Competitive Advantage and the Value Network Configuration

Long Range Planning 39 (2006) 109e131

www.lrpjournal.com

Competitive Advantage and the Value Network Configuration Making Decisions at a Swedish Life Insurance Company Øystein D. Fjeldstad and Christian H. M. Ketels

Executives in industries that facilitate transactions within a network of customers, such as financial services and telecommunications, currently face tough challenges. The factors determining success in their industries are changing rapidly and the boundaries of many of their markets are realigning. General analytical tools exist to support decision-making on these issues, but many of these tools have been devised primarily with manufacturing industries in mind. These tools may therefore require modifications in order to accommodate the underlying value creation logic of transaction services, the so called network-industries. We present a case study of a project in a Scandinavian life insurance company where the value network, an alternate value configuration analysis tool to the established value chain, was used to represent the company’s activities and identify and evaluate the strategic options facing the company. The value network tool, which closely matched the executives’ view of their industry and firm, proved useful in making a significant decision for the company. In particular, value network analysis channelled attention to the composition of the customer set and the mechanisms affecting the composition as being at the heart of the company’s competitive position. Transaction service firms link their customers for a wide variety of purposes, such as risk sharing, financial transactions or communication. The composition and the size of the customer set are therefore important drivers of the value of service to individual customers of such firms. The experience of this case study merits further investigation of the use of different types of activity configurations depending on a given industry’s underlying value creation logic. It provides insights into the types of situations in which value configuration analysis is most helpful, and gives guidance on how to identify issues for which the value network is the more powerful tool to support decision-making. Ó 2006 Published by Elsevier Ltd

0024-6301/$ - see front matter Ó 2006 Published by Elsevier Ltd. doi:10.1016/j.lrp.2006.05.001

Introduction Since its introduction by Michael Porter in his book ‘Competitive Advantage’ (1985), the value chain has become the most widely used tool for representing and analyzing how companies create value.1 The choices that a firm makes concerning its activity configuration, together with the resources that these activities leverage, are central elements of a company’s business model.2 The value chain describes the classic manufacturing firm that creates value for its customers by transforming inputs into products where the price the customer is willing to pay (and thus their value received) is higher than the per unit cost of performing the activities. Porter distinguishes between primary activities, i.e. the physical creation of a product as well as its sale, transfer to the customer, and after-sale assistance, and support activities, i.e. the provision of purchased inputs, technology, and human resources to support primary activities as well as other firm-wide functions. But while the value chain has been applied very successfully in manufacturing industries, researchers and practitioners have questioned its appropriateness for understanding firms with different underlying value creation logics.3

The value network models transaction services where firms act as intermediaries creating value by providing services that support exchanges within a network . of people, organizations or locations. This article discusses the use of the value network configuration, an alternative to the value chain introduced by Fjeldstad and Stabell in response to such concerns about the applicability of the value chain to transaction service structures. The value network configuration has previously been used in the contexts of the telecommunications services industry, e-business, business schools, and in this issue supply management.4 The value chain and the value network representations capture different ways in which firm’s activities interact to create value, depending on the underlying value creation logic and economics of enterprises and their associated industries. The value network models transaction services (such as insurance and telecommunications) where firms create value by providing services that support exchange within a network of nodes, which can be people, organizations or physical locations. Insurance companies allow their customers to share risks, banks intermediate payment, liquidity and risks between individuals as well as companies, telecommunication firms facilitate communication between individuals, and transportation firms move physical objects between locations. These firms are not the networks themselves; they are the intermediaries providing the networking services. The introduction of multiple value configurations, such as the value chain and the value network, reflect the view that firms differ systematically in the way that activities relate to each other according to their underlying value creation logic, that these differences can be captured by different value configurations, and that choice of value creation has implications for the development of business strategy. We investigate these claims in a case study of how the senior executives of a Swedish group insurance company and their consultants used the value network configuration to analyze the company’s situation and develop strategic options for its future.5 Their case enables us to shed light on the following two research questions: -

-

When are value configurations relevant and effective analytical tools in companies’ decision making processes? How does the choice of the value network instead of the value chain as a conceptual tool affect the analysis?

With respect to the first question, two main observations stand out. First, value configurations are powerful tools for analyzing companies’ strategic positioning, although they are of relatively 110

Competitive Advantage and the Value Network Configuration

less importance when studying operational efficiency. Operational efficiency relates to the cost of performing particular activities and can be analyzed using a broad set of benchmarking tools where the organization of activities in value configurations is of little consequence. Strategic positioning depends on how the whole system of activities, i.e. the value configuration, contributes to customer value and to the firm’s cost of providing that value to the chosen market segment.6 Porter highlighted this link between value configuration and competitive advantages, but many subsequent applications have overlooked it. Second, the actual effectiveness of an analytical tool depends on whether or not it is considered valid by the decision makers and corresponds to their intuitive view of the business. There is little systematic knowledge about how consultants and managers actually use the strategy textbook tools,7 and even less about how they are adapted to settings different from those in which the tools have been devised. We address these issues of value configuration validity and relevance in the specific context of our case study.

the value network focuses attention on the size and composition of the customer set . . [and on] identifying, attracting and retaining customers whose membership has a positive value for other clients The second question, the consequences of using the value network instead of the value chain configuration as an analytical tool, reveals two additional main observations, both of which emphasize that companies which use a value configuration that is truly in line with the underlying characteristics of their industry benefit because they can set strategies based on a more appropriate understanding of their value creation dynamics. First, the value network highlights a set of issues that are different from those highlighted by the value chain. In particular, it focuses attention on the properties of the customer set (i.e. its size and composition), as a central element of the choice as to whether to be a member.8 This increases the importance of activities concerned with identifying, attracting and retaining customers whose membership has a positive value for other existing or potential clients. Trying to use a value chain-activity-system to model a company that creates value through a value-network-logic will cause a high likelihood of missing or misjudging the importance of this key element of its value creation process. Second, using the value network rather than the value chain as an analysis tool will also frame the issues raised in a different manner, and thus lead to different answers. The best example is the value of a customer to the company. In a value chain analysis the customer value is given simply by the discounted cash flow of future revenues from sales to the customer, minus direct costs. In a value network, the customer value includes the additional worth of the impact that the customer has on the attractiveness of the customer pool to other existing or future customers. The remainder of the article is organized in three parts: First, we discuss value network analysis, contrasting it to value chain analysis. Second, we present the case study, first describing the actual analysis conducted by the company’s executives and consultants, and then adding an ex-post analysis of the consequences as far as generating strategic options for the company that might have been expected had the alternative value chain configuration been applied. Third, we discuss the key insights for company executives and researchers triggered by this case.

