Andrew Mair Learning from Japan

Andrew Mair Learning from Japan - LEARNING FROM JAPAN?...

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: LEARNING FROM JAPAN? INTERPRETATIONS OF HONDA MOTORS BY STRATEGIC MANAGEMENT THEORISTS ANDREW MAIR NISSAN OCCASIONAL PAPER SERIES NO. 29 1999 NISSAN OCCASIONAL PAPER SERIES PAST PAPERS No.14 David Williams, Reporting the Death of the Emperor Showa. No.15 Susan Napier, The Logic of Inversion: Twentieth Century Japanese Utopias. No.16 Alice Lam, Women and Equal Employment Opportunities in Japan. No.17 Ian Reader, Sendatsu and the Development of Contemporary Japanese Pilgrimage. No.18 Watanabe Osamu, Nakasone Yasuhiro and Post-War Conservative Politics: An Historical Interpretation. No.19 Hirota Teruyuki, Marriage, Education and Social Mobility in a Former Samurai Society after the Meiji Restoration. No.20 Rikki Kersten, Diverging Discourses: Shimizu Ikutaro, Maruyama Masao and Post-War Tenko. No.21 David W. Campbell, Explaining Japan’s Saving Rate. No.22 Penny Francks, The Origins of Agricultural Protection in Japan. No.23 E.S. Crawcour, Kôgyô Iken: Maeda Masana and His View of Meiji Economic Development. No.24 Irena Powell, A Japanese Literary Response to the Vietnam War. No.25 Brian Powell, Intellectuals and Politics in Late Edo Japan: A Playwright’s View. Mayama Seika’s ‘Genboku and Chôei’. No.26 Tomida Hiroko, Japanese Writing on Women’s History. No.27 Tessa Carroll, From Script to Speech: Language Policy in Japan in the 1980s and 1990s. No.28 Ian Neary, Political Culture and Human Rights in Japan, Korea and Taiwan. NISSAN OCCASIONAL PAPER SERIES Every Friday of term from 5.00 pm, a public seminar is held at the Nissan Institute of Japanese Studies in Oxford. The Nissan Occasional Paper Series was established to publish some of the papers presented at this seminar series and to disseminate them free to a wider audience. To be put on the mailing list, please write to the Editor at the address given on the back cover. Starting with Issue No. 27, the papers are available at the Nissan Institute’s World Wide Web site at http://www.nissan.ox.ac.uk in a PDF file format. ANDREW MAIR Dr Andrew Mair is a lecturer in organisation and strategy at Birkbeck College, University of London. A shorter version of this paper appears in the Journal of Management Studies, 36.1. Mark Rebick Editor Learning from Japan? Interpretations of Honda Motors by strategic management theorists Abstract Following the growth of Japanese automobile and electronics manufacturers on world markets during the 1970s and 1980s, western management theorists sought to explain how these companies were able to penetrate markets previously dominated by entrenched North American and European firms, as well as to identify transferable lessons from their success. The Honda Motor Company attracted the particular attention of strategic management theorists, in both the academic and consulting worlds. Strategy theorists frequently claim that their work is very close to real world management practice. This paper evaluates whether the picture of Honda Motors’ business practices that has been painted by the strategy theorists is accurate. It reviews a number of case studies and references to Honda and the inferences drawn from these. It then offers a critical analysis which focuses on both empirical inaccuracies and a theoretical tendency to reduce complex company behaviour to simple one-sided explanations, often quite contradictory to explanations of Honda offered by other strategy thinkers. The final part of the paper addresses the issue of why the western strategy theorists have adopted the particular approaches to learning from Japan that are exemplified in the Honda case. INTRODUCTION Strategic management theorists, whether academics or consultants, pride themselves on the close relationship between their theories and the practice of strategy in real organisations. They maintain strong links to companies and other organisations through empirical research and through intertwined career paths. Like the other sub-disciplines of management, they are often determined to make their theoretical outputs comprehensible, pragmatic and useful to practising and potential 2 managers, through teaching, consultancy, conference presentations and the flagship business school journals which academics cohabit with consultancy professionals and which frequently incorporate empirical case studies based on ‘inside knowledge’ of practice. Yet among the sub-disciplines, it is strategy, through its pedagogy, that has placed the ‘case study’ approach on a pedestal, strongly suggesting that in this domain relevance and practicality are as important as theory. More generally, so closely do management thinkers appear to be tied to management practice that it is not uncommon for members of other academic disciplines to privately and publicly deride the study of management as lacking substantial theoretical foundation. Is this received view of the relationship between the theory and practice of strategic management accurate? Consider the process of management theorising from a systems perspective, with the inputs being empirical data and knowledge gleaned from observation and participation, the transformation being theory building, and the outputs being ideas with practical relevance to organisations. just how effective is this process? The question then becomes: The most commonly perceived weakness perhaps lies in the theory building transformation process. My interest in this article lies partly in this theory building transformation process, but much more in subjecting to greater scrutiny the perceived strengths of strategic management that are believed to lie in proximity to management practice on both the input and output sides. In brief, my thesis is that strategy theorists may not be as ‘close’ to practice on the input side as they might like - and like others - to believe, that they may be too ‘close’ to practice on the output side, and that the causal relationship explaining this paradoxical situation may run ‘backwards’ from the output side to the input side via the theoretical transformation process in a distorted and distorting version of ‘the customer knows best’. To put the matter bluntly, the broad hypothesis to be explored here is that the marketing function of the ‘strategy industry’ - I include in this both the consultancy companies and the academics - blinkers a theory building process the products of which are legitimated with reference to case studies of sometimes dubious value. In other words, a quality control process is carried out 3 on the products of this ‘business of knowledge’, and severe problems are revealed. Japanese automobile and electronics companies have been the focus of much analysis by western management theorists since their successful entry into world markets during the 1970s and 1980s. For strategic management theorists, the Honda Motor Company has played a prominent role. I therefore test this thesis with an in-depth examination of how the Honda Motor Company has been used as a building block for theories and models in strategic management over the last two decades. serves my purpose well, for two reasons. frequently cited in the literature. Honda First, Honda is very Some years ago a professor of strategy at a well-known business school told me that he really liked Honda, ‘a lovely little company’, as he put it, that made an excellent case study for MBA students. I dare say that this opinion, or something similar, is widely shared amongst management educators. Frequent citation has led academics, students, and practising managers alike to feel a certain familiarity with Honda, just as they might feel with General Motors, 3M or General Electric, for instance. Second, Honda has been marshalled as evidence to support a variety of sometimes opposing positions in most of the key theoretical debates in strategic management, namely the debate on strategy process paradigms between the analytical-planning school and the learning-adaptation school, the debate on strategy content paradigms between supporters of industry analysis/market positioning and resource-based approaches, and the debate within the last of these between supporters of ‘core competencies’ and ‘core capabilities’ (these debates are set out perhaps most clearly by de Wit and Meyer (1994)). Honda’s deep - if perhaps unwitting from the company’s perspective - implication in these debates resurfaced in the Summer 1996 issue of California Management Review, where participants fiercely debated the ‘true meaning’ of Honda: once more, and significantly as we will see below, failing to agree. The Honda case is thus a rich vein for examination of the mineral quality being extracted and refined at the coalface of the strategy industry. 4 The article is organised in two parts. The first part reviews the ways that Honda is drafted in to support seven different theoretical positions: advanced by Boston Consulting Group (1975), Pascale (1984), Quinn (1991; 1996), Mintzberg (1987; 1989; 1991; 1996a; 1996b), Abbeglen and Stalk (1985)/Stalk and Hout (1990), Hamel and Prahalad (1994), and Stalk, Evans and Shulman (1992). These seven cases are not exhaustive but they do provide a useful cross-section that represents very significant theories and highly influential strategy thinkers from both the consultancy and academic wings of the strategy industry. The Boston Consulting Group is particularly well represented, since Abbeglen, Stalk, Hout, Evans, Shulman, as well as Goold (1992; 1996) also represent its evolving views.1 The second part of the article critically examines the use made of Honda case studies by the strategy industry. First, the empirical accuracy of some studies is subjected to scrutiny. Second, the various crises that Honda has experienced over its fifty year history are examined, crises which appear to have been ignored by the strategy industry in its rush to support new theories with an unequivocal success story, despite the potentially rich theoretical lessons to be drawn from analysing business failures. Third, and perhaps most significant, the strategy industry’s shared meta-theoretical (epistemological) focus on single theoretical dimensions to explain Honda’s success in a reductionist fashion is highlighted. The outcome of this focus has been a series of misleading and false debates, which in nearly every case - and at this point I return full circle to the thesis of the article - tell us a great deal more about the strategy industry than they do about Honda. Moreover, this meta-theoretical approach proves singularly inappropriate in the case of Honda, because a vital element of Honda’s strategic practice is precisely an innovative transcendence of the false dichotomies of management thinking and practice (Mair, 1994a; 1996). The sting in the Honda tale is that despite - indeed perhaps because of - all the bluster, claim and counterclaim, in twenty years of debate the strategy industry has failed to grasp one of the key underlying competitive drivers at Honda. 