than high-contact activities, because there are fewer uncertainties involved. The uncertainties in high-contact activities stem from irregular arrival patterns, non-standard customer demands and the participation of the customer in the system. Put differently, it is hard to predict when customers will come, what they will want and whether they will be prepared for their role. Moreover, high-contact activities make demands on the skills of employees and facilities accommodating the customer that non-contact activities do not. As interacting with a customer over the phone, a chat or a web cam can be characterised by the same uncertainties and requires interpersonal skills as well, this should be classified as customer contact. The main criterion is the opportunity or need for direct feedback and the associated absence of delays in the communication. Second, our definition points out that actual interaction between customer and service provider is not required; instead, the opportunity for interaction is sufficient. From the perspective of customer-induced uncertainties, whether there is interaction or not does not alter the uncertainties as we described them. In fact, if interaction were considered an indispensable element in the definition of customer contact, activities that are carried out in the customer’s presence (either face-to-face or mediated through technology) but without interaction, would have to be classified as non-contact activities. Yet, it is the customer’s presence in the system, not the interaction, that deprives them of the efficiency potential associated with non-contact activities. Therefore, interaction is not included as a precondition for customer contact. Design of empirical study The three design decisions and the associated performance trade-offs proposed in this paper have been investigated empirically. The main objectives of the study were to find empirical support for the design decisions, to investigate the coherence between them and to elaborate the trade-offs using empirical data. We conducted a multiple case study, involving 15 front office – back office configurations in the financial services sector. The cases were aimed at describing the front office – back office configurations and identifying the design considerations that underlie them. Case studies are a particularly appropriate research strategy for studying such “how” and “why” questions (Yin, 2003). They also provide the benefits of studying the front office – back office configurations holistically and in their organisational context (Meredith, 1998; Yin, 2003). Although virtually every service provider will have to deal with a combination of FO and BO in its service delivery systems, designing the service delivery systems will not be equally complex for the diversity of services that exist in the real world. This study concentrates on the financial services sector, for this industry, on the one hand, illustrates the complexities involved in structuring FO and BO and, on the other hand, will benefit from the insights to be developed. As such, the financial services sector provides a rich empirical domain for this study. This is driven by the prevalence of mixed services (i.e. services that are characterised by a mixture of FO and BO), the large impact of information technology and the strong demands for outstanding performance in terms of efficiency, quality, flexibility, reliability and speed, all at the same time.

The study sample consists of five banks (labelled banks A, B, C, D and E to preserve their identity) from a cooperative banking group in The Netherlands. The banking group is one of the three largest players in The Netherlands for both business and individual clients. It is owned by the approximately 300 member banks and further consists of a number of specialised subsidiaries and a central support organisation. The banks in this group have a reasonable amount of local autonomy with regard to the design decisions relating to service delivery systems. Therefore, they provide a rich environment for studying front office – back office configurations and their underlying design considerations. We employed a theoretical sampling model to find cases that provide the best opportunity to learn (Stake, 1994). Our strategy was based on achieving literal replication (Yin, 2003), using information-rich cases that were distributed for maximum variation (Miles and Huberman, 1994). The banks in the sample were known within the group for displaying careful design decisions, making them information-rich. In addition, we looked for as much variation in design decisions and underlying design considerations as possible. Owing to the local autonomy, a considerably varied sample could be created. In this way, the possible bias from sampling from one organisational context could be reduced. Furthermore, we included banks that differed in terms of size and regional area (Table II). Our unit of analysis is the service delivery system or front office – back office configuration. In each bank the front office – back office configurations for the same three service delivery systems were studied: the service delivery systems for providing mass consumer products, mortgages and business loans. This increased the breadth of our study and facilitated cross-case comparisons. We selected these processes for being common to the banks and other financial institutions (assuming common processes would have received significant design attention and would be easier to generalise) and for variation with regard to the degree of customisation of the service being delivered, as the degree of service customisation often influences design decisions (Huete and Roth, 1988; Kellogg and Nie, 1995). This design led to a total of 15 front office – back office configurations in our study. The data were collected according to a consistent case study protocol, using multiple sources of evidence. First, we used company documents, such as formal process descriptions, strategy plans and documentation from (re-) design projects. Second, we conducted a large number of semi-structured interviews with a list of target respondents that was consistent over cases, i.e. front office employees, back office employees, process designers and bank managers. Across the cases, 79 interviews of 1.5 hours each were conducted. An overview of the respondents is provided in Table III. Third, we added to the interviews and documents through direct observations of work in progress at the case study sites. Triangulating data collection methods and using

Regional area Branches Employees (FTE) Total of assets (EUR billions) Market penetration (per cent)

Bank A

Bank B

Bank C

Bank D

Bank E

Rural 4 91 0.4 40

City 7 120 0.6 20

Rural 13 115 0.8 60

City 10 375 1.9 30

Rural 9 221 1.1 70

Structuring front office and back office work 115

Table II. Description of cases in sample

Team leaders for each process

General manager Process engineer

Process manager

Board members for internal and commercial affairs

B

C

D

E

Total number of interviews in study

Process engineer

A

Table III. Overview of interview respondents

Introductory interviews

Support employee Process engineer Manager

Support employee Process engineer Manager

Consumer advisor

Mortgage advisors (2) Support employees (2) Process engineer Manager Mortgage advisor

Mortgage advisors (2) Support employees (3) Process engineer Manager Mortgage advisors (2) Support employees (2) Process engineer Manager Mortgage advisors (2) Manager

Providing mortgages

Consumer advisors (2) Process engineer Manager

Consumer advisors (2) Support employee Process engineer Manager Consumer advisors (2) Manager

Consumer advisors (2) Process engineer Manager

Providing mass consumer products

Support employee Process engineer Manager

Business advisor

Business advisors (2) Support employees (2) Process engineer Manager Business advisor Support employee Manager Business advisors (2) Support employees (2) Manager

Business advisors (2) Support employees (3) Manager

Providing business loans

116

Bank

79

14

16

11

20

18

No.