Value configuration analysis Comparing value networks to value chains This section introduces the value network in more detail and contrasts its properties to those of the value chain. The value network models the value creation activities of what Thompson labelled Long Range Planning, vol 39

2006

111

a mediating technology,9 which links customers who are or wish to be interdependent. This form of value creation is characteristic of transaction service organizations10 and is reflected in the categorization and organization of primary and secondary activities. According to Fjeldstad and Stabell there are three primary activity categories in the value network: (1) Network promotion and contract management - activities associated with attracting and selecting potential customers for the network. The activities include direct measures, such as traditional sales activities, as well as indirect measures such as marketing and branding activities. These activities provide no direct service to the customers, but invite them to join the network or the risk pool or to modify the terms of their relationship with it. (2) Service provisioning - activities associated with establishing, maintaining and terminating links between customers, and billing for value received. These activities constitute the core contact between existing members of the network and the service provider. Through this point of contact, members access the linking services offered by the network provider, and can also be linked through interconnection to customers of other providers. (3) Infrastructure operation activities associated with maintaining and operating the physical, financial and information infrastructure which allows the company to provide services to its customer base (the network infrastructure of telecommunication operators is the obvious example). The activities are represented as layered rather than sequentially related, consistent with exchange activity models in the literature on communication and organizational behaviour.11 Differences between the two value configurations lead to differences in the associated sources of operational efficiency and competitive advantage. In the value chain, the cost-related interactions across activities associated with the product flow along the entire chain are important, while in the value network, alignment of the size of the network infrastructure with the size of the customer pool and customers’ use of services is key to network yield. Factors impacting the per-unit cost of performing individual activities (e.g., cost economies of scale and learning, location and capacity utilization), however, remain important sources of operational efficiency in both modes of analysis.

In a Value Chain, value creation derives from products, and competitive advantage from how well products match customer needs. The Value Network creates value by enabling exchanges: competitive advantage accrues when the network matches members’ needs. In a Value Chain, value creation derives from products, and the extent to which the products match customer needs defines the source of competitive advantage. The Value Network, by contrast, creates value by enabling exchanges, and the competitive advantage accrues according to the extent to which the network within which such exchanges are enabled matches the needs of its members.12 The two relevant properties are network size and network composition. In value networks, the character of the pool (i.e. who the other members are) can be as important an influence on the benefits of being a member as the characteristics of the network provider’s services. Such network effects are typically associated with Internet firms, but researchers have shown that the size and composition of the customer set affect the value of most types of intermediation activity including voluntary organizations and institutions.13 The effects of the network are termed global when it is the network’s size alone that affects the value of membership, and local when it is the actual identity of the other members that matters.14 Global and local network effects are also relevant for value chains. For example, global network effects can affect value chains when the products are complements (such as game consoles and their respective game software). Local network effects, however, play a particularly strong role when firms add value by linking customers, because customers typically value linking to other particular 112

Competitive Advantage and the Value Network Configuration

customers. Two sided markets where customers belong to two distinct groups and the members of each group are interested only in the composition of the other group present a special case of local network effects. This is the case in credit card service where a credit card provider must ensure that its card is accepted by the stores from which its customers wish to buy.15

The challenge is to build a ‘club of members’ sufficiently large [for] operational efficiency, in which the members sufficiently complement each other to make the customer network itself a key part of the value proposition. Different types of network effects have a strong impact on pricing and other choices that companies make to attract and screen the right type of customers.16 The challenge for any value network is to build a ‘club of members’ that is sufficiently large to yield economies of scale and thus operational efficiency, and in which the members sufficiently complement each other to make the existing customer pool, or network, itself a key part of the value proposition, in addition to the nature and quality of the services offered to customers. The marketing activities of a transaction service provider, including pricing decisions, must ensure both that the network has enough customers and that the customers are good matches for each other. Rothschild and Stiglitz point out how, in insurance, pricing can be an important customer screening mechanism for separating low risk customers from high risk customers. Table 1 summarizes the differences between value chains and value networks with respect to the sources of operational efficiency and competitive advantage as outlined above.

Value configuration analysis in practice Developing strategies for Fo¨renade Liv Founded in 1948, Fo¨renade Liv was one of the smaller players in the Swedish market, providing and managing group life and health insurance products. As a mutual society owned by its customers, Fo¨renade Liv was created to provide insurance for low-wage employees with limited access to comparable insurance products. At the time of the analysis, the company had no ready access to capital required to fund a major expansion and its mission tied it closely to the market it was operating in. Table 1. Competitive advantage versus operational efficiency

Value Chain Competitive advantage

Operational efficiency

Long Range Planning, vol 39

Value Network

 Activities that enable exchange within a network with scale and composition that match members’ needs.  Inter-relationships with complement exchange layers  Interconnection with other value networks  Capacity utilization  Yield management  Cost economies of scale and experience

 Activities that can meet differentiated demand  Linkages with upstream and downstream value chains  Horizontal linkages across segments

2006

113

From 1998 onwards, Fo¨renade Liv provided insurance products through two business units: Fo¨renade Liv Gruppfo¨rsa¨kring AB (‘FLGAB’) offered insurance products through groups, especially large companies and trade unions. The participation by individual company employees or union members in these group schemes was voluntary. Fo¨renade Liv Kollektivavtalsfo¨rsa¨kring AB (‘FLKAB’) offered similar products through the same channels, but the groups were signed up as part of collective agreements without individual group members having the choice to opt out. By the end of 2000, Fo¨renade Liv had financial worries originating at FLGAB, where the income from premiums and investments had begun to fall substantially below insurance pay-outs and the reserves necessary to cover future claims by the insured (see Figure 1). The company’s financial position had suffered from the downward adjustment on equity markets after the bust of the tech boom early in 2000, which had reduced capital returns on the assets held by all insurance companies. However, the data showed that Fo¨renade Liv’s financial position had deteriorated much faster and more dramatically than the overall industry, and thus had further, more company-specific, issues to deal with. The Swedish life and health insurance market that Fo¨renade Liv operated in four distinct layers. At the bottom was the large, government-administrated insurance system that drew on legally determined contributions by the employees. In the second layer (where FLKAB operated) collective insurance was provided by agreements between the unions and the employer organizations (or large companies), with the individual beneficiaries having no direct say in their insurance coverage. The third layer (where FLGAB operated) was group insurance where large organizations, traditionally labour unions, negotiated package deals offering standardized basic protection as a unilateral offer to all individual members of the group, who were free to decide whether they wanted to accept or not. Such group insurance could be provided at low cost, as it did not require the costly assessment of individual risk profiles or the management of complex pricing schemes. However, it also ran the risk of attracting especially ‘bad risks’, such as people with unhealthy lifestyles, because of the lack of controls and price differentiation. To manage this risk, group insurance was limited to basic protection which did not cover the full costs that such ‘bad risks’ might face. In the Swedish market, such group insurance accounted for 35% of all life insurance premiums in 1999. Finally, at the top layer, insurance policies were sold to individuals directly through individual insurance policies of the traditional (fixed amount, dominated by insurance companies) or unit-linked (fixed

Cash Flow minus Required Reserves for Future Client Demands

100%

50%

Forenade Liv FL (w/o FLKAB) Industry average

0%

Jul-01

Sep-01

May-01

Jan-01

Mar-01

Nov-00

Jul-00

Sep-00

May-00

Jan-00

Mar-00

Nov-99

Jul-99

Sep-99

May-99

Jan-99

Mar-99

Nov-98

Jul-98

Sep-98

-100%

Mar-98

-50%

May-98

Normalized (Mar 98 = 100)