5 1. HONDA AND THE STRATEGY WRITERS 1.1 The analytical approach of Boston Consulting Group Boston Consulting Group (BCG) published its analysis of the strategy alternatives facing the British motorcycle industry in 1975 (Boston Consulting Group, 1975). The study had been commissioned by the British government in the face of what appeared to be the collapse of the industry’s competitiveness, especially in once significant overseas markets such as the United States. The BCG report was unusual as a high-level consultancy analysis in that, with the exception of technical and commercial appendices and with some commercially sensitive material deleted, a 120 page report was published in full by the government, thus providing outsiders with a fascinating insight into the world of one of the strategy industry’s foremost players. The report had two main empirical focuses and one overriding theoretical focus. Empirically, it analysed the competitiveness of two motorcycle industries, the British and the Japanese, which were compared and contrasted both in terms of product development and production systems at their respective home bases, and in terms of their sales and distribution systems in the United States. Theoretically, the report analysed the preconditions for the realisation, in the Japanese case, and non-realisation, in the British case, of economies of scale in product development, production, and sales and distribution. Importantly, both empirical and theoretical analyses focused on the period during the late 1960s and early 1970s when the Japanese motorcycle producers effectively displaced the British companies in the North American market for medium and large-sized motorcycles; the British motorcycle market share fell from 11 percent to 1 percent even as the market size (in units sold) more than doubled between 1968 and 1974, leaving the Japanese with 87 percent, and Honda alone 43 percent, by the latter year. BCG’s explanation of the relative success of the two industries pinpointed a nexus of inter-related economic factors. industry Honda furnished the majority of examples. For the Japanese On the side of 6 production, scale, costs, and technology were said to be interlinked. Large volumes permitted lower costs, both through the use of specialised production technology, and through the imputed operation of the ‘experience curve’, an analytical tool favoured by BCG at the time (see Stalk and Hout, 1990) which proposed that costs declined with ‘experience’ (i.e. as accumulated volumes increased). Evidence of Honda’s large volumes and specialised production technology was presented, although, lacking any evidence on Honda’s cost structures, BCG was only able to show that product prices declined over time, which was said to be a ‘useful guide to movements in underlying costs’ (1975:39). Causally, then, higher volumes led to lower costs which were passed on in lower prices, which in turn increased market share and therefore profitability. On the side of sales and distribution, the Japanese maintained unit spending on marketing as US sales mounted. Moreover, it appeared that in the United Kingdom, at least, Honda had been willing to sustain financial losses for a period to encourage market share to grow, and therefore create the higher volumes and consequent lower costs that would make low prices profitable. By contrast, in the United States Honda and the other Japanese companies had been permitted to advance from the smaller motorcycle segments to the medium and now large sized segments in part because the British retreated from a segment as soon as it became unprofitable. Causally, too, segment advance led to higher volumes and eventually to higher profitability. How had Honda, the market leader among the Japanese companies, managed to enter the US market with its smaller motorcycles in the first place, during the early 1960s? This question was briefly discussed in two places. The success of the Japanese manufacturers originated with the growth of their domestic market during the 1950’s. As recently as 1960, only 4 percent of Japanese motorcycle production was exported. By this time, however, the Japanese had developed huge production volumes in small motorcycles in their domestic market, and volumerelated cost reductions had followed. This resulted in a highly competitive cost position which the Japanese used as a springboard for penetration of world markets with small motorcycles in the early 1960’s (1975:xiv). 7 It is often said that Honda created the market -- in the United States and elsewhere -- for what we have called secondary uses of motorcycles, through their extensive advertising and promotion activities; and it is true that Honda presented the attractions of motorcycling as a ‘fun’ activity in a new way, and with a level of media support not previously attempted by motorcycle manufacturers. However, the success of this campaign depended in the last resort on the fact that the lightweight machines that were then the company’s primary product were fun and easy to ride, did not give the mechanical problems that had traditionally been associated with motorcycles, and were cheap to purchase (1975:18). Aside from these brief sentences and some ‘price-experience curve’ graphs going back to 1959, the BCG analysis focuses on events and competitive pressures in the late 1960s and early 1970s during a period of rapid growth in the US market for medium and large-sized motorcycles. The overall issue throughout remains how to explain the ‘cost superiority’ of the Japanese relative to the British. While they do not receive much theoretical prominence and tend to be interspersed with assertions about economies of scale and experience curves, it is worth noting that as early as 1975 the BCG analysts had also drawn attention to a number of management concepts at work in Japan that were only later to receive much more sustained focus in the West and the importance of which was not widely recognised until after the publication of Womack, Jones and Roos (1990). These include design for manufacture (1975:xiii); utilisation of the Sloanist principle of shared core components in products to permit economies of scale plus product variety 1975:xviii); the importance of linking product technology to marketing in new product development, low-inventory logistics systems, close relations between assembly companies and suppliers including close-by locations, products developed according to target final product costs, and the development of supplier capabilities by assembly companies (1975:62-4). The last part of the BCG report also contained a prescient analysis of potential collaboration between British and Japanese motorcycle companies as a strategy option; only a few years later in 1979 Honda would launch a complex collaboration in the European automobile sector with the British car maker British Leyland that would endure at least two decades (see Mair, 1994a; 1998b). 8 When the BCG analysis turns to the detailed strategy alternatives for the British industry, however, the focus remains very much on cost structures, and the mix of market segments, price differentiation, product range, and production systems that will be necessary for the British to compete effectively again. The BCG analysis has proved very influential in moulding an understanding of Honda. The study formed the basis for business school case studies at Harvard Business School and UCLA. Many years later, its analysis of Honda’s US strategy was still being reported even though BCG itself was no longer referred to as the source. Hence Wheelen and Hunger (1995:194) describe Honda’s ‘encirclement’ strategy whereby one by one the company’s motorcycles captured every market segment in the United States except the heavyweight segment controlled by HarleyDavidson. 1.2 Pascale’s ‘Honda effect’ emphasizes a strategic learning process In a direct response to the BCG study, in a 1984 academic publication, Pascale argued that in fact ‘the Japanese’ view strategy differently from Americans and Europeans, and that they find a number of Western concepts, such as ‘portfolio theory’ or the ‘experience curve’, too formulaic, and indeed too easy to ‘read’ (and therefore to counter) from the behaviours of competitors. Pascale’s approach is to set out two different perspectives on Honda’s breakthrough into the American motorcycle market in the early 1960s. The first approach is based on the BCG study, which Pascale argues has superimposed an inappropriately rationalistic interpretative framework onto Honda’s actual strategy. Pascale is particularly troubled by the implication in the BCG study that Honda followed an economically driven strategy in the years around 1960 based on achieving low costs through high volume, and he quotes the first of the short fragments of the BCG report which discusses this period (as cited above; 1975:xiv) as an example. He also, and at much greater length, then quotes a Harvard Business School case study (presented as based on the BCG analysis, yet which adds a wealth of detail, plus a bullish business school ‘success story’ overtone which is 9 not apparent in the original BCG study) which certainly makes Honda appear to have followed a clear and logical strategy. The second perspective describes what Pascale terms the ‘organisational process’ behind Honda’s American success. Asked by Pascale to describe the sequence of events during Honda’s early 1960s US market entry in ‘fine grain detail’, senior Honda managers did just that. To Pascale, their story ‘highlights miscalculation, serendipity, and organisational learning - counterpoints to the streamlined “strategy” version’ represented by BCG. He proceeds to describes the idiosyncratic characters and leadership of company founders Soichiro Honda and Takeo Fujisawa, the product design (rather than production process, pointedly) advantages that Honda developed, in particular the wedding of a hightech small engine to an innovative design in the 1958 Super Cub small motorcycle. The success of this product in Japan permitted the 300,000 capacity Suzuka factory to be built and Honda to distribute it directly to US retail outlets (primarily bicycle shops), bypassing the traditional dealers. Pascale quotes Kihachiro Kawashima, one of the managers charged with exploring the US market from 1958, at length, to present a vivid first-hand impression of Honda’s entry into the United States when the first sales subsidiary was opened in 1959. Kawashima recounts that there was no definite strategy in terms of how to penetrate the US market. The initial inventory was arbitrarily composed of 25% of each of four models: one of which was the new Super Cub, the other three being larger machines. MITI’s currency restrictions forced the company to adopt a very low investment approach. When mechanical problems arose with the larger motorcycles they had initially thought would sell best, the Honda team in the United States explored contacts with non-traditional retail dealers who had contacted them with a view to distributing Super Cubs. Sales took off in the United States during the early 1960s as they had done in Japan from 1959 onwards. Pascale’s interpretation of this second story is clear. Far from deliberately redesigning the US motorcycle market, Honda’s start-up team had ‘backed into’ a success that was not intended in their original strategy. Moreover, the well-known image-changing advertisement (‘You meet the nicest people on a Honda’) associated with Honda’s success did 10 not appear until 1963, after Super Cub sales had started to rise rapidly. The main lesson is therefore that Honda was quick to take advantage of its unexpected success, particularly to exert tighter control over its American distribution system. Indeed for Pascale, the second account is not simply a different perspective. It clearly exposes aspects of real-life strategy-making which are ignored by the ‘Western consultants, academics and executives’ with their ‘preference for oversimplifications of reality and cognitively linear explanations of events’. What matters, and what Pascale believes requires