A Contingency View of Porter's Generic Strategies

A Contingency View of Porter's "Generic Strategies" Author(s): Alan I. Murray Source: The Academy of Management Review, Vol. 13, No. 3 (Jul., 1988), pp. 390-400 Published by: Academy of Management Stable URL: http://www.jstor.org/stable/258087 Accessed: 13/02/2009 08:13 Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=aom. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that promotes the discovery and use of these resources. For more information about JSTOR, please contact support@jstor.org. Academy of Management is collaborating with JSTOR to digitize, preserve and extend access to The Academy of Management Review. http://www.jstor.org ? Academy of Management Review, 1988, Vol. 13, No. 3, 390-400. A Contingency View of Porter's "'Generic Strategies"' ALAN I. MURRAY University of Alberta According to Porter, cost leadership and product differentiation can be pursued simultaneously only under rare conditions: It is also unclear how these strategies can be implemented. In this article Porter's generic strategies are linked to external preconditions. This approach shows that the generic strategies are not mutually exclu- sive and that each strategy may be linked to a variety of strategic means. The implications that these results have for structuring organi- zations are discussed. In 1980 (Hall, 1980; Porter, 1980), the notion of generic strategy swept through the business pol- icy area. It was very appealing because it seemed to offer a solid theoretical framework for a discipline that often lacked theoretical founda- tions. As Hambrick (1983a, p. 688) commented, "Porter's typology of generic strategies seems es- pecially useful, because (1) it builds on previous findings and (2) it is appropriately broad, but not vague." White (1986, p. 220) noted that Porter's generic approach to business strategy incorpo- rated "a few critical dimensions yet has strong theoretical underpinnings." As it stands, Porter's generic strategy concept does not satisfy this desire for a solid theoretical framework. In order to be useful, theory should guide empirical research. Yet, the empirical in- vestigations spawned by the generic strategy concept (Dess & Davis, 1982, 1984; Hambrick, 1983a, 1983b; Miller & Friesen, 1986a, 1986b; Phillips, Chang, & Buzzell, 1983; White, 1986) are not comparable, and their results are contra- dictory. Here it is argued that the generic strat- egy concept can be clarified by linking each strat- egy to a set of environmental preconditions. De- veloping these preconditions also allows the key question (e.g., Miller & Friesen, 1986b) of the compatibility of generic strategies to be resolved and facilitates a discussion of the link between generic strategies and the strategic means (Dess & Davis, 1982, 1984) used to implement them. A Contingency View of Generic Strategies Contingency approaches to strategy are not new (e.g., Hofer, 1975); in fact, a contingency perspective is implicit in the adaptive view of strategy that currently dominates the literature (Miles & Snow, 1978). Its clearest manifestation in the strategy area is the structure-strategy- performance paradigm that was introduced by institutional economists (e.g., Bain, 1956; Caves, 1980; Porter, 1981; Rumelt, 1974). Applying a con- tingency approach to the generic strategy litera- ture also is not new. Especially in his discussion of fragmented industries, and what to do about them, Porter (1980) hinted that the efficacy of generic strate- gies may be contingent on industry structure. In his 1985 treatment, this link became explicit: "Cost advantage and differentiation in turn stem from industry structure" (Porter, 1985, p. 11). However, Porter stopped short of explaining this link. 390 Day (1984) also tied the efficacy of generic strategies to environmental factors, specifically to customer perceptions of product offerings. Hambrick (1983a, p. 702) too concluded that a con- tingency view of generic strategies may be appropriate: "It is simply not accurate to say that all generic strategies are equally viable within an industry ... any broadly 'generic' strategy is really a composite of numerous variations, not all of which are equally suited to a given situa- tion." Phillips, Chang, and Buzzell (1983, p. 42) concluded from their study that: "While our analysis has produced generalizable results showing that certain types of generic strategies do lead to success, the exact manner in which these strategies are translated into success var- ies dramatically by type of business." Finally, Miller and Friesen (1986a, p. 39) also advocated a contingency view: "Although some industrial product and capital goods industries may exhibit relatively pure types, we do not feel convinced that this would be true for consumer durables industries." In addition, Miller and Friesen advanced reasons for their expectation that different strategic configurations would emerge in different industries, yet they too stopped short of presenting a theoretical frame- work to explain the predicted differences. This article attempts to provide the theoreti- cally based contingent approach to Porter's ge- neric strategies that the authors cited above advocated. Beginning with the focus strategy, it is hypothesized that the viability of each ge- neric strategy is dependent on the presence of certain external conditions, specifically on indus- try structures or customer characteristics and preferences. Focused Versus Broad Strategies Of Porter's three original generic strategies (product differentiation, cost leadership, and focus) the focus strategy seems to have caused the greatest confusion (e.g., Dess & Davis, 1982, 1984). Porter (1985) attempted to resolve this con- fusion by positing that the choice of a focused or broad strategy is independent of the choice of product differentiation or cost leadership. Thus, according to Porter, a firm could take either a focused or broad approach to either product dif- ferentiation or cost leadership. Yet even today many researchers confuse a focused strategy with a product differentiation strategy. The problem is one of levels of analysis. Whether pursuing cost leadership