IJOPM 27,1

multiple interview respondents is expected to reduce the effects of single-rater bias and recollection biases (Boyer and Verma, 2000; Leonard-Barton, 1990). The data were analysed through a process of data reduction, data display and conclusion drawing and verification (Miles and Huberman, 1994). Data reduction mainly took place through coding the interviews using software for qualitative data analysis. Subsequently, the “story” for each front office – back office configuration was developed, including a graphical display of the design decisions and a matrix containing the main design considerations. These within-case analyses formed the basis for cross-case analyses in order to detect cross-case patterns. Among other things, the case studies were used to identify the main design considerations for front office – back office configurations in financial institutions. The resulting list can be used for testing on a larger scale. Given our sampling strategy of information-rich banks distributed for maximum variation, such a list can be expected to cover a wide range of considerations in the financial services sector. Finally, the findings were verified in another case, replicating the findings in another context (Yin, 2003). This latter case study took place at one of the main competitors of the banks in the cooperative banking group and focused on the same three service delivery systems. It provided much support for our findings, as we were able to confirm the patterns reported in this paper. The next section presents the primary data from the five main case studies, i.e. the designs of the front office – back office configurations and the underlying design considerations in banks A, B, C, D and E. Results In each of the five banks, the service delivery systems for providing mass consumer products, mortgages and business loans were studied. We examined what they looked like and why. Here, the case study results are presented in three parts. First, the flowcharts that were developed for each service delivery system are presented. Second, we present a table containing the design considerations of the banks. Third, the front office – back office configurations are described in more detail, starting from the contents of the flowcharts and the table with the design considerations. In order to understand the data, some background on the three service delivery systems is required. Therefore, a brief description of the main steps involved in delivering mass consumer products, mortgages or business loans is provided. Providing mass consumer products incorporates many financial services, such as a savings account, credit card or electronic banking agreement. The service delivery system for these products involves: . a customer intake to schedule an appointment; . preparing for meeting the customer; . a sales meeting to identify the customer’s needs and potential for additional sales; . registration of the new products in the sales support system; . arranging consent for providing credit facilities to the customer; . printing the sales contract to be signed; and . archiving the contract.

Structuring front office and back office work 117

IJOPM 27,1

118

Providing a mortgage also entails a number of process steps, including: . a customer intake to schedule an appointment; . preparing for the meeting; . a scheduled meeting with the customer; . calculating a mortgage proposal; . arranging consent for the mortgage; . turning the proposal into an offer; . preparing the mortgage deeds; . transferring the funds; and . finally archiving the file. Although mortgages can be seen as customised products in the eyes of the customer, they basically are choices from a pre-defined set of options. Structures and rates have been set in advance and require little adaptation. Business loans are much more tailored to, and designed for, each individual customer than mortgages or savings accounts. They often require several meetings with the customer, the development of a tailor-made financial solution and careful assessments with regard to rates and risks. In general, the service delivery system for this product consists of: . a customer intake to schedule a meeting; . preparing for the meeting; . the actual meeting; . making a financial appraisal of the client’s position; . developing a structure for the loan; . arranging consent for the loan; . preparing an offer; . preparing the additional deeds required; . transferring the funds; and . archiving the file. To examine the design decisions in the banks, flowcharts for each of the three service delivery systems have been developed. The flowcharts display the activities involved in a service delivery system and the design decisions regarding customer contact, decoupling and grouping. They are shown in Figures 1 –3. When read from top to bottom, the first row in a figure shows the steps or activities in a particular process. The second set of rows illustrates the customer contact decision, i.e. which activities are FO and which ones are BO. Similarly, the third set of rows shows the decoupling decisions: how each bank has allocated the activities to employees. This reveals which employees are involved in the process and where work is handed over. The most common functions are advisor, commercial support (CS) and administrative support (AS). In the process for providing mortgages, several banks make use of a shared service centre (central centre). In the process for providing business loans, some banks employed assistants to assist the business advisors. Finally, the bottom part of a figure

BO

Bank A,D,E

CS

Bank B

Comm. support

FO FO

BO

Figure 1. The design decisions for providing mass consumer products

Advisor

Bank B

CS

Archive

Comm. support

CS

Central centre

AS

Comm. support

Central centre

AS

FO

BO

FO

BO Advisor

Advisor

Bank C

Advisor CS

Admin. support

Advisor Advisor

Comm. support

CS

Central centre

AS

CS

Central centre

AS

Bank B

AS

Bank C

AS

Assistant

Ass.

Adv.

Advisor Assistant

Deeds

Assistant

Adv.

Admin. support

Admin. support

Adv

Admin. support

Advisor Assistant

Figure 2. The design decisions for providing mortgages

BO

Adv.

Advisor

Bank D Bank E

FO

Offer

BO

Assistant

Advice

FO

Bank A

Transfer

Grouping: advisors and commercial support employees form market group, administrative support employees and the central centre are functional groups Consent

Bank E

Explanation

Grouping: all employees in different functional groups

Appraisal

Bank B,C,D

Preparation

Grouping: support employees form market group, advisors and central centre are functional groups

Intake

Bank A

Archive

Bank A

Transfer

BO

Deeds

FO

Offer

Bank C,D,E

Consent

BO

Proposal

Preparation

FO

Advice

Intake

Grouping: all employees in different functional groups

Bank A,B

Bank A…E

119

Advisor

Advisor

Bank A…E

Bank E

BO

BO

FO

Bank C

Bank D

Archive

FO

Contract

Bank C

Consent

BO

Data entry

Preparation

FO

Advice

Intake Bank A,B,D,E

Structuring front office and back office work

Ass.