150%

Source: Swedish Insurance Federation

Figure 1. Free capital from insurance premium revenue 114

Competitive Advantage and the Value Network Configuration

investment, dominated by banks) type. Fo¨renade Liv’s life insurance products would not be affected ex-post by whether or not the insured had additional policies on other layers. However, the availability of other contracts did affected potential customers’ demand for group insurance products. The nature of the Swedish life and health insurance market reflected two basic features of insurance: life insurance contracts typically limit coverage by only covering particular types of loss, or only specified parts of an insured’s loss. Such practices can be explained by the fact that some insurers restrict insurance to areas where asymmetric information is less of a problem,17 or where quantity restrictions could be used to avoid attracting high risk customers, i.e. avoid adverse selection.18 As a result, many insurance markets have a ‘layered’ structure in which contracts offered in one layer provide added risk coverage but specify that they do not extend to risks covered by other (lower layer) insurance contracts. By 1999 Fo¨renade Liv understood clearly that its core market - where FLGAB was competing was undergoing significant change. First, banks had entered the group insurance business, exploiting their existing branch networks and customer relationships. Second, technological change was increasing the number of access channels customers expected to be able to use, including internet and 24/7 phone services. Third, the increased public debate on the viability of the public pension systems had changed public perception of the importance of life insurance, turning it from a lowattention issue into an important consumer product, changing the context in which life insurance marketing messages were being received. Fo¨renade Liv needed to understand what its strategic options for being able to continue operating in this environment were, whilst staying true to its mission as a mutual society. Analyzing Fo¨renade Liv’s strategic options using different value configuration frameworks In 1999, faced with increasing threats to their business caused by recent underlying changes in their markets, Fo¨renade Liv hired a consulting company to support its board and senior management in developing strategic options in line with the company’s mission. The consultants’ work was organized around a number of workshops (see Appendix), in which documents prepared by the consultants were reviewed by the senior leadership of the company. The consultants were asked to provide recommendations on methods and analysis, leaving the company management to judge whether the findings were sufficient for the board to take a decision, and then make the final decision on which approach to take. The following chapter documents the actual analysis and methodological approach; it does not add to or change the analysis ex-post. However, the authors also offer a sub-section which develops a hypothetical value chain analysis of the insurance company to illustrate the differences to the approach actually taken. The overall analytical review process started with an analysis of the market and the overall industry, the key elements of which are outlined above. The industry analysis was followed by an analysis of activities and capabilities to understand company strengths and weaknesses, which is discussed below. (See Appendix for further details of consultation procedures.)

the value network approach . immediately resonated with the leadership team as being a good match for the company’s business as the organization and management of a risk pool Value network analysis in practice At the initial stage of its strategic review, the consultants proposed modelling the company’s activities using the traditional value chain approach, in which they had considerable experience. However, it soon became apparent that this model presented problems in terms of recognition, and the Long Range Planning, vol 39

2006

115

value network approach was proposed as an alternative. This concept immediately resonated with the leadership team as being a good match for the company’s business as the organization and management of a risk pool to provide insurance coverage to multiple clients, perhaps because of the company’s nature as a mutual society.19 The first task was to represent the activities of the group insurance business in the three different groups suggested by the value network framework (see Figure 2, taken from the project material). The team adapted Fjeldstad and Stabell’s value network representation where the activities performed are represented as layered rather than sequential. Infrastructure operations, in this cases, includes investing the capital accumulated from insurance premiums, assessing risks and setting prices, and developing and packaging new insurance products. Service provision here includes activities related to customer service both for individual clients and for the groups that they are part of, handling premium payments, and regulating claims. Finally, marketing and contract management includes activities related to persuading new groups to join, or to achieving higher penetration within groups already served. If the group already existed for another reason, such as membership in a trade union, this activity could gain significantly in effectiveness. The second task was to identify the key source of the company’s worsening relative economic performance, by asking the following questions: What is the relative effectiveness of the company within specific activity categories, and what is the relative value that the sum of these activities provides to customers? The first question tested whether Fo¨renade Liv had fallen behind in terms of efficiency in any of its individual activities. The short answer is that this did not seem to be the case. Table 2 shows that its overall costs relative to premium income had traditionally been in line with or below those of competitors. Staff numbers were low and the company had been very economical in its spending on marketing and sales (see Table 3). Its actuarial skills were commensurate with those in the industry, and relations between its sales staff and their counterparts in the groups, whether enterprises or trade unions, were perceived as good. While any line-item comparisons between the data for Fo¨renade Liv (which only handled group insurance) and those for its competitors (for whom group insurance only represented a minority share of overall business) were problematic, interviews within

Value Network primary activities adapted in insurance case Marketing Selling to groups

Selling to individuals in groups

Build brand

Service Premium administration

Customer service groups

Customer service individulas

Payments to insured

Infrastructure Productdevelopment

Productpackaging

Risk analysis/ pricing

Asset management

Figure 2. Group insurance Value Network representation 116

Competitive Advantage and the Value Network Configuration

Table 2. Cost position of Swedish Life Insurance companies Admin costs as % of premium income, Companies with more than 2% market share, 1998.

Company

1st Qtr 1998

KPA Liv AB SEB Trygg Liv gamla SHB Liv FLGAB Folksam Liv ¨G FLO Sparliv Skandia Liv Landia Wasa Liv Livia SEB Trygg Liv SalusAnsvar Grpliv SPP Livfo¨rs. AB

3.41% 3.98% 7.48% 7.88% 10.11% 11.89% 13.61% 14.48% 15.31% 15.56% 15.81% 25.96% 43.09% 67.39%

the industry confirmed that the recent decline in performance was not due to any inefficiency in its individual activities. Overall, it seemed that Fo¨renade Liv had a relatively strong position in servicerelated activities, and that it was seen as roughly on par with its peers in terms of underlying infrastructure, but as relatively weak in marketing. This profile of strengths and weaknesses seemed natural given the company’s roots. As a mutual society owned by its clients it focused on services, kept investments in administrative infrastructure at the standard industry level due to its limited access to capital, and, lacking a profit motive, did little to aggressively expand its business. If Fo¨renade Liv’s problem resulted from weak performance in one individual activity, marketing and sales would have been the prime candidate. But the low expenditures in this area seemed perfectly in line with the company’s overall positioning as a low cost provider of base insurance. The low cost had also predated the recent financial difficulties. Between 1998 and 2000, however, the company’s situation had clearly deteriorated. Fo¨renade Liv had lost premium income while administrative costs had risen, partly because of the company’s investment in a major new IT system to handle the many relatively small payments that characterize the group insurance business. There was also a gut feeling that a greater share of customers had begun to claim insurance benefits, which increased administrative costs as well as benefit costs. To the analysis team, the problem seemed to be related to the systemic interaction of activities in the value creation process, an issue where the value configuration chosen offered a significantly better framework for understanding the nature of the interaction between activities and network structure. Using the value network, the analysis was able to focus on understanding how Fo¨renade Liv’s activities affected the network it organized; i.e. the risk composition of their customer portfolio, and what had changed in that relationship over time. Table 3. Fo¨renade Liv’s costs over time