more research, is how organisations deal with miscalculations, mistakes and serendipity, just as Honda was quick to learn and react to the unexpected. Pascale generalizes the lesson both from this one strategic event (the unexpected boom in demand for Super Cub in the United States) and this one company (Honda) to describe a ‘Japanese approach’ to strategy-making in general, in which Japanese manufacturing success is due less to an overarching strategic plan implemented with unfailing success than to a willingness to experiment and feed back input on the results from lower rungs in the organisation. In practice, therefore, strategy evolves with the organisation rather than being the bold and unerring guidance of a leader. Pascale’s article and interpretation has of course been highly influential, particularly in that part of the strategy industry which approaches issues from a sociology/organisation background as opposed to an economics/industry analysis background. It is excerpted at length in Pascale (1990) and reprinted again in 1996 in the special issue of California Management Review on Honda. In his own contribution to that debate, Pascale (1996) expands on his argument that the significance of Honda lies in its support for new and flexible organisational forms, as first developed, in different terms, in Pascale (1990). The article is also reprinted in Mintzberg and Quinn (1991; 1996). Mintzberg has made frequent reference to Pascale’s argument (discussed separately below). Other strategy writers have picked up Pascale’s story, sometimes adding their own interpretations. Thus Whiteley (1991:53) repeats the story, which he claims to be ‘thoroughly documented’ by Pascale, as an example of ‘getting the front-line to 11 listen to customers’ and ‘communicate upward’: what Whiteley now calls ‘low-ranking sales staff’ are said to have been able to get the company to change its strategy. Floyd and Wooldridge (1996) cite Pascale’s story as an example of ‘strategy from the middle out’. They claim that the 50cc (Super Cub) motorcycles were not brought to United States at all except for the Honda employees to use (Rumelt (1996) in his review of Pascale, implies the same). Supposedly, only when seen by a Sears representative did the Honda team ask for Super Cubs to be sent from Japan. Floyd and Wooldridge interpret this Honda story as a combination of deliberate strategy and emergent strategy (Mintzberg is referenced, although for him (see below) this is only a story of emergent strategy.) Other writers are a little less willing to commit themselves to Pascale’s version of events but still view the story as significant. Thus Whittington (1993:33-4) repeats it as an example ‘according to Pascale’ in reviewing theories of adaptiveness compared to top-down planned strategies. Indeed for a number of years it has become almost de rigueur to at least mention Pascale’s study of Honda as “the” case to counterpoise to the design/planning schools of strategy, usually by introducing an emergent/incrementalist approach to argue that the planning approach is not all it is claimed to be (Kay, 1993a; Silbiger, 1994; Stacey, 1993). 1.3 Quinn’s case study portrays an idiosyncratic Honda Quinn (1991) published a long case study of Honda within a university strategic management textbook which took Pascale’s basic focus on emergent strategy and organisational process and expanded it in an attempt to explain all of Honda’s history. Quinn’s main sources were Pascale (1984), Sakiya (1982 edn) and Sanders (1975), supplemented by media articles. Soichiro Honda is described as a colourful individualist and eccentric mechanic, who lived a hermit-like existence for part of his life, was irreverent, against authority, but was nonetheless a technical perfectionist. Accident-prone and lacking business sense, he allied himself with Takeo Fujisawa as his business manager. Honda Motors was an entrepreneurial company that grew during the 1950s on the basis of 12 the hot-tempered Mr Honda’s technology and designs. At the same time, with no MITI or keiretsu support, Fujisawa engaged in risky financial management to support investments. Launch of the Super Cub in 1958, based on advanced technology and innovative design, propelled Honda to becoming the world’s largest motorcycle producer in 1959. Honda now entered the US market, again independently, without MITI support, and with the US staff all having to ‘muck in’. Indeed Japanese currency restrictions kept conditions difficult despite the popularity of the Super Cub by 1962. At this point in the case study a box with questions for the student reader is introduced: ‘How should Honda develop its U.S. presence? What should its advertising, pricing, distribution, inventory, product and service policies be in the U.S.? How should these be related to its Japanese prices, production facilities, development activities, and the continuing restriction by the Ministry of Finance on exports of Japanese Yen?’ (p.290). The questions are not answered, and discussion moves on to Honda’s entry into the automobile business, describing MITI’s efforts to obstruct Honda and Honda’s response of rapidly designing a car, the entry into FI car racing, the company’s first mass production car in 1967, and the 1972 CVCC engine which beat Honda’s competitors to meeting new US air pollution standards. A number of unusual features of Honda’s organisation are then briefly described: the focus on engineering and technology; Mr. Fujisawa’s ‘expert’ system for promotions outside the traditional hierarchy and ‘paperweight’ (flat) organisational structure; the ‘SED’ (sales, engineering, development) product development process; the focus on individualism in research and development. It is explained that Honda introduces new products in a ‘trial and error’ way, that workers design their own production equipment (‘Workers had automated whole body-panel and body-assembly sections [unsupervised by engineers]’ p.293), and that workers invented what later became best-selling Honda products. Quinn’s discussion of ‘production organisation’ focuses on the sense of equality, the absence of seniority, the willingness to ‘muck in’, and 13 democracy (‘On matters which affected them, associates [all employees] were asked their opinion’ p.293). Quinn explains that the owners retired to let younger managers take the reins in 1973. The new president followed Mr Honda’s traditional passion for technology, liked to drink sake, and would shout to make himself heard. projects. Various committees were set up to oversee specialist Honda became successful in the United States because of the ‘legendary’ quality of its products. However, the competitive nature of the market in Japan made it difficult to make profits there. Investments by other Japanese (and Korean companies) in North America are sketched, and Honda’s US expansion plans described. The success of the US operations is ascribed to the fact that ‘Honda had attacked all its costs in the United States aggressively’ p296. The result was that the Honda Accord became the best selling single model in the United States by the late 1980s. The case study ends with brief discussions of the state of competition in the US and European automobile industries (including such inaccuracies as: ‘Quotas had restricted imports of Japanese cars into individual countries in the past. However, Spain and some of the “Mediterranean countries” with low labour costs were pressing to break these quotas in 1992’ p.298). The student of strategic management is then set further discussion questions (e.g. ‘4. What implications does Honda’s past history have for the future of the auto industry?’ p.299). An appendix of statistical charts gives very general information on Honda, some of its global rivals, and the characteristics of regional markets. In short, in Quinn’s interpretation, Honda is a veritable ‘dream factory’ (after Tom Peters), fostering an individualist and bottom-up anarchy which is somehow both creative and productive. A new version of this case study was published in a new edition of Mintzberg and Quinn’s book (Quinn, 1996). The general line of the argument is retained, and the case reads like an atheoretical and uninterpreted potted history to the point of a list of random information. In their introduction to the case study, Mintzberg and Quinn (1996:614) propose Honda as an example of entrepreneurship, which, 14 along with other cases in the book, is one of the ‘most important startups of the modern era. These companies literally changed the entire profile of competition in their fields’. Like Pascale’s before it, Quinn’s study has found its way into other writings. Hence Stacey (1993:1) introduces his book on strategy and organisation dynamics, with its ‘chaos theory’ theme, by citing both Pascale’s (1984) and Quinn’s (1991) analysis of Honda, and comparing Honda’s growth during the 1970s and 1980s with the stagnation of General Motors, as a ‘well known example of the strategy game’ in which an upstart company is able to outflank an industry giant. 1.4 For Mintzberg, Honda epitomizes the value of learning over planning Henry Mintzberg has frequently drawn on Pascale’s story of the ‘Honda effect’; indeed he considers Pascale’s study to be perhaps the most influential article published in the management literature (Mintzberg, 1996a:78), and has drawn upon it not only in his general writing on strategy but in direct debates with key proponents of the ‘planning school’ such as Igor Ansoff and with an author of the BCG report itself, Michael Goold. By 1987 Mintzberg was proposing Honda as an example of a company which epitomized the way good strategy is ‘crafted’ rather than ‘planned’. Mintzberg refers to . . . . Richard Pascale’s account of how Honda stumbled into its enormous success in the American motorcycle market. Brilliant as its strategy may have looked after the fact, Honda’s managers made almost every conceivable mistake until the market finally hit them over the head with the right formula. The Honda managers on site in America, driving their products themselves (and thus inadvertently picking up market reaction), did only one thing right: they learned, firsthand (Mintzberg, 1987:70). Thus Honda’s successful entry into the US motorcycle market in the early 1960s is portrayed as an example of a ‘grass-roots approach to strategic management’. Significantly, Mintzberg continues: Of course, this view is overstated. But it is no less extreme than the conventional view of strategic management, which might be labelled the hot-house approach. Neither is right. Reality falls between the two. Some of the most effective strategies we 15 uncovered in our research combined deliberation and control with flexibility and organisational learning (Mintzberg, 1987:70). Notwithstanding this comment, Mintzberg has generally emphasized the first of the cited statements, rather than the second (although the second continues to surface, in a somewhat uneasy relationship with his view of Honda). Mintzberg (1989:32-4) again refers to Pascale’s research, repeating the first statement above. Pascale is cited as ‘someone writing about the reality of management, and the effect can be stunning’ (1989:358), and Honda is praised as a company with common sense, not overly-rational, willing to learn rather than impose distant corporate views. In similar vein, Mintzberg and Quinn (1991:95) describe the Pascale study as an ‘upsetting’ article which ‘challenges head on not only the whole approach to strategy analysis . . . but also the very concept of strategy formulation itself’. On