Clerk Adv.

Assistant

Admin. support Admin. support

Bank A & E

Grouping: advisors and assistants form sales teams, support employees are different functional group

Bank B

Grouping: advisors, assistants and support employees form sales teams

Bank C & D

Grouping: all employees in different functional groups

Figure 3. The design decisions for providing business loans

IJOPM 27,1

120

shows the grouping decisions in each bank, allocating the employees involved in the process to either functional or market groups. For example, for the process for providing mass consumer products (Figure 1) bank A designed mainly FO. The entire process, except the customer intake, is carried out by a consumer advisor. The CS employees and the consumer advisors are grouped in different functional groups. In addition to the actual design decisions, described in the flowcharts, we were interested in the banks’ design considerations that underlie these decisions. We asked why they had made a particular design decision and developed matrices for each design decision for each process in each bank. Table IV presents a summary of these matrices. The first column shows the design choices, such as a decoupled process or a functional group, and lists the reasons the banks put forward for making a particular choice, such as cross-selling for FO. The other columns show which reasons the banks specifically mentioned for providing mass consumer products (indicated by C), mortgages (indicated by M) and business loans (indicated by B). For example, Table IV shows that with regard to the service delivery systems for providing mass consumer products, bank A designed FO to promote cross-selling and enable instant service delivery. In addition, the bank designed a coupled process to avoid handovers of work and grouped the employees in functional groups to realise economies of scale. Design considerations FO activities Cross-selling Information exchange Instant service delivery BO activities Efficiency Risk control through checks Coupled processes Avoiding handovers of work Avoiding idle time Broad tasks Concentrated customer knowledge Decoupled processes Enable centralisation Freeing sales capacity Matching workers and tasks Separation of duties Specialisation effects Market groups Workflow coordination Functional groups Continuity of delivery Cross-fertilisation Economies of scale Uniformity

Table IV. Main design considerations in cases

Bank A C C

Bank B C

M M

B

M

B

C

Bank C C

M M

B

M

B

Bank D C

M

B

C M

B

C M

B

M M

B B

Bank E

M

B

M

B

M

B

C M

B

C C

C

C

C

C

B

C

B

M

B

B B

M

M

M B

B

M M

M M M

B B

B B B

B

C

M

B B

B

M

M C

C

M

M M M

B

M M

M

B

M

B

M C

M

B B

C

B

Notes: The design considerations are presented in alphabetical order; C, mentioned for the process for providing mass consumer products; M, mentioned for the process for providing mortgages; B, mentioned for the process for providing business loans

The front office – back office configurations and the design considerations for each of the three service delivery systems are explained in more detail below. Providing mass consumer products From Figure 1, it becomes clear that there are not many differences between the processes for providing mass consumer products in the five banks. Despite the differences in terms of size, market penetration and regional area, the banks roughly made the same design decisions. They also showed similar reasoning (except bank C), see Table IV. In banks A, B, D and E, the process for providing mass consumer products mainly contained FO to enable cross-selling of additional products. Administrative activities were integrated in the sales part of the process to eliminate follow-up work, which increased the speed and efficiency of the process. This was enabled by the information system in use. The processes were largely coupled to avoid handing over work. In fact, decoupling this process was not evident, because most activities took place during the sales meetings with customers. Yet, bank B decided to allocate the preparation of sales meetings to CS employees, in order for the consumer advisors to have more time for meeting with customers, their core sales activities. The banks referred to this as “freeing sales capacity”. With regard to the grouping decisions, the banks grouped the support employees and consumer advisors in functional groups to realise economies of scale. Compared to the other banks, bank C made some different choices. This bank defined a few BO which the other banks chose for FO, to better control the risks involved in providing credit facilities. In addition, the consumer advisors carried out every activity in the process themselves to avoid handovers and create broad tasks. Providing mortgages From Figure 2, it becomes clear that several similarities can be identified in the processes for providing mortgages, although they are not as large as in the processes for providing mass consumer products. The processes mainly consisted of BO, to increase the efficiency of the process. The banks defined a few FO to exchange information with the customers (Table IV). Yet, banks A and B also defined the calculation of mortgage proposals as FO to eliminate follow-up work, while the other banks did not. For the decoupling decisions, two general strategies can be identified. On the one hand, banks A, B, D and E decided to decouple the process into advisory activities, CS activities and AS activities. These banks also subcontracted part of the process to the Mortgage Support Centre of the banking group (“central centre” in Figure 2). The main reason for the decoupling was to free sales capacity, but also to match worker skills with tasks, benefit from specialisation effects and comply with regulations regarding the separation of duties. Bank C, on the other hand, chose a different strategy. It designed a coupled process in which the mortgage advisors carried out most of the activities themselves to avoid handovers and to deliver quality service, following from the fact the mortgage advisors would have all information regarding their customers and their mortgages in their own heads. Additionally, bank C reasoned it was more efficient to fill the idle time of mortgage advisors with administrative activities than to set up a central administrative unit. Finally, with regard to the grouping decisions it becomes clear that most banks applied functional grouping of the employees that were consecutively involved in the process of providing