Marketing Administration

1997

1998

1999

2000

3.4% 6.2%

2.4% 14.2%

0.1% 17.3%

0.0% 15.1%

Share of insurance premium income. Long Range Planning, vol 39

2006

117

Fo¨renade Liv’s business model as a stand-alone group insurance company operating in the traditional market had relied on its ability to offer standardized insurance products at unified rates that were attractive to ‘good’ risks, i.e. young, healthy employees, as well as ‘bad’ risks, i.e. old, more illness-prone employees (Figure 3). Standardization as well as the sales approach of group insurance allowed them set lower prices than what would have been possible for individually sold insurance policies. Unified rates led to an implicit redistribution of insurance costs from ‘high’ to ‘low’ risk customers (as shown in the shaded area), but the overall cost advantage of the group insurance format was large enough that these policies were still attractive to the ‘low’ risks despite the implicit subsidy they were paying to the ‘high’ risks. The entry of the banks during the mid 1990s fundamentally changed the dynamics of the insurance market by breaking up the broad customer pool into sub-groups by risk level. Banks formed groups around their existing customer base, predominantly young ‘low risk’ customers taking out loans to finance real estate purchases. They passed through the cost advantages inherent to the group insurance product and could offer a price based on the low risk profile of this smaller customer group. Fo¨renade Liv couldn’t match this offer without financial loss as it had to budget for the implicit subsidy to ‘high’ risk customers. The banks’ ability to attract and select such clients increasingly changed the composition of the risk pool for traditional group insurance companies targeting work-related groups, such as Fo¨renade Liv. With fewer ‘good’ risks to subsidize the ‘bad’ risks, the pressure was on to increase premiums to reflect the growing average risk in the client pool. This process threatened to unravel the previous equilibrium situation: as risk levels were rising within traditional group insurance, its lower distribution costs were no longer sufficient to compensate for the implicit subsidy of an increasing share of ‘bad’ risks. As good risks left the pool, either to sign up for individual policies (which, moreover, could be distributed cheaply via new direct channels such as the internet), or to opt for new group offers to non-traditional groups (such as young house buyers), premiums for the remaining clients were forced upwards, creating incentives for the next group of good risks to seek better alternatives. This vicious circle was likely to continue until only the bad risks remained in the pool and would have resulted in a premium level so high as to price these clients out of the market. Figure 4 illustrates the changing dynamics of this unravelling market. The upper dotted line reflects the actuarially fair pricing of individual low-risk contracts and the lower dotted line the actuarially fair pricing of such contracts when provided to young house owners as a group. The lower dashed line reflects the pricing of group insurance to the

Cost of insurance to insured

The Traditional Market

Redistribution

Actuarially fair pricing of risk for individual

Cost advantage of group sales channel Actuarially fair pricing of risk for group

Risk = Age = Cost of insurance to insurer

Figure 3. Group insurance before the entry of banks 118

Competitive Advantage and the Value Network Configuration

The Unravelling Market

Cost of insurance to insured

Actuarially fair pricing of risk for individual Actuarially fair pricing of risk for remaining group Deterioration of average risk profile in remaining group Cost advantage of sales channel to new group

Actuarially fair pricing of risk for old group

Actuarially fair pricing of risk for new group

Risk = Age = Cost of insurance to insurer

Figure 4. Group insurance after the entry of banks e market unravelling

original group (Figure 3) and the upper dashed line the pricing after the good risks have been enticed away from the pool. The problem faced by Fo¨renade Liv corresponded closely to an insight noted by M. Rothschild and Nobel Prize winner J. Stiglitz: that a pooled equilibrium is impossible when there are cost-effective ways, including self-selection, of separating the high and low risk customers (see Appendix 2 for further detail). The analysis delivered a disturbing message to Fo¨renade Liv’s leadership team. A change in their competitive environment had fundamentally eroded the competitive advantage they had enjoyed in organizing risk pools for customers with heterogeneous risk profiles, which allowed all customers to benefit from the low distribution cost model of a group insurance sold through employmentrelated groups. Before moving to the next task, developing strategic options for the company, we will discuss what the assessment of the firm’s position might have been had it been represented from the perspective of a value chain. While in the actual case the team was quick to choose the value network instead as the appropriate analytical tool to map their company’s activities, it is instructive to undertake this ‘thought experiment’ to investigate the consequence of different value configurations in representing company activities. Analyzing group insurance as a value chain e ‘the road not taken’ Stone, Foss and Machtynger provide a representative example of an insurance company as a value chain.20 Their representation includes four sequential steps: customer acquisition, underwriting, customer relationship management and claims management and service. Although this is a straightforward list of activities conducted by firms in the insurance industry, it is not a straightforward application of the value chain model. First, the activities listed do not match the primary activity categories of the value chain. And second, the listed sequence does not describe the production of an increasingly valuable product or service from the acquisition of components through production to post purchase-service. The only real connection between the model and its use is a sequential listing of the activities by which an insurance company creates value. The first part of the analysis, i.e. finding out whether the per unit cost of producing and distributing group insurance products was too high, turns out to be independent of the type of value configuration framework applied. The second part, however, i.e. understanding how activities interact to provide sufficient value for customers, reveals important differences. A value chain analysis would most likely have come to the conclusion that unsatisfactory product quality was at the heart of the company’s problems, as indicated by the low rate of prospective new customers (in the last Long Range Planning, vol 39

2006

119

year before the strategic review, the net stock of groups served by the company had even experienced some shrinking), and the low penetration rate achieved within the groups (with fewer than 50% of eligible members in most groups deciding to sign up for insurance in most groups).

from a value chain perspective, the problem with trying to identify which activities affect product value is that important quality attributes are associated with who the other customers are A typical value chain analysis of the origins of product quality would assess how the various sequential steps add value to a product as it passes through the value chain. Michael Porter refers to ‘policy choices’ about product features and performance offered and services provided as the most important drivers of uniqueness. But, from a value chain perspective, the problem with trying to identify which activities along the insurance chain affect the product value is that important quality attributes are associated with who the other actual and prospective customers are. In the words of Rotschild and Stiglitz: ‘By their very being, high-risk individuals cause an externality: the low-risk individuals are worse off than they would be in the absence of the high-risk individuals’. The main drivers of value for customers in insurance are therefore the composition and scale of the insurance pool into which the customer is invited. Using a literal traditional value chain analysis and looking at the activities that impact product features as the product passes through the value chain, the company would have been likely to conclude that the value chain activities behind creating their products were not at the root of their problems. Arguably they could invest more in marketing and sales but it was not obvious whether that would improve the company’s financial situation or just increase the absolute scale of the problem by adding more customers with low (or negative) value. The problem was in the market: for parts of the potential client base, the company’s offer seemed to be getting less attractive. Since there had been no substantial change in the products in the recent past, something else must have happened to erode the offer’s attractiveness in this market. But the value chain analysis provided no guidance as to how to identify the source of this market disruption. In contrast, the value network perspective naturally put this disruption at the heart of the analysis.

. . the value chain analysis failed to identify the source of the [insurance] market disruption, [while] the value network put this disruption at the heart of the analysis. Developing strategic options for Fo¨renade Liv The ultimate value of the value network (as well as the value chain) approach depends on its ability to analyze a given situation in a way which provides useful guidance in developing actionable alternatives. To the team, the results of the value network analysis described above suggested a number of directions that the company could pursue: -

120

First, focus on service provisioning, as an outsourcing solution for other companies including insurance firms, dropping infrastructure operation and marketing Competitive Advantage and the Value Network Configuration

-

-

Second, drop only marketing activities, the weakest activity category in their value network configuration, and become a supplier of insurance products to other companies. Third, change the range of services while keeping the existing value configuration largely intact, in order to become attractive to a client pool that could be economically sustained.