the other hand, they claim not to wish to encourage rejection of ‘strategy analysis’, but to ‘balance the message’ with ‘practical lessons from the field’. Mintzberg (1991) again draws on Pascale’s Honda story in his debate with planning/design school strategist Igor Ansoff, published in Strategic Management Journal. Implying that most strategy thinking is either theoretical debate or over-elaborate statistical analysis, Mintzberg argues that Pascale’s story is perhaps the one factual empirical study published of real strategy formation. He calls it ‘Richard Pascale’s account by several Honda executives about how they developed on site the strategy that captured two-thirds of the American motorcycle market’, and it is compared with the ‘brilliantly rational strategy’ imputed by BCG’s analysts who ‘apparently never bothered to ask’. Mintzberg argues that if Honda had in fact been ‘rational’ in its planning it would not have attempted to sell the small motorcycles in the United States at all. The Honda case thus reveals the necessity of ‘emergent learning alongside deliberate planning’. The problem with the BCG analysis is that it ignored the emergent learning phase which, Mintzberg argues, had to come first, prior to the formal planning upon which BCG concentrates. Hence while both planning and learning are necessary, learning must 16 come first; accordingly, he concludes with a score of ‘1-0’ for informal learning over formal planning. Responding in the pages of Strategic Management Journal, original BCG author Michael Goold (1992) argues in favour of a synthesis or reconciliation between the design and learning schools of strategic management, but claims that ‘polemics and prejudices’ are standing in the way (i.e. Mintzberg claims also to support some type of combination but in fact argues for the superiority of learning). Goold contextualizes the BCG study, arguing that the particular questions asked of Honda were driven by the particular needs of the British motorcycle industry, admitting that the analysis of Honda was therefore incomplete in that it did not analyse how the Honda strategy was formulated. Yet Mintzberg’s own approach is criticised as impractical in its apparent proposition that trial and error is the best form of strategy. The advantage of the BCG approach was the way it helped participants reflect upon what strategies might work for British industry. Goold concludes that both planning and ‘learning from others’ are necessary in equal measure. As the debate shifts to California Management Review, Mintzberg (1996) notes that Goold has now emerged to identify himself as a BCG author. Mintzberg argues that BCG’s analysis was misleading because it apparently inferred how Honda formulated its strategy. Read that report and the implication is that you should lock yourself in your office and do clever competitive analysis. never would have produced its strategy that way (1996:96). Honda Mintzberg does not in fact believe that organisations should conduct random experiments but instead lauds the value of ‘exposing oneself to the chance to be surprised by the marketplace and so to learn’ (1996:96). He continues to argue that Honda ‘seemed to do everything wrong’, ‘the story violates everything we believe about effective strategic management (and much that BCG imputed to those clever Japanese)’ (1996:96). In fact the Japanese simply used common sense, learning on the ground, whereas ‘we’ (in the West) tend to be overlyrational and try to dictate strategy to our organisations from afar. 17 Mintzberg admits that Honda used its Japanese production experience and volume, but only after learning that it was Super Cub model that would sell in the United States. Mintzberg then proceeds to criticise BCG on the basis that the British motorcycle industry declined further during late 1970s after publication of the BCG report, and that managers within the British motorcycle industry believed that outside consultants brought in during the early 1960s had contributed little or nothing, having no real understanding of the industry. He concludes that a learning approach is necessary if possible strategies are not to be rejected too soon because of the analytical approached adopted. Similarly, in learning from past strategies or from other companies, investigating how decisions were actually made is superior to imputing explanations based on an imposed logical system. He concludes that there is still too little emphasis on the learning aspect of strategic management. Goold (1996) attempts again to place the original BCG report in context, asking what practical advice to a British motorcycle industry in a strategically difficult and political sensitive environment would be offered by BCG, Mintzberg and Pascale, and Richard Rumelt (1996), who as the author of a UCLA business school case study based on the BCG report, has also contributed to the California Management Review debate. The only advice that might be drawn from Pascale’s and Mintzberg’s work would be based on their interpretation of Honda’s success as apparently serendipitous; accordingly the implication would be to try out alternative new motorcycle models and marketing methods, learn from the experience and build on any good results. This trial and error approach Goold dismisses as inappropriate for a British industry that was almost bankrupt and so did not have this luxury. Similarly, Rumelt’s view that the key competitive advantage lies in a superior product does not appear to recognise the uncertain technology capabilities of the British industry at the time. Admitting that BCG’s analysis also failed to provide a workable answer, both the Pascale/Mintzberg and the Rumelt approach are argued by Goold not to be very useful in the real context faced by the British motorcycle industry. 18 Goold proceeds with his attempt to combine the different approaches, each of which he believes has something to offer but also has its limitations: the BCG approach is useful for testing the viability of, but not generating strategies; the emergent approach is useful in stressing the importance of learning and adaptation, but not in choosing alternatives when an industry is already in crisis; and Rumelt’s approach focuses on the importance of innovation but has little to offer in terms of where innovation comes from. Goold (1996:102) therefore claims to be an ‘unrepentant synthesizer’ of what are necessarily partial approaches to strategic management. 1.5 Abbeglen, Stalk and Hout use Honda to support theories of competitive cycles, product diversity and time-based competition We return to a new phase in the strategy analysis approach of the Boston Consulting Group in the work of Abbeglen and Stalk (1985) and Stalk and Hout (1990). Like Pascale, Abbeglen and Stalk (1985:42-52) turn to Honda to provide cases of dramatic incident and competitive breakthrough. They begin their analysis by describing a typical view of Japanese companies as focused on cost cutting, ‘willing to sacrifice short-term profits to satisfy their relentless obsession for growth and increased market share’ (p42) (a view certainly promoted by BCG, 1975). In fact, they argue, in Japan successful companies establish a ‘winner’s competitive cycle’ in order to survive. This involves a virtuous circle of increased volume, decreased cost, increased profitability and financial power, followed by reinvestment to fuel growth (a view also promoted by BCG, 1975). While this cycle can be found at work throughout the world, it has played a significant role in competition in high-growth Japan, and has been both to topple market leaders or to defend an existing powerful competitive position. Honda furnishes examples on both accounts. Abbeglen and Stalk’s first case study refers to competition in late 1950s Japan, when Honda overtook the erstwhile domestic motorcycle leader, Tohatsu, despite starting from an unpromising financial position. Following a dramatic industry shake-out, Tohatsu was declared bankrupt. The explanation is laid out squarely in terms of industry 19 economics. Tohatsu was complacent, conservative, and did not attempt to increase sales volumes in the fast-growing domestic market of the late 1950s. Honda, by contrast, borrowed heavily, increased volumes and market share rapidly and therefore reduced costs; on this basis Honda was able to ‘overwhelm the competition’ (p.45). Later, it is explained that Honda’s upset of Tohatsu was due to a ‘combination of aggressive financial policies and fundamentally sound product and marketing strategies’ (p.157-8). Hence Honda took major financial risks to invest in growth, which was repaid by higher volumes, lower costs and, as competitors fell by the wayside, higher profits. Rate of growth, and not existing market share, is the best measure of expected company performance. The second case study analyses how Honda retained its leadership position in the Japanese motorcycle market in the face of a concerted attack by rival firm Yamaha. During the 1960s and 1970s Honda had focused its growth efforts and the resources (human, technical and financial) it had developed in the motorcycle sector on entering the automobile industry. As Yamaha continued to invest and to expand its own motorcycle model range, it caught up with Honda in domestic market share, to the point where, by 1980, Yamaha believed it could become domestic market leader. Yamaha invested in a large new factory convinced that Honda was now too focused on its automobile activities to offer much resistance. However, by 1982 Honda had begun mount a strong defence of its position, and over a period of only eighteen months pushed its market share up from 38 to 43 percent, while Yamaha’s fell from 37 to 23 percent. Abbeglen and Stalk explain Honda’s success in terms of a ‘counter strategy that was both simple and innovative’ (p.49). The simple parts included massive price cuts, promotions, and larger inventories maintained at dealers. The innovative part was ‘the use of product variety as a competitive weapon’ (p.49). Honda rapidly expanded its model range and renewed many older models, at a rate that Yamaha, which had not yet attained market leadership, could not match and therefore rapidly fell behind. Despite the heavy spending on price cutting and new models in the domestic motorcycle market, Honda, financially 20 supported by its success in the automobile sector, improved its overall financial position between 1981 and 1983. Honda continued the pressure of relentless new model introductions in 1983 and 1984, to the point that Yamaha publicly admitted failure. Honda’s success is therefore explained by four factors: Honda was larger and had more resources; Honda was more diversified and could now draw financially on its automobile sector activities; Honda responded to Yamaha just as the latter’s new plant was completed (while its costs were still being borne but results had not been obtained); and Honda introduced so many new products that Yamaha’s offering appeared obsolete. The latter two factors reveal that Honda excelled at both timing and speed of reaction. Stalk and Hout (1990:58-9) repeat in summary form this story of the ‘H-Y war’, now focusing more on the ‘massive new product introductions’ than the price cuts and advertising expenditures. While the latter are mentioned in passing, the economics of scale and debt-led growth strategies that played important roles in Abbeglen and Stalk (1975) (and BCG (1975)) have now been downplayed, and the ‘H-Y war’ becomes an example of the ‘exploitation of the benefits of flexible manufacturing to the point that a new competitive thrust emerged - the variety war’ (p.58). The BCG analysts have essentially changed the theoretical framework they are imposing upon the same events, towards the new BCG interpretation of Japanese manufacturing success as an outcome of ‘timebased competition’. 