Structuring front office and back office work 121

IJOPM 27,1

122

mortgages to realise economies of scale, improve uniformity in working procedures and ensure continuity of service delivery. The banks reasoned that in functional groups it is easier for employees to cover for each other or take over work in case of absence than in market groups. This contributes to the reliability of service delivery. Providing business loans Figure 3 shows that the customer contact decisions in the processes for providing business loans did not differ for the five banks. They made exactly the same choices. Most activities were BO, because the work involved in delivering business loans is either too complex for front office execution or would take too long. However, there was quite some variation between the decoupling decisions in the five banks. Yet, this variation can be reduced to two general strategies (Table IV). On the one hand, banks A, B and E designed a decoupled process to free sales capacity. Complementary reasons were matching worker skills and tasks, benefiting from specialisation effects and establishing separation of duties. In this strategy, business advisors worked closely together with assistants that carried out quite a few activities in the process of providing business loans, so that the business advisors could concentrate on visiting existing customers, approaching new customers and working out complex cases. Bank B also assigned AS employees to the couples of business advisors and assistants to relieve the assistants. To facilitate the handovers of work that result from these decoupled processes, the banks established sales teams (market grouping). As business loans are highly customised products, a great deal of coordination is required. In fact, without the sales teams, a decoupled process would probably not have been feasible. Banks C and D, on the other hand, chose a coupled process. Bank C generally preferred coupled processes to avoid handovers of work between consecutive employees in a process and reasoned that with the low-levels of growth and complexity in its market and the number of business advisors available, coupled processes would be more efficient than assistants. Yet, the business advisors were freed from paperwork to prevent over-qualification. Bank D designed a coupled process for a different reason. It argued that a coupled process in which the business advisors carried out most of the work themselves would contribute to the quality of the services delivered, because the advisors have all relevant customer knowledge and deal with all customer-related matters. Still, advisors were freed from the paperwork, such as making offers and deeds. With regard to the grouping decisions, banks C and D chose functional grouping of employees to facilitate cross-fertilisation among business advisors and continuity of service delivery and scale economies regarding the support employees. Analysis In the first part of the data analysis the front office – back office configurations in the cases are investigated to look for empirical support for the three design decisions proposed in this paper. More specifically, we examine whether they can be recognised as separate design decisions. Secondly, the design considerations put forward by the banks are compared with the design considerations derived from literature in Table I. This leads to a refined set of trade-offs that is empirically grounded. Third, we investigate how the design decisions may contribute to particular performance objectives to see whether dichotomous perspectives still hold. Each of these outcomes is addressed below.

Three design decisions instead of clear front office and back office parts At the outset of this paper we argued that depicting service organisations to consist of two separate front office and back office parts might be an over-simplification of what is happening in the real world. Instead, we proposed that three separate design decisions (the customer contact, decoupling and grouping decisions) would better capture the way in which service organisations deal with contact and non-contact activities in service delivery systems. Reviewing the case material, it can quickly be concluded that none of the five banks that was studied operated under a structure that strictly separated BO from FO, allocating the tasks to different employees and grouping them in different departments. Their design choices were more refined than that and showed variation is possible. A more detailed analysis is provided below. The cases show that the banks pay close attention to the occurrence of customer contact in a service delivery system. They do so, because it affects their efficiency and sales potential. Therefore, it can be classified as an actual design decision. A significant example of the attention for the impact of customer contact can be found in the front office – back office configurations for providing mass consumer products in banks A, B, D and E. Here, virtually all activities, including the administrative ones, are performed front office. Another example regards the front office – back office configurations for providing mortgages in banks A and B where mortgage advisors now calculate a mortgage proposal in the customer’s presence. Hence, the distinction between FO and BO is still important. Yet, the banks in this study did not base their division of labour on this factor. Instead, they treated the allocation of tasks as a relatively separate subject. From Figures 1-3, it is easily derived that the distinction between FO and BO does not have to be the basis for decoupling, in the sense that FO are allocated to one employee and BO to another. The banks in this study based their decoupling decisions on different grounds, often leading to a mix of FO and BO combined in one job. For example, the mortgage advisors and CS employees in bank A both conduct FO and BO. The remaining BO are further divided between CS employees, AS employees and the Mortgage Support Centre. This illustrates that the decoupling decision is far more complicated than allocating FO and BO to different front office and back office employees. In our cases, the decoupling was often based on the different skills, knowledge and attitudes required for either commercial or administrative activities. Apparently, in practice commercial and administrative tasks do not overlap neatly with FO and BO. With regard to the grouping of employees, in the current literature on the design of front office – back office configurations, i.e. the customer contact approach by Chase (1978) and the decoupling strategies of Metters and Vargas (2000), the basic premise is that decoupling automatically leads to the establishment of different departments at different locations under different supervisions. We, however, proposed that the grouping decisions would not have to follow naturally from the decoupling decisions. Based on the five case studies we can conclude that this assumption holds. The banks show that, although functional groups were more common than market groups, decoupled processes do not necessarily have to lead to the establishment of separate groups. In the case of the provision of business loans, for example, banks A, B and E chose market grouping of the employees that were involved in the process. Bank E made a similar decision for the grouping of mortgage advisors and CS employees. Hence, the grouping of employees should be treated as a separate design decision.