The team evaluated all of the options that this approach generated according to their economic feasibility and their compatibility with the company’s mission. The first option, i.e. becoming a provider of administrative services, turned out to have some market potential. However, a segment that consisted mainly of administrating large volumes of relatively small payments for public or nearpublic institutions was viewed as an unattractive option. It depended on political decision-making and legislative changes to create market structure and pricing, and if the market were to be opened, other outsourcing specialists might decide to enter it. There was a considerable likelihood that international outsourcing specialists would jump at the opportunity. Furthermore, this activity profile did not match the company’s mission. The second option, i.e. becoming a provider of insurance products to other insurance companies by focusing on infrastructure and service activities, was also seen as having the potential to meet some future demand. Insurance companies and other financial services providers with strong brands and marketing/sales organizations were looking for additional insurance products to broaden their own portfolio. Attractiveness, however, was again an issue: lacking direct access to end consumers and their needs, Fo¨renade Liv would become increasingly dependent on its sales partners and loose negotiating power. While the company could ensure that its current clients would continue to be served, its influence on the nature of the offer would diminish over time. The third option, i.e. remaining a provider of insurance products with the full set of activities in infrastructure operation, service provision and marketing, was seen as the only option left, although it clearly required significant changes within the company. One sub-option e changing the definition of the groups targeted in order to gain back a foothold among ‘good’ risks e was ruled out as being in conflict with the company’s mission. A second sub-option could see the company extending its portfolio to sell other, more profitable products to its existing base. In principle, this was seen as both economically feasible and in line with the company’s mission, but it would require significant investments in new systems, new products, and especially marketing and sales activities. With its mutual society ownership structure the company lacked the capital to finance such investments, and decided instead to look at partnering models to provide the capabilities and assets that Fo¨renade Liv did not have the capital to build up on its own. In 2001 the search for a partner led to Folksam AB, a large insurance company with a broad product portfolio serving the Swedish market, purchasing Fo¨renade Liv Gruppfo¨rsa¨kring AB and fully integrating its activities into its own operations. This option represents a further development of the last option e finding a partner and merging the customer pool, in this case with a larger insurance firm that must have perceived synergies between its own activities and Forenade Liv’s customer set. We were not privy to the assessments made by the acquiring company, but as a general comment value network synergies can arise because a customer set may have a poor composition or insufficient scale for one type of exchange, but a good composition and scale for other types of exchange: in insurance terms, a particular group may be low-risk with respect to one type of insurance service and high-risk with respect to another.

Discussion This article contributes to improving our understanding of how companies’ activity configurations can be represented in ways that support effective decision making by company executives. We organize the discussion of our findings along two dimensions: their relevance to either research or practice, and their relationship to the two research questions raised in the introduction: when are value configurations relevant and effective analytical tools in companies’ decision making, Long Range Planning, vol 39

2006

121

and how does the choice of the value network instead of the value chain as a conceptual tool affect the analysis? Figure 5 summarizes the insights from the case study. Insights for researchers Two implications from this case study relate to our understanding of value configurations in general, while a third implication relates to the consequences of using the value network instead of the value chain as an analytical tool:  Designing intuitively valid activity system representations. For an activity model to be effective in use it must support representations of the firms that the executives involved see as valid, otherwise they are likely to reject the tool as irrelevant and ignore it. The insurance executives were sceptical about value chain models of their firm because it did not match their view of the insurance business. However, the value network immediately resonated with them, even though it was presented as a newer and more experimental tool. They quickly developed sufficient confidence in the tool to use it in a critical decision process. (One can speculate that the success of the value chain, in many other situations, was similarly supported by its intuitive appeal to many executives in manufacturing companies.) Many of the existing analytical tools have been developed in the context of industrial firms rather than service firms, and other researchers have noted that tools and languages must be adapted in accordance with the technological, economic, and organizational properties of firms, recognizing that particular models may be local and specialized rather than broad and generalizable.21  Using value configurations as building blocks for activity systems. Porter uses the term activity system to describe interrelated choices about activities, resources, location, technology, etc. that explain how a company creates competitive advantages in serving its target market.22 Our findings suggest a possible avenue for further advances in the understanding of strategic positioning. Understanding and considering different value configurations could become the starting point for gaining a more systematic understanding of which of these choices are critical and how they interact for different classes of companies. Value configurations could define systematic categories as the starting point for describing a company’s activity system instead of developing such a system entirely from scratch for each individual business.

Implications Practice

Research

• Designing intuitively valid Effective use of value configurations

• Analyzing competitive advantage,

activity system representations.

not operational efficiency

• Using value configurations as building blocks for activity systems

• Evaluating value configurations as abstract analytical tools. Impact of using value network vs. value chain

• Gaining competitive advantage from customer set composition

• Considering effects on the attractiveness of the network when assessing customer value

• Emphasizing activities and choices that affect the customer set

• Identifying potential markets with value configuration analysis

• Disrupting markets through configurational change

Figure 5. Issues and Findings 122

Competitive Advantage and the Value Network Configuration

 Linking tools to theories. We have made the argument that in certain situations the value network configuration may prove a more appropriate analytical tool than the value chain configuration. How do we determine the appropriateness of a particular tool for a given context? In our view, the choice of one model over another depends on two equally important factors: (1) whether it gives better answers to the strategic questions faced by a company and (2) whether it offers a higher likelihood that issues which are critical to the company in question receive attention. While the first condition is obvious, the second reflects the fact that such analytical tools also frame the questions asked. Thus making the appropriate choice will result in better answers to better questions.

Using the appropriate analytical tool will result in better answers to better questions. Note that answering these ‘tests’ is different from stating whether or not a company is a value network or a value chain in an absolute sense. Value configurations are abstract constructs that describe how a company’s activities create value for customers. Determining the configuration of a company is both an empirical issue and a strategic choice. (We return to the latter in the discussion of insights for practitioners.) This issue reflects a deeper underlying relationship between theories and tools. As a result of social construction processes where concepts travel back and forth between academics and practitioners, tools carry strong assumptions about the theories and contexts that they reflect.23 Over time, the tools’ underlying assumptions and their original application contexts may become separated. Therefore researchers need to assess whether, and to what extent, the assumptions behind a tool, concept or logic warrant use outside the context within which they have been derived.24 The two above tests respectively reflect framing of questions raised and the analytics involved in deriving conclusions. Specifically, our case study illustrates that in a world of diverse forms of value creation we must separate a theory of activity-based advantages from the value chain manufacturing metaphor. New tools, such as the value network, provide alternative schemas for framing the understanding of the firm and its environment, and may therefore play a role in transformational renewal.25 Insights for practitioners Turning to practitioners, the first implication relates to the effective use of value configurations in general. The subsequent implications relate to the specific choice of the value network as a representational framework.

Porter’s perspective that activity configurations are analytical tools to understand the [creation of] competitive advantage, rather than operational efficiency, has too often been lost Analyzing competitive advantage, not operational efficiency The experience from this case supports Porter’s argument that activity configurations are analytical tools used to understand the interplay of a company’s activities in creating competitive advantages, not the operational efficiency of individual activities. This is a perspective that has too often been lost in subsequent applications. In our case study, the company’s performance in individual Long Range Planning, vol 39

2006

123

activities was quite comparable to the performance of its peers and clearly not the source of its economic problems. Instead, the issue at stake was how changes in the fit between these activities, the customer set and the market environment eroded the company’s established competitive advantage. Gaining competitive advantage from customer set composition In the case, the decision to use the value network as an analytical tool directed the attention of the executives and their consultants to problems that the company faced in its customer set. The value chain framework is unlikely to have been as powerful in unlocking the key problem that the company faced, which was a critical worsening, not in the performance or combination of its internal activities, but the composition of its client portfolio, brought about by the declining attractiveness of its offer in the face of a shifting market. More generally, there are systematic differences in the key dimensions of strategic positioning and key sources of competitive advantage differentiating the value configurations that we believe executives need to consider: -

-

Value chain: strategic positioning is defining the specific set of customer needs targeted by a product. Competitive advantage is achieved by the appropriate integration of activities that create products with properties that satisfy the targeted customer needs. Value network; strategic positioning is defining the specific needs for exchange of a specific set of customers. Competitive advantage is achieved by organizing activities that attract and retain customers that mutually contribute to the attractiveness of the network.