1.6 Stretch, leverage, and the core competencies of Hamel and Prahalad Like so many recent strategy writers belonging to the ‘resource-based’ school, a key interest of Gary Hamel and C.K. Prahalad is how small companies can enter and compete successfully in existing industries that are apparently dominated by large entrenched companies, and how the traditional competitors can learn and should respond. Honda becomes one of a small number of key breakthrough case study companies in their analysis. The company plays a prominent role from the start of Hamel 21 and Prahalad (1994), where it serves as an example of the fact that it is indeed possible for small companies to break into apparently stable markets (p. xvi, 139, 163). Honda strategy is then used widely to illustrate, exemplify and legitimate Hamel and Prahalad’s principal theoretical concepts: stretch, leverage and core competencies. Hamel and Prahalad are not slow to criticise other contemporary strategy prescriptions, including re-engineering, downsizing, time-based competition, adaptive incrementalism, and core capabilities, even though advocates of the last three, at least, also claim the support of Honda for their theory. Honda, it is argued, sets itself stretched targets, such as competing head-on with the dominant players in its industries, and it leverages its resources to reach the targets. Honda is cited as a company characterised by ‘unreasonableness of ambition and creativity in getting the most for the least’ (p170). At root, it is this characteristic rather than some mythical ‘Japanese management system’ that explains Honda’s success. Leverage rather than mere resource allocation is the’ key. For instance, Honda only spends one-quarter as much as General Motors on research and development yet is a technology leader whereas the GM is not (p163, p172). Honda is cited as an example of three of Hamel and Prahalad’s five ways to achieve resource leverage. First, the fact that Honda has launched fewer new models than GM and Ford (and yet its products are superior) is said to show that Honda learns better (faster and cheaper) from experience than do its competitors: ‘Honda makes a mockery of the experience curve’ (p180) (a swipe at BCG en passant). Second, Honda ‘recycles’ technologies by reusing them across its product range (motorcycles, automobiles, and various power products), leading to an ‘unmatched R&D efficiency’ (p187). Third (and apparently in a nod to BCG), Honda is said to have built up a strong position in the ‘undefended territory’ of the American motorcycle market before, it is implied, using this as a springboard to take on the Big Three American car makers (p190). The key form of leverage appears to be the second type, which is expanded into the concept of ‘core competencies’. has escaped ‘the myopia of the served market’: Honda, it is argued, 22 Conceiving of itself as a world leader in engines and powertrains, Honda leveraged that competence into cars, lawn mowers, garden tractors, marine engines, and generators (p91). Indeed, for Honda, competition is perceived not in terms of the success of individual product lines but a ‘decades-long’ battle to build deeper ‘core competencies’ (p221). True core competencies are not simply technologies that are shared across product lines, but are a domain in which the company is also superior to its competitors, and Honda again provides an example; compared to Ford, Honda’s level of competence in its chosen field of engines and powertrains is said to be substantially superior (p226); later it is argued that core competencies must be ‘competitively unique’ (p229). Moreover, there is said to be a cumulative effect whereby the very process of diffusing Honda’s competencies across product lines has enhanced them: hence Honda’s ‘overall understanding of combustion engineering has multiplied’ (p229). Finally, it is argued that having been obliged to solve certain technical problems in motorcycle engine design (maximizing power/weight and power/size ratios), Honda could then transfer this knowledge into car engine design (the implication being that this gave Honda a competitive advantage in automobiles) (p229). Hamel and Prahalad argue that core competencies are more significant strategically than the ‘core capabilities’ identified and analysed by the alternative ‘resource-based’ school of strategy championed by Boston Consulting Group’s George Stalk and his colleagues (discussed below) (p224). In a carefully worded argument that subjects the core capabilities idea to a more rigorous test than their own core competencies idea, Hamel and Prahalad argue that Honda’s capabilities in dealer network management are of only secondary importance: ‘Very few customers choose Honda over competing marques because of some unique capability on the part of Honda’s dealers’ (p225: emphasis added). Yet, although ‘few customers could express in words exactly why the Honda driving experience may be better than that experienced in, for example, a Chevrolet Lumina’ (p225: emphasis added), ‘Honda’s ability to produce some of the world’s best engines and powertrains does provide customers with highly valued benefits’ (p225: emphasis added). 23 It is the stretched targets, leveraged resources and core competencies that lie behind Honda’s competitive advantage. More concretely, and as the key link to competitive advantage, they deliver successful new products. Stretch, for example, is argued to have played a vital role in the design and development of the Honda NSX, a high-technology sports car launched in 1990. Thus in a discussion of ‘foresight’, Hamel and Prahalad argue that while a US company was slow to introduce new models because it was responding to its customers, ‘its customers were following more imaginative competitors’ (p109) such as Honda. The Honda NSX is enthusiastically presented as example of a car similar to a Ferrari but sold at a fraction of the price, an instance of how Honda has gone being ‘beyond customer led’, following the company’s own apparently unerring ability to lead its customers into new market niches; ‘one gets the feeling that Honda is more interested in outpacing competitors than bench marking them’ (p109). Indeed Hamel and Prahalad argue that there are three types of companies: 1) those that try to lead their customers where they do not want to go (‘these are companies that find the idea of being customer-led an insight’: p109); 2) companies that respond to customers; and 3) companies that lead customers to where they want to go but do not yet know it. Honda belongs to the third group. The implication is clearly that The NSX, described as a ‘world beater sports car’ ‘that could often be purchased for a lot less than a top-end Porsche 911’ is compared with the products of Porsche, a company that became too complacent in its product offerings: ‘Porsche’s brand premium wasn't always backed up by a competence-based performance advantage’. Accordingly, Porsche experienced a dramatic sales decline in the United States in the years between 1986 and 1993 (p230). Finally, a series of further examples of good strategy-making at Honda are cited. Hence Honda’s product development staff are said to ‘empathize with human needs’. Honda gets very close to the company’s customers so that they can be offered more appropriate products: Again and again, Honda has worked hard to ensure that those charged with product development possess deep insights and real empathy for the customers they are seeking to serve (p113). 24 In a discussion of the significance of ‘core product share’ as opposed to ‘market share’, Honda’s 1980s and 1990s British partner in the European automobile industry, Rover Group, is presented as an example of a company where ‘brand share’ hid dependence on Honda for ‘core product’ such as ‘key auto components such as the engine and the design of vehicle platforms’ (p241). As far as Honda’s own brand name is concerned, the company has capitalised on its core competencies by leveraging the same brand name into all the products that share them (p277-84). Indeed, so important is the Honda core competence in engines to consumers, apparently, that ‘whatever the product, the engine has always received prominence in Honda advertising’ (p287). By the mid 1990s Hamel and Prahalad’s now influential approach to strategy and concomitant references to Honda was finding its way into the management textbooks. Kotler (1994:67) used Honda in his box summary of the core competencies idea, citing Honda as a company that is ‘nurturing its major core competence, namely making engines’. Doyle (1994:29) reviewed Hamel and Prahalad’s argument, and also presented Honda as an example: ‘Honda’s car, motor-cycle, power tool and outboardmotor businesses, for example, rest upon shared design and development skills’. 1.7 Stalk, Evans and Shulman’s core capabilities Stalk, Evans and Shulman (1992) raise the case of Honda in a direct rebuttal of the Hamel and Prahalad ‘core competencies’ thesis (as presented in articles published prior to Hamel and Prahalad (1994)). Honda is their first example, and like Hamel and Prahalad they depict Honda as a tiny company after the Second World War which is now ‘challenging General Motors and Ford for dominance of the global automobile industry’. Their explanation of Honda’s success, however, differs markedly. Stalk, Evans and Shulman focus squarely on the strategic significance of particular business processes; these are the core capabilities. A box case study is devoted to Honda, to show explicitly through example how capabilities differ from core competencies. Hamel and Prahalad’s idea 25 of core competencies, they argue, focuses on the technologies and production skills that underlie a company’s product lines. However, for Stalk, Evans and Shulman this argument cannot explain Honda’s ability to move into new product areas. Instead, ‘what distinguishes Honda from its competitors is its focus on capabilities’. The authors focus on ‘one important but largely invisible capability . . . Honda’s expertise in ‘dealer management’ - its ability to train and support its dealer network with operating procedures and policies for merchandising, selling floor planning, and service management’. This capability is said to have been replicated as Honda moved into new product areas. A second capability is what the authors call ‘product realisation’, arguing that Honda has replaced the traditional sequential product development process with two separate processes: ‘planning and proving’ (i.e. research) and ‘execution’ (i.e. development) which take place in parallel. Execution has a disciplined four-year cycle of ‘product revisions’. The speed of Honda’s process for launching its Acura division is compared to the slowness of General Motors’ launch of its Saturn division. Competencies and capabilities, it is argued, are complementary and can form parts of a ‘grand unified theory’, with the former focusing on technologies and production expertise, and the latter ‘more broadly based, encompassing the entire value chain. In this respect, capabilities are visible to the customer in a way that core competencies rarely are’ (this is the argument rebutted by Hamel and Prahalad). The core capability in ‘dealer management’ is further discussed later in their argument in terms of Honda’s provision of operating procedures, policies and service management, and training of staff, to ensure a more business-like approach to sales. In a manner similar to the idea of leverage, it is argued that the same capability has been transferred and reproduced as Honda moved into new market sectors. 