Structuring front office and back office work 123

IJOPM 27,1

124

Thus, following the 15 front office – back office configurations that were studied, it can be concluded that distinguishing between the proposed design decisions better captures the impact of customer contact on service delivery systems than the front office – back office model. The trade-offs involved The second part of our analysis concentrates on the trade-offs that are associated with the design decisions. The design considerations put forward by the banks (Table IV) are investigated to see whether they are consistent with the trade-offs summarised in Table I. In addition, we look for design considerations that can be added. This leads to a refined set of trade-offs that is empirically grounded. The elaborated trade-offs are presented in Table V. Besides the individual considerations, Table V also shows the performance objectives they contribute to. For example, defining FO to promote cross-selling contributes to the performance objectives of sales. Defining FO to establish instant service delivery contributes to the efficiency and speed of a service delivery system. With regard to the customer contact decision, the banks often mentioned cross-selling as the main motivation for FO, particularly for mass consumer products. Having the customer on-site or on the phone was considered a major opportunity to introduce additional financial services or identify significant sales moments in the future, e.g. moving house or having children. For the more customised services, Design consideration

Table V. Trade-offs in the design decisions

Front office activities Cross-selling Instant service delivery Coupled processes Avoiding handovers Avoiding idle time Broad tasks Concentrated customer knowledge Market groups Workflow coordination Back office activities Control and rationalisation Counterchecks Decoupled processes Counterchecks Enabling centralisation Matching workers and tasks Specialisation effects Functional groups Continuity of service delivery Cross-fertilisation Economies of scale Uniformity

Contributes to SLS E þ SP E þ Q E E þ Q Q E þ Q E RC RC þ Q E E þ Q þ SLS E þ Q þ SLS R E þ Q E Q

Notes: E for efficiency; Q for quality; R for reliability; RC for risk control; SLS for sales; and SP for speed

FO were in place to exchange information with customers. However, this actually was a necessity, not a strategic choice, as the customised nature of the services being delivered calls for an exchange of information between the customer and service provider about the specifications of the service. Therefore, we do not consider exchanging information with customers to be part of the performance trade-off. Instant service delivery can be seen as a new and significant motivation for designing high-contact activities. The banks argued that being able to deliver a service in one stop not only contributes to the speed of delivery, and therefore customer satisfaction, but also to the efficiency of the service delivery system, because follow-up work is avoided. The front office employees can directly attend to new customers, without having to engage in paperwork or handing over work to a centralised back office department. From this perspective, designing FO can actually be more efficient than adding BO. The banks generally reported efficiency motives for BO, as BO were considered to be easier to control and rationalise than FO. In fact, BO were the default design option, with FO only being defined if they could be justified. An additional reason for BO is risk control. Particularly relevant for financial institutions, it was argued that BO allow for additional checks to be carried out by supervisors or co-workers, to prevent credit facilities being given to the wrong customers. To conclude, the trade-off related to the customer contact decision involves the weighing of, on the one hand, cross-selling and instant service delivery and, on the other hand, control and rationalisation and counterchecks. With regard to the decoupling decisions, the banks’ considerations are largely consistent with Table I. In line with Metters and Vargas’ (2000) typology, they decided to couple different kinds of activities in one job to prevent idle time or to deliver quality service, as customer knowledge was concentrated in one place. They also designed coupled processes to create broad tasks, which could increase both the quality and efficiency of service delivery. In addition, some banks emphasised avoiding handovers of work between employees. Avoiding handovers reduces the time that is needed for service delivery and prevents information losses or the creation of noise when information is transferred, improving the quality of the service being delivered. Also consistent with Metters and Vargas’ (2000) typology, the banks decided to decouple their service delivery systems to prepare for centralisation of BO to increase the efficiency of their processes or to benefit from matching workers and tasks, i.e. employing experts. Matching workers and tasks contributes to both the efficiency and quality of service delivery. Furthermore, the banks mentioned specialisation effects and learning curves as the main reason for decoupled processes, also contributing to quality and efficiency. An important new factor in the decoupling decisions is the employment of decoupling to increase sales, by freeing sales capacity. Increasing sales was one of the banks’ major strategic priorities and this clearly influenced the design of their front office – back office configurations, not just for mass consumer products, but also for mortgages and business loans. In order to increase sales, the banks decoupled their processes to ensure that sales experts could be hired for sales jobs and that these jobs would contain nothing but sales. Yet, we do not consider this a separate design consideration in the decoupling trade-off. In fact, it is a combination of matching workers and tasks and specialisation effects. Therefore, we will not include it in the trade-off, but add it as a performance effect to the other considerations (Table V).

Structuring front office and back office work 125

IJOPM 27,1

126

Finally, another reason for decoupling that can be derived from the cases is the separation of duties. This particularly applies to the financial services sector where regulations and legislation are in place to separate the actual transfer of funds from the commercial parts of a service delivery system to prevent conflicts of interest and improper use of assets. Apart from these risk control regulations, decoupling can help prevent errors in general, because one employee can easily check the work of the previous employee. The banks referred to this as built-in counterchecks. To conclude, the trade-off related to the decoupling decision involves the weighing of, on the one hand, avoiding handovers, avoiding idle time, broad tasks and concentrated customer knowledge and, on the other hand, counterchecks, enabling centralisation, matching workers and tasks and specialisation effects. The design considerations regarding the grouping of employees also correspond with the findings from literature. The case material shows that the banks opted for market grouping of employees to improve workflow coordination. This particularly occurred when customised services, such as business loans, were being delivered through a decoupled process. In bank B, the employees in the sales teams even shared the same office to facilitate their interaction. However, functional grouping was basically considered the default design option, with market groups only being defined when a great deal of workflow coordination was required. The main motivations for choosing functional groups were realising scale economies, increasing uniformity in the ways employees carry out their activities and facilitating cross-fertilisation between colleagues. These factors will contribute to the efficiency of the process and the quality of the service being delivered. One addition can be made to the motivations for functional groups. Several banks grouped employees in functional groups to improve the continuity of service delivery. In functional groups employees can more easily take over each other’s work or share a high workload than in market groups. This makes them less vulnerable in case of absence of employees. Achieving high-levels of continuity of service delivery in market groups would require significant cross training of employees. Therefore, functional groups were considered a valuable way of increasing the reliability of a service delivery system. To conclude, the trade-off related to the grouping of employees involves the weighing of, on the one hand, workflow coordination and, on the other hand, continuity of service delivery, cross-fertilisation, economies of scale and uniformity. Performance objectives and coherent front office – back office configurations The third part of the data analysis regards the expected performance effects of the design choices. In the introduction to this paper, we suggested that the front office – back office model of service organisations is largely based on dichotomous thinking: a front office contributes to one set of performance objectives, a back office to another; service organisations focus on either low-costs or high service. We also argued that today’s organisations cannot afford to make such trade-offs. Here, the trade-offs derived from the case studies (Table V) are analysed to see whether a dichotomy still holds. To that end, we investigate the performance objectives that are weighed against each other and the way in which the banks made their design choices. From Table IV it is easy to conclude that the trade-offs do not represent dichotomous situations, where each side contributes to a different set of performance