As an example, consider the well documented case of Direct Line, a UK direct insurance company that was very successful in car insurance but failed to succeed in the home insurance market.26 In a value chain interpretation, the direct sales channel which the company introduced to the insurance business could be seen as an innovation that dramatically reduced the cost of sales and in this way created value for its customers. This was consistent with the company’s success in the car insurance business but failed to shed light on their failure (or at least lack of comparable success) in the home insurance business. When we take a value network perspective, the differences in these markets become more apparent: in their car insurance business, Direct Line could attract a customer portfolio of good risks by using a pricing scheme that would offer low prices to experienced drivers only. This target group is easily identifiable, it views car insurance as a commodity and price is a critical decision-making factor for them. The marketing approach used attracted a pool of customers with the desired characteristics. The market for home insurance is quite different. It is harder to organize pools of good risks e screening is more complicated and housing insurance is often obtained in the context of other financial relationships, e.g. when taking out loans to finance housing. Home insurance customers also tend to stick with insurers. Neither the relationships with the experienced drivers nor Direct Line’s lower sales costs provided enough value to entice enough good risks to join the new risk pooling network that Direct line was trying to run. Considering effects on the attractiveness of the network when assessing customer value Because of its focus on the customer set, the value network logic also leads to a different assessment of the value of an individual customer. In a value chain environment, the value can be measured by the discounted cash flow of revenues from this customer minus the per-unit cost of producing the product or service. In a value network environment, in addition to the net cash flow contribution, the customer’s value also includes the impact of their membership on the overall attractiveness of the customer base to other customers. In the case, Fo¨renade Liv had set a uniform price for all customers even though one group (young workers) had a higher customer value due to their positive contribution to the overall attractiveness of the customer pool. This worked as long as there was a sufficient cost advantage for this group from sharing the activities with the entire customer pool and there were no competing offers targeted at the lower cost group. Fo¨renade Liv’s positioning choice had always raised the question of whether there should be differential pricing for different network participants. In other network industries, like telecommunication, designing 124

Competitive Advantage and the Value Network Configuration

differential pricing to attract and retain those customers who have the highest positive impact on the attractiveness of the network to others is an important competitive tool.

In network industries differential pricing is an important competitive tool [to gather] customers who have the highest positive impact on the attractiveness of the network to others. Emphasizing activities and choices that affect the composition of the customer set A strategic focus on the customer set versus individual customers’ needs has important implications for the choices that a company identifies as core. In the case studied, the banks entering the market focused on a specific customer set and were successful by offering insurance or banking products that increased the attractiveness of this customer set to existing and new target customers. Companies like Fo¨renade Liv that instead defined their market position through a particular form of insurance had trouble competing. In other network-based industries, financial services in general or telecommunication networks, the same logic will apply. In a value chain perspective, product or service properties are at the centre, while the customer set is of interest mainly in terms of customers’ willingness-to-pay for different product properties and total volume of demand captured. In contrast, the value network perspective puts the customer set at the centre, and then draws attention to individual products and services as instruments to strengthen and extend this network, and obtaining profitability from organizing it. The case illustrates the importance of the ability to screen customers according to the impact of their risk profile on the profile and attractiveness of the overall customer pool. Banks, Fo¨renade Liv’s new competitors, had, by way of their existing customer relationships, access to a different (and in the event superior) customer screening technology. Identifying potential markets with value configuration analysis Many transaction service industries are in the midst of fundamental change processes. While government regulations have traditionally defined the boundaries of their markets, legal and technological changes are now driving market realignments. Value configurations are a useful tool to structure the analysis of markets that might emerge, and using the value network as the appropriate representation of the underlying technology can improve the framing and the analysis. Potential markets can be associated with complete value configurations: creating and making available products (value chain) or enabling inter-customer linkages (value network), or with individual activities that are ‘outsourced’ elements of other companies’ value configurations. In the case, Fo¨renade Liv considered potential markets at both levels. The company’s project team analyzed the option of remaining a complete value network business; this alternative required investments, which the company was unable to make. In addition, the project team analyzed two options that relegated the company to being an outsourcing service provider for other value networkbusiness; both alternatives were discarded for not being in line with the company’s mission. There are similar issues in other value network-industries. In telecommunication, for example, network providers need to consider their relationship to service providers that perform only selected portions of the value network activity-set. Provisioning of media services represents another and richer example. A network operator can: (1) provide media services by buying content and thus treating the media companies as suppliers along a value chain, (2) sell services to the media companies and treat them as customers of their value network, or (3) leverage media companies as complementing partners who stimulate demand for the network service, i.e. treat companies, such as googleÔ, as complement network layers. There are significant strategic implications related to detailed activities, standards, ownership and pricing dependent on which framing is chosen. Long Range Planning, vol 39

2006

125

[The arrival] of a fundamentally different underlying technology can be profoundly disruptive, introducing fresh value creation logics and new forms of competitive behavior. Disrupting markets through configurational change Companies that introduce a fundamentally different underlying technology into an existing market can have a profoundly disruptive effect on incumbents.27 Such changes introduce fresh value creation logics, and thus facilitate new forms of competitive behavior. An interesting example is SkypeÔ, which challenges traditional telecommunication companies by providing voice over internet (VOIP) software and services. The company reconfigured the telephony network value system by separating the voice layer from the network operator layer, arriving at cost economies that traditional telecommunication companies cannot match. The company also differed from early entrants into the VOIP market, who had introduced a valuechain-logic by only providing software which required users to keep track of complicated internet-addresses. SkypeÔ disrupted the market not only by giving away software, but also by combining the software with a value network logic and providing essential value network services, such as maintaining a database of customers’ contact-names, and interconnecting to fixed and mobile phone services. In the Fo¨renade Liv case, the situation was slightly different. The two ‘outsourcing’ alternatives that were developed for Fo¨renade Liv explicitly moved the company away from its basic value network logic, but they also moved it into new markets. Recognizing the change in underlying technology was important for understanding whether the company would be able to develop a competitive advantage in these markets.

Conclusions A single case study, as we discuss in the Appendix, can identify new issues and generate ideas, but it can not give definitive answers. The case of Fo¨renade Liv and the use of the value network configuration in developing strategic options for the company, in our view, motivate both broader research and practical development to make a set of different value configurations available to business executives. The use of the value network instead of the value chain in this case study led to different insights because of one fundamental difference between these two value configurations. Value chains sell something that they produce and own, i.e. a product or a service, while value networks sell something that they organize but don’t technically own, i.e. access to a network of other customers. This leads to very different types of value creation economics. For a value chain, product value is independent of the number of customers, although cost levels might be affected by economies of scale. For a value network, service value depends on the specific and overall number of other customers in the network. This point was made in Fjeldstad and Stabell’s article which initially developed the value network as an analytical tool: Mediation services offered by value networks represent the extreme case [of network externalities] because the dependency among customers is the main product delivered. Stated differently, in value networks, the other customers are the key part of the product. The services of a value network mainly deliver the customers’ opportunities to exercise those dependencies. Size and composition of the customer base are therefore the critical drivers of value in the value network. (p 431) 126