or product differentiation, the strategist using a focused ap- proach must first differentiate the product offer- ing from offerings aimed at other segments of the same market, hence the confusion between a focused strategy and product differentiation. Southland's 7-11 stores, for example, exemplify product differentiation based on convenience, but this is only when they are compared with food retailers targeting other market segments (e.g., supermarkets). When they are compared with other firms competing in their own niche (i.e., other convenience stores), it becomes clear that 7-11 stores strive for cost leadership. Given that both cost leadership and product differentiation have broad and focused variants, the obvious question is: Which variant should a firm choose? A conjoint analysis framework (Green & Srinivasan, 1978; Shocker & Srinivasan, 1974, 1979; Srinivasan & Shocker, 1973) can help to answer this question. This approach repre- sents a product as a "point" in "attribute space" in which each dimension is a product attribute (including price) valued by customers. By using Linmap (Srinivasan & Shocker, 1973) and sim- ilar statistical techniques, firms can deduce sur- veyed customers' attributed weights that repre- sent the importance of each attribute to each customer. But, conjoint analysis permits not only the derivation of attribute's weights for individual customers but also the derivation of individual customer's "ideal points." An ideal point is that combination of product attributes (including price and the trade-offs thus implied) which the cus- tomer prefers to all other points in the attribute space. Because the attribute space is common for all customers, it is possible to collapse numer- ous customers' ideal points into a single space. This feature is useful because it permits firms to 391 identify if the market can be segmented on the basis of customer needs. If all customers' ideal points fall into roughly the same region, one product configuration should suit all customers. But, if ideal points form two or more clusters, the firm may be able to offer more than one product configuration suc- cessfully. Obviously, if all customers prefer a roughly similar product offering, a focus strat- egy is not viable. Thus, the viability of the focus strategy is tied directly to exogenous factors; in this case it is the heterogeneity of customer preferences. However, even if multiple segments are identi- fied, a focus strategy may not be viable. It may be that sufficient economies of scope exist so that a multisegment producer can outcompete a com- petitor that focuses on a single niche. Only if synergies between segments are low or nega- tive will a focus strategy be viable. In Figure 1 the conditions which determine whether a broad strategy that includes one or several product offerings will be more successful than a focused strategy that includes one or a few product offerings are summarized. It can be seen that although market heterogeneity is a nec- Number of Product Offerings One Several Heterogeneous Broad Homogeneous market with market positive synergies between segments Strategy Heterogeneous Heterogeneous market with market with negative synergies Focused negative synergies between groups of between segments segments and positive synergies within groups Figure 1. External factors supporting a focused strategy. essary condition in order for a focus strategy to be viable, it is not a sufficient condition: Syner- gies across segments must be absent or nega- tive before a focus strategy can be adopted. Structural Preconditions for Cost Leadership In a previous attempt to provide a contingency basis to the generic strategy concept, Day (1984) linked customer price sensitivity to the viability of a cost leadership strategy. This approach con- fuses the necessary and sufficient conditions that are needed in order for a cost leadership strat- egy to be successful. Price sensitivity is, at best, a necessary but not sufficient condition for cost leadership. Greater price sensitivity increases the advantage a cost leader has over other firms, but it is not sufficient to justify adopting a cost leadership strategy. For example, in the gasoline retailing industry, customers are extremely sensitive to prices, but the optimal scale of operations is low relative to the size of the total market. In this situation, in- stead of a single cost leader emerging to enjoy supra-normal profits, many firms have similar cost structures. Yet, in order to maintain sales, these firms are forced to slash prices until profits virtu- ally disappear. There are external preconditions for a cost leadership strategy but customer price sensitiv- ity is a minor consideration. A cost leadership strategy is viable only if cost structures vary across competitors within an industry in ways other than in direct ratio to output. Cost struc- tures can deviate from a direct ratio due to varia- tions in the quality of management across com- petitors, as a result of economies of scale, or as a result of economies independent of scale such as learning effects, or preferential access to in- puts or distribution channels. The first factor, competitors' incompetence, usually is not a strong foundation for a sustained cost leadership strategy because it is dependent on factors beyond management control. Each of the remaining factors provides a more substan- tial basis for a cost leadership strategy, but it is argued that each is determined by industry structure. 