objectives that are mutually exclusive. Although increasing sales seems to be solely associated with FO and decoupled processes, this is not the case for efficiency and quality. Both FO and BO, coupled and decoupled processes and market and functional groups can contribute to efficient processes. The same applies to the quality of the service being delivered. This can be enhanced through both coupled and decoupled processes and both market and functional groups. Therefore, making a choice on these trade-offs is not a simple case of weighing one performance objective against another and choosing the one that is most desirable. Instead, it involves careful assessment of the different ways in which the design decisions contribute to the performance objectives and determination of which design choices are most appropriate. From the cases it can already be concluded that the banks took the nature of the service being delivered into account. For example, cross-fertilisation and workflow coordination were mainly mentioned for business loans, the most customised service in this study, but not for the standardised services. Furthermore, the banks’ strategic priorities played a role, for example, whether a bank emphasised increasing sales (most banks) or risk control (bank C). Another conclusion that can be drawn from the case studies is that the banks coordinated their design decisions when making choices on the trade-offs. The cases show that, although a front office – back office configuration consists of three relatively independent design decisions, the banks in this study did not consider them in isolation. For example, grouping decisions were used to either emphasise decoupling decisions, like a further separation of different kinds of jobs by installing functional groups, or alleviate their drawbacks, such as when market groups are established to facilitate the handover of work. In the latter case, the grouping decision provides compensation for what is lost in the decoupling decisions. The banks also coordinated the decoupling decisions and the customer contact decision. When only FO were involved, for example, decoupling the process was avoided. Hence, the banks considered the design decisions in concert, as intrinsic parts of a coherent front office – back office configuration. In this way, they tried to create front office – back office configurations that would result in a particular performance, instead of trading off one performance objective for another. Conclusions The main purpose of this paper was to investigate how the distinction between contact and non-contact activities influences the design of service delivery systems and to identify key design decisions for structuring front office and back office work. Building on the current literature, we proposed three design decisions: deciding where customer contact occurs in process, deciding which activities are decoupled from each other and deciding how employees are grouped together, and validated them empirically. We conclude that identifying three separate design decisions provides a better understanding of, and support for, today’s practices than the front office – back office model. We have showed that FO do not necessarily have to be allocated to front office employees that work in front office departments, and vice versa for BO. Instead, different front office – back office configurations are possible. However, this does not mean that customer contact is no longer an important factor influencing design decisions and that the terms “front office” and “back office” should no longer be used. We concluded that the occurrence of customer contact does not have to be the grounds for decoupling a process or grouping employees. Yet, high- and

Structuring front office and back office work 127

IJOPM 27,1

128

low-contact activities still make different demands on the design of facilities, staff and technology in service delivery systems. From a customer’s perspective, there is obviously a clear difference between the parts of the organisation that he or she actually experiences and the parts that remain behind the scenes. Therefore, a great deal of attention should still go into defining and designing the points where the customer and the organisation meet. Elaborating the guidelines proposed by Chase et al. (1984, p. 544), we propose that FO are justified when: . service production and consumption are absolutely inseparable; . marketing benefits are afforded by contact with the customer; . contact with the customer is in principle avoidable but in practice necessary; and . the speed and efficiency of service delivery improve through the elimination of follow-up work. Furthermore, the terms “front office” and “back office” should not be discarded, for they have a strong value in practice, emphasising the distinction between on stage and off stage elements of service delivery. However, it should be realised that the terms can be used to address several different things, such as buildings (“the walk-in offices are our front offices, the support centre is the back office”), people (“I am the front office, because I am on external duty and have sales responsibilities; the people from the support department on internal duty are my back office”), or activities (“I now do most of my work front office, whereas I used to have a lot of back office work”). Whatever they are meant to address, it is not likely to be an organisational form that is based on a strict separation of contact and non-contact activities. In addition to the three design decisions, a set of trade-offs based on the empirical evidence from the case studies has been formulated. We conclude that making choices on these trade-offs involves careful assessment of the ways in which FO and BO, coupled and decoupled processes and market and functional groups contribute to particular performance objectives. We showed they are not black and white choices between two mutually exclusive sets of objectives. This makes them more suitable for today’s competitive arena in which service organisations have to do well on multiple performance criteria. This study is based on evidence from the financial services sector, which limits the generalisability of the findings to other service industries. However, according to Yin’s (2003) replication logic, analytical generalisation is possible, making theoretical (as opposed to statistical) inferences about other industries. In fact, we have no reason to believe that the three design decisions cannot be found outside the financial services sector or that the trade-offs will be very different. It might be the case that for particular organisations, such as hospitals, particular considerations, such as increasing sales, are not very relevant, or that industry-specific elements can be added, such as counterchecks for the financial services sector. Yet, this will not change the basic trade-offs. It is expected that the design decisions and the trade-offs will apply to a broad range of mixed services and providers of mixed services. Nevertheless, further research should ascertain whether they do and what changes might have to be made. Two other areas that are primary candidates for future research are the impact of contingency variables, such as the service being delivered and the organisation’s strategic priorities, on the design decisions, and the actual performance effects of the