2. DECONSTRUCTING THE HONDA DEBATES The most striking impression when reviewing these interpretations of Honda in close succession is that several of them appear contradictory, 26 to the point that their proponents are led to fiercely debate ‘the meaning of Honda’. Pascale’s (1984) insight that the same reality can be interpreted in different ways characterises not just the particular events in which he is interested but the way Honda is used in general. What can be learned - both about Honda, and about the strategy industry - from the fact that apparently contradictory theoretical and practical lessons can be drawn from a single company? Before we examine this question in detail, it is instructive to deal with two initial questions: to what extent are the strategy studies reviewed above ‘accurate’ (true in their own right), and to what extent are they ‘representative’ (typical of Honda)? 2.1 Dubious facts and missing crises Some of the discussions of Honda do raise significant issues of empirical research. Pascale, for instance, insists that the Super Cub was inexpensive because its engine was small and lightweight, whereas overall production costs for internal combustion engines do not vary greatly by size (hence the much greater profitability for manufacturers of large cars, which can command significantly higher prices). Moreover, Sakiya reports (1987:123) that Fujisawa did in fact intend the Super Cub to be a significant competitive weapon in the United States from the outset. Quinn makes a number of errors, failing to distinguish Honda’s organisation of basic research from that of product development (see Mito, 1991), dubiously asserting that Honda’s production workers (rather than the specialist subsidiary Honda Engineering) design production equipment, and revealing a poor understanding of competition in the European car market. When he moves beyond an exaggerated paraphrasing of Pascale, Mintzberg (1996b) juxtaposes statistics which show the respective success and failure of the Japanese and British motorcycle industries before and after 1975 in order to test the success of BCG, concluding that since the British industry collapsed after 1975 BCG’s analysis was not successful, but in doing so commits the fundamental error of inferring causality purely from correlation with a 27 single variable. Abbeglen and Stalk (1985) fail to distinguish between two important periods during the second half of the 1950s, an initial period of successful cost cutting at constant volumes which preceded and indeed ultimately permitted the cost reductions due to sales and volume increases at the end of the decade (Sakiya, 1987:173-5). It is perhaps Hamel and Prahalad’s empirical grounding that leaves the most to be desired. The most questionable evidence concerns the NSX ‘super car’, of which they make much. Far from the triumph lauded by Hamel and Prahalad, the NSX, launched in 1990, proved a costly market failure for Honda, one of the company’s Japanese ‘bubble economy’ excesses. Output collapsed with the collapse of the bubble economy in 1992, with sales also declining rapidly from early peaks in the United States: output of 35 per day in 1991 declined to 5 per day by 1992 and never recovered. The favourable comparisons with Ferrari and Porsche are equally problematic. Ironically, Honda was obliged to sign a technology accord with Ferrari to try to recoup some of its NSX investment. While Porsche’s sales in the US market did dip in the years around 1990, NSX sales in the United States peak sales of 1,940 units were only half of Porsche’s lowest annual sales. Alongside NSX, there are several other examples of Honda actually ‘losing touch’ with its clientele from the mid 1980s until the mid 1990s (see Mair, 1998a), and the description of Rover’s supposed technological dependence on Honda is far from accurate; Rover actually rebuilt its technological competencies during its partnership with Honda, and largely independently (see Mair, 1998b). Can these studies be said to ‘represent’ Honda. Perhaps most immediately striking is the close focus of Pascale and Mintzberg on a single aspect of Honda strategy over a very short period in one place (recall that, moreover, BCG (1975) in fact only devotes two short passages to these events). Viewed from a greater distance, however, perhaps even more striking is the complete absence of any discussion of Honda’s strategic errors and financial crises. Throughout all seven analyses, Honda is presented as an unmitigated ‘success story’, even when the analysts come so close to the strategy failures that one could be forgiven for thinking that they have deliberately ignored them. 28 At least four significant crises at Honda - 1954, 1966, 1969, early 1990s - are worthy of analysis for their strategy implications and deserve mention here. The 1954 financial crisis was largely the outcome of precisely the high investment ‘winner’s competitive cycle’ strategy proposed by Abbeglen and Stalk, but in the absence of competitive products. The key to understanding the failure of Honda’s strategy in the early 1950s and its success a few years later lies at the level of the product offering; the 1958 Super Cub made the difference between near disaster and spectacular success (see Sakiya, 1987:81-89, 94-105). The Super Cub boom in the United States was immediately followed by a spectacular collapse of sales in 1966, with monthly sales falling from an average 20,000 to 8,000 and as low as 2,000. Fujisawa placed the blame squarely on Honda’s complacency over the Super Cub design, which had not been updated. The result was a huge excess of stocks (300,000), enormous cashflow - especially foreign currency - problems and potential bankruptcy. When recovery came in the late 1960s it was based on the larger motorcycles Honda had initially - and correctly - expected to sell, as analysed by BCG (Mito, 1990: 37-8; Sakiya, 1987: 139-40) Soon after Honda’s successful move into mass automobile production with its N360 mini car in 1967, Mr Honda banned all work on water-cooled engines as he enlisted his engineers in the search for a masterpiece powerful air-cooled engine with which to crown his career. The outcome was a niche ‘sporty’ car launched at precisely the moment, in 1969, when a mass market car was needed to capitalise on the N360 success. The 1300 sold only 115,000 units before it was withdrawn in 1975 (compared to over 1 million N360s in four years). Only after a major internal confrontation that left Mr Honda isolated were the company’s younger engineers permitted to develop water-cooled engines. Fujisawa believed that Honda had committed a major strategic error by missing the opportunity to offer a car like the water-cooled 1972 Civic three years earlier (Sakiya, 1987: 186-7). The failure of the NSX was in fact symptomatic of a deeper crisis for Honda between 1991 and 1996. Declines in unit sales were compounded by 29 a sharp rise in the yen’s value that put severe pressure on costs incurred in Japan. Profits declined to 0.6 percent of turnover in 1994. While in part the crisis was due to world-wide recession, long-standing underlying strategic problems related to Honda’s product strategy now came into the open. Notwithstanding the company’s carefully fostered product technology image, commercial success had always rested on well designed and well made mass-market products: Super Cub (1958) Civic (1972), and Accord (1976) in particular. The financial crisis brought the relationship between mass products and technology-related niche products into sharp focus since sales of the mass products were no longer sufficient to financially support advanced research projects and some technologically advanced and niche projects were now losing considerable sums of money. Moreover, defining the company’s growth niches in terms of sporty and luxury cars, Honda R&D had steadfastly refused to enter the ever expanding market for recreational vehicles and mini vans in North America. The resulting financial crisis led to abrupt strategic reversals in product development, including the cancellation of sporty car projects and a belated move into the recreational vehicle and mini van segments, under the revealing banner ‘market in, product out’ (see Mair, 1998a). 2.2 Reductionist epistemologies As types of methodological errors, facts that are dubious and problems that are ignored are well known empirical issues. In principle they could be rectified relatively easily simply by adopting a more rigorous inductivist method. Why this has not occurred raises the equally well known problem of the theory-laden nature of observation. In brief, it would appear that the analysts of the strategy industry see only what they want to see. This brings us directly to an epistemological issue that goes to the heart of the ways-of-thinking, or implicit philosophies, that dominate management thinking in the West (or certainly the ‘Anglo-Saxon’ countries2), and which appears to play an important role in thinking on strategic management. Simply put, this is the acute one-sided reductionism revealed in each of the seven strategy industry analyses of 30 Honda, in which complex polarities - planned and emergent strategies, competencies and capabilities, etc. - are persistently reduced to only one of their poles in such a way that theory-building degenerates into a brute struggle over which pole is the more significant. This particular theory-building process is intimately linked to the factual poverty and missing crises, because to make it ‘succeed’ (i.e. for one theory to ‘win’), data has to be ‘aligned’ to fit a theory (i.e. one end of a pole); accordingly, data that cannot be made to align or is too inconvenient has to be ignored or fudged. If the strategy industry authors are aware that they are doing this consciously or subconsciously, presumably they feel justified by the righteousness of their theoretical approach. Alternatively, as I suggest below, the driving factor in this theory-building process may be found in the sociology of knowledge, specifically the relationship between the strategy industry and its market. In the normal course of events, one-sided reductionism may not cause particular theoretical and practical problems in those Anglo-Saxon countries where company managers share a similar implicit metatheoretical approach (because it is at least aligned to their practice). Explanations couched in these terms may therefore be more acceptable and easily digested (because of their familiarity). However, this approach is strikingly inappropriate in the case of Honda, a company where what we might call ‘two-sidedness’ or non-dualist thinking is deeply ingrained (Mair, 1996; Mito, 1990; Nonaka and Takeuchi, 1995). A meta-theoretical approach which insists upon one-sided reductionism in theory-building will never offer convincing explanations of a company that not only does not accept it in its management thinking, but indeed positively strives against it (see below). What is the evidence for the strong claims being made here concerning one-sided reductionist approaches in the analyses of all the strategy thinkers? And how might Honda strategy be better understood in a non- reductionist, non-dualist manner. 