392 Perhaps economies that are independent of scale can provide the most durable basis for a cost leadership strategy. These can be grouped into three categories: access to raw materials, access to product or process technology, and ac- cess to distribution channels. But whether or not each of these economies can offer a firm founda- tion for cost leadership depends on external factors. In the case of access to raw materials, this foundation depends on the supplier industry's characteristics. If the supplier industry enjoys considerable economies of scale and if entry barriers are high, the potential for high transac- tion costs (Williamson, 1981) exists, suggesting that cost advantages may be derived from back- ward integration. Alternatively, if there is high variability across suppliers in the basic (excluding transaction costs) cost of raw materials (as is the case in many extractive industries), cost advan- tages, again, can be gained if access to materi- als can be controlled. Thus, for example, a cor- nerstone for success in the oil business tradition- ally has been access to crude oil reserves that have the lowest extraction costs ("Why Things," 1982). Where upstream industries are character- ized by low barriers to entry and where econo- mies of scale are few, little opportunity exists to generate differential cost structures. In order for a firm to achieve differential cost structures, independent of economies of scale in the production process itself, it must either pro- duce defensible technological breakthroughs or pursue learning effects, or, preferably, both. Here again, how viable this will prove to be depends on the industry or, more specifically, on the industry's maturity. Abernathy and Utterback (1978) argued that as an industry matures both the rate and charac- ter of innovation change. Initially, innovations come rapidly and tend to be radical in nature. Later, they are slower in coming and are more incremental in nature. Thus, predicating a cost leadership strategy on heavy research and de- velopment (R&D) spending may be sensible when an industry is new but may not pay off in more mature industries. Being first with a new technology, even if this is formally protected by patents or copyrights, only provides the firm with a temporary cost ad- vantage because imitation is inevitable. In order to sustain any cost advantage innovation may provide, the firm must buttress the initial advan- tage by capitalizing on learning effects. But, learning effects also depend on industry char- acteristics: Experience curve slopes vary widely from prod- uct to product.... In some industries the slope may be as steep as 60%; in others it may not occur at all. (Ghemawat, 1985, p. 144) Cost declines with experience seem to be most significant in businesses involving a high labor content performing intricate tasks and/or com- plex assembly operations (aircraft manufacture, shipbuilding). (Porter, 1980, p. 12) If an industry is not characterized by a suffi- ciently "steep" learning or experience curve that will provide a significant cost advantage for those firms that are "lower" on the curve, a cost leader- ship strategy based on pursuit of learning effects will collapse. Here again, how viable the cost leadership strategy will be is tied to exogenous factors. Finally, cost advantages independent of scale can be gained from preferential access to distri- bution channels, but this also will depend on industry structure. Forward integration can pro- vide cost advantages by capturing the best loca- tions for distribution, leaving the less desirable sites for the competitors. Economies of scale provide the other major durable basis for a cost leadership strategy. Ev- ery industry is characterized by unique optimal scales of operations, which are determined by product and market characteristics and by pro- duction and marketing technologies. For example, manufacturing a wide-body air- craft demands a greater scale of operations than manufacturing a light, recreational aircraft, in part, because a wide-body aircraft is technologi- cally more sophisticated. Fast food restaurants can achieve a greater scale of operations than French restaurants because customers tend to prefer a standardized product from the former 393 and a distinctive product from the latter. Nuclear power generation demands a greater scale of operations than solar power generation because the processes involved in the former occur most efficiently at a much larger scale of operations than is the case for the latter. If the optimal scale that these exogenous fac- tors dictate is small relative to the size of the market, many firms can achieve the same opti- mal cost structure, and as a result, a fragmented, highly competitive industry will develop. Any firm that tries to gain a dominant market share in such an industry will experience diseconomies of scale and will suffer accordingly (Porter, 1980). However, as the optimal scale of operations approaches one-half of the size of the market, only one competitor can enjoy operating at full capacity at the optimal scale of operations. In order to become the cost leader, a firm must build operations which approach that optimal scale. Of course, given that the optimal size rela- tive to the market size is high when cost leader- ship is viable, any aspiring cost leader must achieve a dominant market share so that the cost benefits gained from its efficient scale opera- tions are not nullified by overcapacity. A cost leadership strategy based on econo- mies of scale can only guarantee superior perfor- mance if only one firm can achieve it (Porter, 1985), and this can occur only when the optimal scale of operations exceeds one-half of the size of the market. But, the optimal scale of opera- tions is determined by production, distribution, and product and market constraints. Similarly, although cost leadership also can depend on economies independent of scale such as experi- ence or learning effects, preferential access to raw materials, or distribution channels, these, in turn, depend on industry structure. Market-Based Preconditions for Product Differentiation The price sensitivity which Day (1984) linked to cost leadership is really the inverse of perceived product differentiation, which Strategic Planning Associates link to the product differentiation strategy. This relationship between price sensi- tivity and differentiation can be clearly identi- fied if the conjoint analysis framework, described above, is used. Attribute weights can be stan- dardized so that they sum to one. If this is done, the inverse relationship between price and other attributes is clear; that is, the weight accorded the price attribute is one minus the sum of the weights attached to all other attributes. If those other weights are low, the customer will be ex- tremely price sensitive and will care little ab