trade-offs. Not only should the trade-offs as specified in this study be tested on a larger scale to see whether they are comprehensive, measuring the impact of the particular choices on process performance quantitatively is required to provide support for design guidelines. By proposing and empirically validating three design decisions, this paper has provided the starting points for developing such design guidelines and has contributed to our understanding of the impact of customer contact on service delivery systems. References Ancona, D.G. and Caldwell, D.F. (1992), “Demography and design: predictors of new product team performance”, Organization Science, Vol. 3 No. 3, pp. 321-41. Boyer, K.K. and Lewis, M.W. (2002), “Competitive priorities: investigating the need for trade-offs in operations strategy”, Production and Operations Management, Vol. 11 No. 1, pp. 9-20. Boyer, K.K. and Verma, R. (2000), “Multiple raters in survey-based operations management research: a review and tutorial”, Production and Operations Management, Vol. 9 No. 2, pp. 128-40. Campion, M.A. and Thayer, P.W. (1987), “Job design: approaches, outcomes, and trade-offs”, Organizational Dynamics, Vol. 15 No. 3, pp. 66-79. Chase, R.B. (1978), “Where does the customer fit in a service operation?”, Harvard Business Review, Vol. 56 No. 6, pp. 137-42. Chase, R.B. (1981), “The customer contact approach to services: theoretical bases and practical extensions”, Operations Research, Vol. 29 No. 4, pp. 698-706. Chase, R.B. (1996), “The mall is my factory: reflections of a service junkie”, Production and Operations Management, Vol. 5 No. 4, pp. 298-308. Chase, R.B. and Bowen, D.E. (1989), “Integrating operations and human resource management in the service sector”, in Snow, C.C. (Ed.), Strategy, Organization Design, and Human Resource Management, JAI Press Inc., Greenwich, CT, pp. 273-95. Chase, R.B. and Hayes, R.H. (1991), “Beefing up operations in service firms”, Sloan Management Review, Vol. 33 No. 1, pp. 15-26. Chase, R.B. and Tansik, D.A. (1983), “The customer contact model for organization design”, Management Science, Vol. 29 No. 9, pp. 1037-50. Chase, R.B., Northcraft, G.B. and Wolf, G. (1984), “Designing high-contact service systems: application to branches of a savings and loan”, Decision Sciences, Vol. 15 No. 4, pp. 542-56. Ferdows, K. and De Meyer, A. (1990), “Lasting improvements in manufacturing performance: in search of a new theory”, Journal of Operations Management, Vol. 9 No. 2, pp. 168-84. Grove, S.J. and Fisk, R.P. (1992), “The service experience as theater”, Advances in Consumer Research, Vol. 19, pp. 455-61. Hill, A.V., Collier, D.A., Froehle, C.M., Goodale, J.C., Metters, R.D. and Verma, R. (2002), “Research opportunities in service system design”, Journal of Operations Management, Vol. 20 No. 2, pp. 189-202. Huete, L.M. and Roth, A.V. (1988), “The industrialisation and span of retail banks’ delivery systems”, International Journal of Operations & Production Management, Vol. 8 No. 3, pp. 46-66. Jackson, S.E., Joshi, A. and Erhardt, N.L. (2003), “Recent research on team and organizational diversity: SWOT analysis and implications”, Journal of Management, Vol. 29 No. 6, pp. 801-30.

Structuring front office and back office work 129

IJOPM 27,1

130

Johnston, R. (1999), “Service operations management: return to roots”, International Journal of Operations & Production Management, Vol. 19 No. 2, pp. 104-24. Johnston, R. (2005), “Service operations management: from the roots up”, International Journal of Operations & Production Management, Vol. 25 No. 12, pp. 1298-308. Johnston, R. and Clark, G. (2001), Service Operations Management, Pearson Education Limited, Harlow. Kellogg, D.L. and Nie, W. (1995), “A framework for strategic service management”, Journal of Operations Management, Vol. 13 No. 4, pp. 323-37. Larsson, R. and Bowen, D.E. (1989), “Organization and customer: managing design and coordination of services”, Academy of Management Review, Vol. 14 No. 2, pp. 213-33. Leonard-Barton, D. (1990), “A dual methodology for case studies: synergistic use of a longitudinal single site with replicated multiple sites”, Organization Science, Vol. 1 No. 3, pp. 248-66. McDermott, C.M., Greis, N.P. and Fischer, W.A. (1997), “The diminishing utility of the product/process matrix. A study of the US power tool industry”, International Journal of Operations & Production Management, Vol. 17 No. 1, pp. 65-84. Menor, L.J., Roth, A.V. and Mason, C.H. (2001), “Agility in retail banking: a numerical taxonomy of strategic service groups”, Manufacturing & Service Operations Management, Vol. 3 No. 4, pp. 273-92. Meredith, J. (1998), “Building operations management theory through case and field research”, Journal of Operations Management, Vol. 16 No. 4, pp. 441-54. Mersha, T. (1990), “Enhancing the customer contact model”, Journal of Operations Management, Vol. 9 No. 3, pp. 391-405. Metters, R.D. and Vargas, V. (2000), “A typology of de-coupling strategies in mixed services”, Journal of Operations Management, Vol. 18 No. 6, pp. 663-82. Miles, M.B. and Huberman, A.M. (1994), Qualitative Data Analysis: An Expanded Sourcebook, Sage, Thousand Oaks, CA. Mintzberg, H. (1979), The Structuring of Organizations, Prentice-Hall, Englewood Cliffs, NJ. Nie, W. and Kellogg, D.L. (1999), “How professors of operations management view service operations?”, Production and Operations Management, Vol. 8 No. 3, pp. 339-55. Parasuraman, A., Zeithaml, V.A. and Berry, L.L. (1985), “A conceptual model of service quality and its implications for future research”, Journal of Marketing, Vol. 49 No. 4, pp. 41-50. Roth, A.V. and Van der Velde, M. (1991), “Operations as marketing: a competitive service strategy”, Journal of Operations Management, Vol. 10 No. 3, pp. 303-28. Safizadeh, M.H., Field, J.M. and Ritzman, L.P. (2003), “An empirical analysis of financial services processes with a front-office or back-office orientation”, Journal of Operations Management, Vol. 21 No. 5, pp. 557-76. Sasser, W.E., Olsen, R.P. and Wyckoff, D.D. (1978), Management of Service Operations, Allyn and Bacon, Inc., Boston, MA. Schmenner, R.W. (1986), “How can service businesses survive and prosper?”, Sloan Management Review, Vol. 27 No. 3, pp. 21-32. Shostack, G.L. (1982), “How to design a service”, European Journal of Marketing, Vol. 16 No. 1, pp. 49-63. Shostack, G.L. (1984), “Designing services that deliver”, Harvard Business Review, Vol. 62 No. 1, pp. 133-9. Skinner, W. (1974), “The focused factory”, Harvard Business Review, Vol. 52 No. 3, pp. 113-21.