31 BCG Goold (1996) accepts that the original BCG report focused entirely on structural economic factors to the neglect of the strategy formation process. Pascale Pascale does point to a surprise learning experience in which there were preconceived expectations and yet Honda was ready for (with four motorcycle models), and seemed relatively quick to react to, market signals that were different (size of machine, appropriateness of a different distribution network). He does suggest a logic to the decision-making that accompanied this strategic change and an appropriate organisational structure for this kind of strategic learning. Yet Pascale fails to pursue the fact that, with the Super Cub, Honda was already practised at making the small machines in large numbers for one market, had already established a new type of distribution system - in Japan - through retail bicycle shops which bypassed the traditional motorcycle distributors (for the predecessor Cub model in 1952), and had already deliberately opened up a new market niche for small motorcycles - in Japan - based on practicality and ease of use while competitors focused on larger machines (these two points are discussed by Sakiya, 1987: 84, 117). From this perspective, the lesson for Honda was not that the United States was a quite novel market which required a radically innovative strategy, but that the United States was far more like Japan than expected. Less learning was required than Pascale implies. Moreover, Pascale fails to place Honda’s entry into the United States in the wider context of Honda’s overall international strategy including its general strategy towards the US market. An element of company philosophy was already to ‘test its products against the best’ (hence 32 Honda’s early racing activities and very early exports). Fujisawa deliberate chose to focus the meagre foreign exchange available to Honda on testing the US market, imposing this strategy on the company and overriding the strategy proposed (bottom-up) by his managers of concentrating international efforts on European and Asian markets (Sakiya, 1987). To contextualize Pascale’s argument in this way radically reduces the significance of ‘serendipity’ and bottom-up decision-making’, confining them to quite specific, and not quite so remarkable, aspects of the US strategy. While Pascale appears to present ‘both sides’ of the planning/learning polarity he clearly favours the implications of his own investigative work. He never seriously attempts to show that both BCG’s results and his own findings may be correct (his weak attempt is discussed below) because they are only contradictory to a certain reductionist mindset and not in practice. At first glance Pascale appears to have opened up thinking (hence his popularity in strategy pedagogy), yet he has simultaneously closed off any further steps. This is all the more remarkable given the theoretical approach of Pascale’s 1990 book (which might be summarised as ‘how companies should transcend contending opposites in order to progress’). Indeed Pascale clearly states that Honda utilises paradox and contradiction as an explicit management tool (1990:27). When it comes to the events in the United States circa 1960, however, this interpretation is for some reason not applied. Quinn Quinn’s interpretation is even more thoroughly one-sided, pursuing the Pascale ‘emergent strategy’ theoretical model to its extreme by applying it to all of Honda’s activities and processes. Having adopted this theoretical approach with a vengeance, it comes as no surprise to find that Quinn presents a broad and shallow survey, what is in effect a pastiche tied together neither by theory nor by empirical reality. In Quinn’s portrayal, there are no operations systems and organisational 33 structures at Honda, with the only logic holding the analysis together being individuals and their innovations. This is an example of a not-untypical ‘business school’ type case study constructed from fragments of knowledge based on books (often popular), magazine and newspaper clippings, which then asks students to make strategic decisions on this basis. Any real strategy analyst would surely need quite different information available. In particular he or she would be aware of the vital role played by what I have called the ‘Honda Production System’ in operations management and ‘Fujisawaism’ in organisational form (see Mair, 1998a). These are the management systems which are necessary to permit the individualism and innovation to function, and in the case of Fujisawaism were deliberately designed to permit it to flower. In the absence of a deeper and more rounded understanding of such management systems, no student or real strategist, whether competitor or benchmarker of Honda, could have confidence in Quinn’s one-dimensional study. Mintzberg Given Mintzberg’s well rehearsed arguments (e.g. Mintzberg, 1991) that strategy must ‘walk on two legs’ (structure and strategy) and that planning and learning need to take place in tandem, his insistence on confining ‘the facts’ about Honda to a narrow one-dimensional perspective, essentially an exaggeration of Pascale’s basic argument, requires some explanation. True, as Goold (1992) observes, it may be necessary for Mintzberg to stress the ‘learning’ aspect of strategy to counter the continued dominance of the ‘analytical’ strategists. Yet Honda offers a particularly poor exemplar; it might have been better to take the next step and present Honda precisely as an example of the advantages of ‘walking on two legs’. Mintzberg appears to have missed a golden opportunity to do just this, still preferring in 1996 to fight a 1975 battle. Like Pascale, then, Mintzberg appears to favour of non- dualist thinking in principle but not in the case of Honda’s entry into the US motorcycle market. 34 Abbeglen and Stalk and Hout Remarkable by its absence in the first of Abbeglen and Stalk’s case studies, that of Honda’s growth during the late 1950s, is the vital product, the Super Cub, which made the difference compared with the mid 1950s crisis. Accordingly, Honda’s success is explained purely in terms of a strategy which at that time (mid 1980s) played a significant role in BCG’s consultancy armoury, the ‘winner’s competitive cycle’, or growth through debt leading to low costs and profits through economics of scale. In fact, it appears to have been a combination of four factors - strategic operations management to cut costs, product design, product technology, and volume (and not the last alone, despite Abbeglen and Stalk) - that lay behind Honda’s rapid domestic growth. Stalk and Hout already downplay part of the story of the H-Y wars previously told by Abbeglen and Stalk, and this tendency towards onesided reductionist explanation is taken further by other management writers who have adopted it. The story of the H-Y wars has been picked up and summarised by several strategy consultants, including HarveyJones (1988:14), Heller (1989), Heller and Carling (1995) and Peters (1992:60). In each case the focus is solely in the introduction of a wave of new product introductions rather than price cutting, advertising, and inventory build-up. academic texts. It has also found its way into Thus Whittington (1993:69-70) tells the story as an example of how the Japanese do not necessarily make strategic investment decisions based on ‘rational’ Western accounting criteria, and Doyle (1994:192-3), citing Stalk and Hout (1990), argues, contrary to the original case study, that Honda is a ‘fast-track innovator’ that did not cut its prices to compete with Yamaha but instead introduced more fashionable and sophisticated products. Hamel and Prahalad While the concepts of stretch, leverage and core competencies may be intuitively attractive, Hamel and Prahalad present no evidence that 35 Honda in fact owes its success to their adoption. How much explanatory weight can Honda’s ‘organic’ corporate diversification around internal combustion engine technologies bear? Hamel and Prahalad offer no causal linkages, only a correlation between Honda’s focused form of corporate diversification and its commercial success. Yet the fact is that there has only been one strategically significant technology diffusion at Honda, the company's move into the car industry over thirty five years ago. While Honda does make an impressive list of other products powered by internal combustion engines, only cars and motorcycles have ever accounted for more than a very small proportion of sales, with all other products taken together consistently accounting for less than ten percent of corporate turnover. Moreover, the core competency argument, together with Honda’s supposed closeness to its customers, must both be placed in considerable doubt by the company’s conspicuous failure to ‘leverage its competencies’ into the growing sports-utility vehicles (SUVs) and mini vans sub-sectors until the mid 1990s despite repeated calls for such products from the American front-line dealership network from the mid 1980s onwards. Honda was similarly stubborn in delaying for years the introduction of six-cylinder engines in the American Honda Accord and diesel engines in its cars sold in Europe. Instead, Honda continued to define its product narrowly in terms of ‘sporty’ vehicles, invested heavily in the NSX (see above), and insisted that the technology of the four cylinder engine was quite adequate whereas the diesel engine was an outdated technology. Analysis of Honda’s actual strategy from the mid 1980s to the mid 1990s reveals a narrow self-definition and a technological stubbornness the result of which was that the company was unable to respond to the ‘less knowledgeable’ consumer. Neither can conclusions be drawn about the value of their theory from the empirical comparisons that Hamel and Prahalad offer between Honda and other companies. If the goal is to show the power of core competencies, it is not very fruitful to compare Honda to Ford and General Motors, companies which differed from Honda not only in their product offerings but also in most of their business processes (core capabilities) (see Womack et al, 1990), invalidating any inference that 36 core competencies lay behind Honda’s success in North America. An appropriately stiff test (following Karl Popper) of Honda’s strategy would be to observe how Honda has fared compared with Toyota in the Japanese domestic market. If Hamel and Prahalad were correct in their thesis about the best way for small companies to break into established markets, Honda ought to have made significant headway against Toyota, a company well known to be a very process - rather than product - driven company (Shimizu, 1998). The fact is that Honda has never succeeded in occupying more than a minor place in its domestic vehicle market, with a five-ten percent share, compared to Toyota’s dominating thirty percent share (Table 1). Table 1: Toyota and Honda domestic market shares Year Four wheel vehicle market share in Japan Toyota (%) Honda (%) 1970 27 9 1975 33 6 1980 30 5 1985 30 8 1990 32 9 It is also difficult to see how Honda’s engines - good as they may be pass Hamel and Prahalad’s test for core competence status: ‘competitive uniqueness’. Other Japanese companies, notably the globally less successful Mazda and Mitsubishi, have pursued strategies in part reliant on differentiation rooted in advanced engine technologies to distinguish them from rivals. Moreover, if unique core competencies can play such a powerful role, surely other Japanese car makers which adopted similar product technology strategies would have experienced greater success than they have