Competitive Advantage and the Value Network Configuration

Viewing insurance companies through this lens, portrays them as network (pool) organizers that reduce risks by combining individual risk profiles rather than as businesses that get paid by their customers for taking risks. The value chain has shown the substantial benefits that activity-oriented tools can provide. We are convinced that alternative value configurations that correspond better to the underlying network logic of the industries that have gained substantial ground in the economy in recent years can have significant value for companies facing strategic challenges in those dramatically changing markets. Such challenges often call for much more than bringing a new product to market or upgrading a specific function or skill. They demand the ability to develop new business models based on a fresh analysis of underlying value creation logic(s). The importance of re-evaluating existing tools in any field undergoing change, be it research, engineering or strategic management, is well captured by the famous psychologist Abraham Maslow: ‘I suppose it is tempting, if the only tool you have is a hammer, to treat everything as if it were a nail’. Executives in such situations face the dual challenge of having tools appropriate for managing their current business and obtaining tool-kits that allow them to take advantage of exciting new opportunities or face up to dire threats from competitors with different value creation logics. Those who can broaden their tool kit, and can understand which tools to use for which situations, have much to gain.

[Executives must] broaden their tool kit, and understand which tools to use for which situations. Acknowledgements We wish to thank Anita McGahan and Jan Rivkin, as well as the journal’s Editor and his academic referees, for helpful comments on an earlier version of this manuscript. The first author thanks the Haas School of Business, University of California Berkeley, where parts of this article were written during his period as a Visiting Scholar.

Appendix The analytical review process undertaken by the executives and the consultants The process adopted for the interaction between project team and the company leadership was not fundamentally different from other comparable situations. Discussions were structured around different workshops on market opportunities, internal capabilities, and decision options. Figure 6 outlines the process which was followed. More importantly, we believe that, the interaction process itself would not have been different with a value chain instead of a value network approach. As the article explains, what was different was the framing that the tools provided and the outcome of the analytical process. A note on applying insurance and information economics to the case There is obviously a strong link between modern information economics and insurance competition. The Swedish market for group life and health insurance provides an interesting example of the mechanisms described in the economics literature on contracts under incomplete information. Below, we give a brief outline of the link. Akerlof’s famous ‘lemons’ paper explicitly mentions the group insurance market (although in a slightly different form than here). Old employees know their risk profile while young employees don’t e this destroys the equilibrium for old employees as healthy old employees will find it beneficial to opt out of the pool. The case documented in this article had a slightly different structure: while the types e old and young employees e were known to everyone, the traditional Long Range Planning, vol 39

2006

127

The Process of the Analytical Review Joint project team

Board working group

Meeting 1

Meeting 2

Meeting 1

• Formulate

• Identify potential

•

challenge for the company

• Analyze

overall market opportunities and threats

• Analyze

future markets

• Short-list

markets with sustainability and possible fit to company’s capabilities

Meeting 2

Meeting 3

Detailed • Detailed • Preparatio evaluation of evaluation of action company of market recommendations attractivenes activities & for the board for the three capabilities relevant for the prioritized target markets three prioritized target markets

Board

Off-Site

• Review of

the analysis

• Discussion of action recommendations

• Decision

company’s internal strengths and weaknesses

TIME Figure 6. Project process

group insurance pooling provided a cost advantage that was so great that the policies were attractive even for the younger employees. But at the national level there was nothing to stop new competitors from entering the market with an offer aimed solely at young employees. This is what happened when the banks exploited their existing customer relationships to target young employees. As a result, the implicit subsidy, which the young employees remaining in the traditional group had to finance, reached a level beyond that which could be compensated by the cost advantage in distribution. Rothschild and Stiglitz have described another situation that is also of relevance in this context. Their model, which comprised insurance contracts of different risk groups self-selected through a menu of price and quantity combinations, pointed out that under specific conditions there might not be a stable equilibrium. Again the case documented in this article is structured differently: group insurance was created to extend coverage to those who could not afford it at actuarially ‘fair’ terms. Hence the Rothschild and Stiglitz result, where bad risks get fair terms and good risks are slightly underinsured, would fail to address the objectives of the initiators of group insurance. Fo¨renade Liv explicitly wanted to create groups that included good and bad risks at equal terms: but, as Rothschild and Stiglitz showed, such a pooled equilibrium cannot be sustained when the good risks can be separated from the bad risks. Research method We used a single case study to describe and analyze how a team of company executives and their external consultants applied core strategic analysis tools to analyze a strategic situation and develop and evaluate alternative options for the company’s future. One of the authors was a member of the consulting team. We believe that therein lays the strength of our approach because it allowed us rich access into a context of managerial decision-making typically closed to outsiders. On the negative side, such an approach makes it harder to obtain the detached objectivity of an outside researcher. We tried to overcome the potential bias problems by the second researcher, who was not part of the team, interviewing the team member where appropriate. The original project documentation, to which we had complete access, has been used as a data source. The case method, and in particular a single case, obviously has its limitations. In the case analyzed, the team of executives and their consultants had made explicit choices about models or lenses by which to interpret a business context, a firm and its industry.28 We can only speculate about the consequences of imposing alternative lenses ex-post, but that discussion is based on the 128

Competitive Advantage and the Value Network Configuration

participating author recounting the team’s motives for its choice of lens, including a brief description of why the dominant lens was rejected. It is worth pointing out that if the participating author had a bias, it was in favour of the rejected tool. The Value Chain is part of the standard toolbox of his consulting firm. Our analysis discusses what the particular lens chosen reveals about the case context and the strategic choices derived.29 Our data consist of the reports and presentations made by the consulting team to the client, transcripts of interviews with the participating author, transcripts from interviews with managers in the course of the consulting project and follow-up interviews, and a selection of salient articles on the economics of insurance.