Skinner, W. (1992), “Missing the links in manufacturing strategy”, in Voss, C. (Ed.), Manufacturing Strategy. Process and Content, Chapman & Hall, London, pp. 13-25. Stake, R.E. (1994), “Case studies”, in Denzin, N.K. and Lincoln, Y.S. (Eds), Handbook of Qualitative Research, Sage, Thousand Oaks, CA, pp. 236-47. Thompson, J.D. (1967), Organizations in Action. Social Science Bases of Administrative Theory, McGraw-Hill, New York, NY. Voss, C.A. (2003), “Rethinking paradigms of service. Service in a virtual environment”, International Journal of Operations & Production Management, Vol. 23 No. 1, pp. 88-104. Yin, R.K. (2003), Case Study Research: Design and Methods, Sage, Thousand Oaks, CA. Corresponding author Leonieke G. Zomerdijk can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

Structuring front office and back office work 131

IJOPM 27,1

132

International Journal of Operations & Production Management Vol. 27 No. 1, 2007 pp. 132-133 q Emerald Group Publishing Limited 0144-3577

Note from the publisher Emerald at 40 This year Emerald Group Publishing Limited celebrates its 40th Anniversary. As anyone with more than two score years under their belt (or approaching it) will know, 40 represents a milestone. It marks a point at which we have, or we are supposed to have, come to terms with the world and reached a good understanding of what we want from life. And for those of a more contemplative persuasion, it prompts us to reflect upon our earlier years and how we got to where we are. It is perhaps not so different for a company. In many ways, to reach the age of 40 for an organisation is quite an achievement. Emerald’s own history began in 1967 with the acquisition of one journal, Management Decision. The company was begun as a part-time enterprise by a group of senior management academics from Bradford Management Centre. The decision to found the company, known as MCB University Press until 2001 was made due to a general dissatisfaction with the opportunities to publish in management and the limited international publishing distribution outlets at this time. Through the creation and development of the journals, not only was this particular goal achieved, but the foundations of a successful business were laid. By 1970 the first full-time employee was appointed and by 1975 there were five members of staff on the pay roll. In 1981, there were 20 members of staff and three years later the company had grown to a size which meant we had to move to larger premises – one half of the current site at 62 Toller Lane. Through the 1990s Emerald came of age. In 1990, the first marketing database was introduced and several years later we acquired a number of engineering journals to add to our increasing portfolio of management titles. The IT revolution also began to impact upon the publishing and content delivery processes during this period. Writing in 2007, it seems hard to remember a time when information was not available at the click of a button and articles were not written and supplied in electronic format – and yet it was only 11 years ago that Emerald launched the online digital collection of articles as a database. The move was seen as pioneering and helped to shape the future of the company thereafter. The name of the database was Emerald (the Electronic Management Research Library Database) and in 2001, we adopted this name for the company. So, how does Emerald look in 2007? Emerald has grown into an important journal publisher on the world stage. The company now publishes over 150 journals and we have more than 160 members of staff. Emerald has always stressed the importance of internationality and relevance to practice in its publishing philosophy. These two principles remain the cornerstones of our editorial objective. The link between the organisation and academe that was so crucial in the foundation of the company continues to influence corporate thinking; we uphold the principle of theory into practice. Emerald also continues to carry the tag of an innovative company. Through our professionalism and focus on building strong networks with our various communities, we have launched and developed initiatives such as the Literati Network for our authors and a dedicated website for managers. These innovations and many more, help

to set us apart from other publishers. For this reason, we feel confident in stating that we are the world’s leading publisher of management journals and databases. It is important to us that we continue to strengthen the links with our readers and authors and to encourage research that is relevant across the globe. In more recent history we have, for example, awarded research grants in Africa, China and India. We also opened offices in China and India in 2006, adding to our existing offices in Australia, Malaysia, Japan and the USA. We would like to thank the editors, editorial advisory board members, authors, advisers, colleagues and contacts who, for the past 40 years, have contributed to the success of Emerald. We look forward to working with you for many years to come. Rebecca Marsh Director of Editorial and Production

Note from the publisher

133