References 1. M. E. Porter, Competitive Advantage, The Free Press, New York (1985); For an examination of the Value Chain in use see for example A. Hinterhuber, Value Chain Orchestration in Action and the Case of the Global Agrochemical Industry, Long Range Planning 35(6), 615e636 (2002). 2. C. Markides and C. D. Charitou, Competing with dual business models: A contingency approach, Academy of Management Executive 18(3) (2004). 3. C. B. Stabell and Ø. D. Fjeldstad, Configuring value for competitive advantage: On chains, shops, and networks, Strategic Management Journal 19(5), 413e437 (1998); J. G. Harris and R. J. Burgman, Chains, Shops and Networks: The Logic of Organizational Value, Accenture Institute for High Performance Business, Chicago 1e9 (2005); C. K. Prahalad, The Blinders of Dominant Logic, Long Range Planning 37, 171e179 (2004); P. Molineux, Creating customer value, Spectra e The Journal of the Management Consultancies Association (Summer), 19e23 (2002); OECD, Information Technology Outlook, (2004). 4. E. Andersen and Ø. D. Fjeldstad, Understanding inter-firm relations in mediation industries with special reference to the Nordic Mobile Communication Industry, Industrial Marketing Management 32, 397e408 (2003); OECD (op. cit. at Ref 3); A. Afuah and C. L. Tucci, Internet Business Models and Strategies, McGraw-Hill, Irwin (2000); A. Ordanini, The Effects of Participation on B2B Exchanges: A Resource Based View, California Management Review 47(2), 97e114 (2005); P. Lorange, New Visions for Management Education, Pergamon (2002); L. Huemer, Supply Management: Value Creation, Coordination and Positioning in Supply Relationships, Long Range Planning 39(2), doi:10.1016/j.lrp.2006.04.005 (2006). 5. The case is based on the second author’s experience working as a consultant for Monitor. He was part of the consulting team and proposed the use of the value network model. At that time the other author was not aware of the project. 6. M. E. Porter, What is strategy?, Harvard Business Review 74(6) (1985). 7. R. Whittington, Strategy as Practice, Long Range Planning 29(5), 731e735 (1996). 8. The importance of membership composition qualities is eloquently expressed in the famous quote by comedian Groucho Marx: ‘I sent the club a wire stating, please accept my resignation. I don’t want to belong to any club that will accept me as a member.’ 9. J. D. Thompson, Organizations in Action, McGraw-Hill, New York (1967). 10. D. C. North, Institutions, Journal of Economic Perspectives 5(1), 97e112 (1991). 11. C. P. Alderferer, An intergroup perspective on group dynamics, in J. Lorsch (ed.), Handbook of Organizational Behavior, Prentice-Hall, Englewood Cliffs, NJ 190e222 (1987); A. Tannenbaum, Computer Networks, Prentice Hall, Englewood Cliffs, NJ (1981). 12. N. Economides, Network Economics with Application to Finance, Financial Markets, Institutions & Instruments 2(5), 89e97 (1993); M. L. Katz and C. Shapiro, Network Externalities, Competition and Compatibility, American Economic Review 75(3), 424e441 (1985); M. Rothschild and J. Stiglitz, Equilibrium in Competitive Insurance Markets: An Essay on the Economics of Imperfect Information, Quarterly Journal of Economics 90(4), 629e649 (1976). 13. N. Economides, The Economics of Networks, International Journal of Industrial Organization 14(6), 673e699 (1996); N. Economides (op. cit. at Ref 11); D. Encaoua, M. Moreaux and A. Perrot, Compatibility and competition in airlines Demand side network effects, International Journal of Industrial Organization 14(6), 701e726 (1996); J. Rohlfs, A Theory of Interdependent Demand for a Communication Service, Bell Journal of Economics and Management Science 5(Spring), 16e37 (1974); re Internet firms, see Afuah and Tucci (op. cit. at Ref 4); re voluntary organisations, see J. M. McPherson and L. Smith-Lovin, Women and Weak Ties: Differences by Sex in the Size of Voluntary Organizations, American Journal of Sociology 87(4), 883e904 (1982); re institutions see North (op. cit. at Ref 8). Long Range Planning, vol 39

2006

129

14. J. Farrell and P. Klemperer, Coordination and Lock-In: Competition with Switching Costs and Network Effects, in M. Armstrong and R. H. Porter (eds), Handbook of Industrial Organization, Elsevier, Amsterdam (2003). 15. J.-C. Rochet and J. Tirole, Platform Competition in Two-Sided Markets, Journal of the European Economic Association 1(4), 990e1029 (2003). 16. see Katz and Shapiro, and Rothschild and Stiglitz (op. cit. at Ref 10). Economic effects of customer set size and composition are sometimes referred to as network effects or network externalities. This terminology is not used in the insurance literature, where the main concerns have been with the elicitation of the information necessary to compose risk pools. 17. A. G. Akerlof, The Market for ‘Lemons’: Quality Uncertainty and the Market Mechanism, Quarterly Journal of Economics 84(3), 488e500 (1974). 18. Rothschild and Stiglitz (op. cit. at Ref 10). 19. In effect, they recognized that their business was in managing a club: ‘One can think of managing a mediating firm as managing a club. The mediating firm admits members that complement each other, and in some cases exclude those that don’t. The firm establishes, relationships among members. Supplierecustomer relationships may exist between the members of the club, but to the mediating firm they are all customers. Depositors are not bank suppliers, they are just as much bank customers as those borrowing money. By acting as an intermediary, bilateral interactions between the mediator and its customers are used to enable multilateral interactions between customers.’ Stabell and Fjeldstad p. 427 (op. cit. at Ref 3). 20. M. Stone, B. Foss and L. Machtynger, The UK consumer direct insurance industry: A role model for relationship management?, Long Range Planning 30(3), 353e363 (1997). 21. M. Adams, Activity-based costing (ABC) and the life insurance industry, Service Industries Journal 16(4), 511 (1996); R. Ramirez, Value co-production: Intellectual origins and implications for practice and research, Strategic Management Journal 20(1), 49e65 (1999); G. V. Krogh and J. Roos, A tale of the unfinished, Strategic Management Journal 17(9), 729 (1996); B. Løwendahl and Ø. Revang, Challenges to existing strategy theory in a postindustrial society, Strategic Management Journal 19(8), 755 (1996). 22. Porter (op. cit. at Ref 5). 23. A. Giddens, The Consequences of Modernity, Stanford University Press (1990). 24. R. A. Bettis and M. A. Hitt, The new competitive landscape, Strategic Management Journal 16(5), 7 (1995); R. A. Bettis and C. K. Prahalad, The dominant logic: Retrospective and extension, Strategic Management Journal 16(1), 5 (1995); R. Cooper and G. Burrell, Modernism, Postmodernism and Organizational Analysis: An Introduction, Organization Studies 9(1), 91 (1988); S. Fuchs and S. Ward, What is deconstruction, and where and when does it take place? Making facts in science, building cases in law, American Sociological Review 59(4), 481 (1994). 25. H. W. Volberda, C. Baden-Fuller and F. A. J. van den Bosch, Mastering Strategic Renewal: Mobilising Renewal Journeys in Multiunit Firms, Long Range Planning 34, 159e178 (2001). 26. D. Channon, Direct Line Insurance PLC: new approaches to the insurance market, in C. Baden-Fuller and M. Pitt (eds), Strategic Innovation, Routledge, London (1996). 27. C. M. Christensen, The innovator’s dilemma: when new technologies cause great firms to fail, Harvard Business School Press, Boston, MA (1997). 28. G. Morgan, Images of Organization, Sage, Thousand Oaks (1987). 29. G. T. Allison, Essence of decision: Explaining the Cuban Missile Crisis, Little, Brown & Co., Boston (1971).

Biographies Øystein D. Fjeldstad is Telenor Chair of International Strategy and Management, Department of Strategy and Logistics and Dean of Academic Programs at the Norwegian School of Management. He holds a Ph.D. in Business Administration from the University of Arizona. Fjeldstad is a former manager of Andersen Consulting (now Accenture). His main research is on models of value creation and competition in telecommunication, financial services and professional service industries. BI Norwegian School of Management, Nydalsveien 37, 0442 Oslo, Norway tel +47 46410401, e-mail: [email protected] Christian H. M. Ketels is the Principal Associate at the Harvard Business School, Institute for Strategy and Competitiveness. He received his PhD (Economics) from the London School of Economics (LSE) where he developed 130

Competitive Advantage and the Value Network Configuration

microeconomic models analyzing the relationship between international competition and firm efficiency. Ketels worked as a strategy consultant for Monitor Company from 1997 to 2000 in the UK, Germany and Scandinavia, and was part of a team that founded Cell Strategy (now part of Adcore AB), a Swedish-based Internet Strategy Consultancy, in 2000. His research interests focus on the relationship between company strategy and location and on competitiveness at different geographic levels, especially in Europe. Harvard Business School Ludcke House, Soldiers FieldBoston, MA 02163, USA tel (617)384-5935, e-mail: [email protected]

Long Range Planning, vol 39

2006

131