in fact: Subaru with its four wheel drive, Mazda with its rotary engine. And what of the companies which have also diversified by sharing core competencies across product divisions: Isuzu with its four wheel/off-road competence and its dual truck and car markets, Suzuki with its motorcycles and cars? Why are these companies not more 37 successful? What explains Toyota’s continued dominance of the Japanese market, and what explains the relative commercial failure of other Japanese companies, in Japan and globally, which could equally be described in terms similar to those used by Hamel and Prahalad for Honda? In short, technological product focus and diversification based on core competencies do not seem to be deciding competitive factors. Indeed Hamel and Prahalad surely go too far in arguing that dealer management does not play the significant role argued by Stalk, Evans and Shulman. There is abundant evidence of the degree to which Honda values the management of its distribution system. A great deal of Takeo Fujisawa’s managerial effort during the 1950s and 1960s was spent gaining control of domestic distribution networks, and a major effort went into preparing them for the introduction of cars during the late 1960s (Sakiya, 1987). This was particularly important for a company with no keiretsu linkages and which believed that the complex technical nature of the product, with its significant requirements for servicing, demanded a direct line to the customers (rather than one mediated by the traditionally dominant Japanese trading companies). Several distribution practices first developed in Japan were transferred to North America, despite different formal relationships with dealers there (see Shook, 1988). It is also quite clear that, as argued by Stalk, Evans and Shulman, the organisation of the product development process (as opposed to the basic technologies research that is the implied emphasis of Hamel and Prahalad) is very significant indeed for Honda (Mito, 1990; Nonaka, 1991; Nonaka and Takeuchi, 1995). Moreover, Honda’s product-technology strategy appears to have been associated with many failures as well as successes. Indeed a hypothesis that suggests itself for further testing is that Honda’s global commercial success has been achieved despite its engine technology/sporty car focus rather than because of it. Honda’s strengths in operations management and organisation have received virtually no attention in the Anglo-Saxon world, likewise the company’s reliance for global growth on the Civic and Accord mass market cars (and before them the mass market Super Cub), which together continue to account for three quarters of unit sales. While the company’s self- 38 image stresses Mr Honda’s achievements and technological drive, the new motto for the mid 1990s at Honda - ‘market in, product out’ - recognised how commercially damaging the product technology focus so praised by Hamel and Prahalad had become. Finally, it is noteworthy that, like Pascale (1990) and Mintzberg (1991), Hamel and Prahalad too (1994; final chapter) claim that their overall approach to strategic management has been informed by a nonreductionist approach which combines dichotomous polar concepts. Yet, like Pascale and Mintzberg, this approach is absent from their interpretation of Honda. Stalk, Evans and Shulman Stalk, Evans and Shulman reasonably argue that Hamel and Prahalad’s product-oriented competencies approach is one-dimensional, and argue that there are important operational and organisational business processes which help explain Honda’s success but which have no place in Hamel and Prahalad’s theory. Their proposal that the two approaches, core competencies and core capabilities, may be complementary to each other, appears to be a positive, if limited, move in the right direction. However, this is not without irony given BCG’s own reductionist tendency to try to explain and re-explain Honda’s competitive successes in terms of BCG’s currently dominant and usually relatively simple managerial paradigm. 2.3 Learning from Honda: pushing forward the frontier From one conventional perspective, alternative approaches to strategy process and strategy content can be ‘tested’ against successful companies. From this perspective, further study of Honda might be called for, to help make a final decision about whether the case of Honda supports planning or learning, positioning/industry analysis or internal resources, competencies or capabilities. This is an 39 implication of Rumelt’s (1996) analysis of the Honda debates; more research is needed. An alternative conventional perspective would be to test out a contingency approach, to hypothesize that perhaps Honda adopts different strategy processes and types of strategies as a function of different decision-making environments. From this perspective, each of the polar theoretical explanations might be valid at particular times and places, but it would be an error to generalise from particular incidents and situations to the company and its development at large. This resolution is partly attempted in Mintzberg's (1996b) and Pascale’s (1996) suggestion that learning and planning alternate over time, with learning coming first. An unusual alternative to these conventional approaches is to hypothesize that two apparently polar opposite and contradictory explanations can be true simultaneously. A number of observers of the strategy debates over Honda have in fact taken initial, if halting, steps in this direction. Thus Kay (1993a; 1993b) contrasts BCG and Pascale and suggests that neither may be correct: there was neither a Japanese master plan nor pure luck and idiosyncrasy. Likewise Silbiger (1994:340-1) reviews the debate between BCG and Pascale and concludes that both have some truth despite the inherent attractiveness of the Pascale thesis. Curiously, direct participants in the debate have made similar noises. Pascale (1990), Mintzberg (1991), and Hamel and Prahalad (1994) all claim that their meta-theoretical approach to strategic management is based on the resolution of contending opposite concepts. Moreover, Stalk et al (1992) and Goold (1992; 1996) state strongly that the resolution of the Honda debates lies in the combination or synthesis of opposing views. Yet none makes a serious effort to follow this approach through for the case of Honda. Why does the debate appear to have reached such an impasse? The Strategic Management Journal/California Management Review debate between Goold and Mintzberg well illustrates suggests just how difficult the resolution of these polarised debates can become. Each states clearly that some kind of combination of learning and planning will lead to the 40 better strategic management. Yet each only appears to view this combination from a viewpoint whereby their own initial starting point retains the dominant position. Hence while Goold admits that learning has an important role to play, he restricts his definition of learning to precisely the type of formalised analysis of other companies’ strategies (‘we can use both planning and learning from others in selecting the strategies to try’ (Goold, 1992)) that Mintzberg criticises. Similarly, while Mintzberg (1991) appears to admit an important role for planning, he not only delimits its role and concludes with a score of ‘learning 1, planning 0’, but argues that learning must always precede formal planning (‘ . . . the critical period of emergent learning that had to inform the deliberate planning process. In other words, strategy had to be conceived informally before it could be programmed formally’ (Mintzberg, 1992)). Pascale makes a similar argument to Mintzberg’s, that normalisation of strategies best comes post hoc, to rationalise ‘the strategic nugget that has been stumbled upon’ (1996). Only after the fact, he claims, did Honda come to see its small engine dominance as a distinctive competence to be exploited purposively in the car and motorcycle industries. Later in the paper we suggest how the very functioning of the strategy industry itself may hinder progress beyond such impasses. First we need to begin to clarify how advancing our knowledge of, and learning from, Honda strategy requires us to match the complexity of our strategy theory with the complexity of reality at Honda, rather than proposing such simple theories that reality is reduced out of all recognition. Table 2 illustrates a series of Honda ‘solutions’ to some of the polar puzzles of management. found in Mair (1996). (A fuller argument and further examples can be A similar analysis with respect to Honda’s multinational organisational form, set against the context of the literature which simplifies the model of the new Japanese multinational to the ‘global company’ or ‘stateless corporation’, can be found in Mair (1997)). 41 Table 2: How Honda Has Resolved Management Dichotomies {PRIVATE }Domain Dichotomy Puzzle solution Product design model renewal vs facelift rolling four year model replacement cycle with two year lagged cycle for mechanical parts renewal (Mair, 1996) Organisation top down vs bottom up ‘middle-up-down’ management (Nonaka, 1988; Nonaka and Takeuchi, 1995) Organisation individual vs group on factory floor stress on individualism in groupist Japan, stress on teamwork in individualist USA (Mair, 1994a; Shook, 1988) Organisation individual vs group at senior management level alternating emphases with reforms by each new company president (Mair, 1996) Operations efficiency vs humane work ‘free-flow’ production line which stops at each work station (Mito, 1990) Operations product volume vs variety small batch (60) production system Strategy and human resource management product flexibility vs long-term employment ‘flexifactory’, where products are changed but factories and workforces are retained (Mair, 1994b) Supplier relations single sourcing vs multiple sourcing ‘parallel sourcing’ (Richardson, 1993; described for Honda in Mair (1994a) Spatial organisation of suppliers geographical proximity vs dispersal ‘just-in-time region’ (Mair, Florida and Kenney, 1988) Global organisation economies of scale vs local responsiveness ‘strategic localisation’ (Mair, 1997) The weak hypothesis is therefore that Honda seeks ways to make apparent contradictory polarities in strategic management concepts mutually compatible. The strong hypothesis is that Honda has found ways to make the polarities mutually supportive, so that they are in fact positively rather than negatively correlated. This suggests that the ‘normal science’ of an appropriate ‘paradigm’ (Thomas Kuhn) under which the strategy industry could learn from Honda would be to analyse and reconstruct how Honda does not choose between 42 the polar positions of the dichotomies of strategic management formulation/implementation, strategy/structure, planning/learning, positioning/resource based, capabilities/competencies - but synthesizes them in its strategy making. This normal science would of course include analysis of the problems that arise when Honda fails to implement such an approach, notably the apparent over-domination of the product-led aspect of strategy as revealed by the crisis of the 1990s. An initial pointer along these lines, testing the weak hypothesis, is given in Table 3, which takes a more rounded look back at the events of forty years ago when Honda achieved its dominance of the Japanese motorcycle industry and began in earnest the process of internationalising its markets. Table 3: Honda strategy in the motorcycle industry, 1950-1965: supposedly contradictory theories are not incompatible 1) Mr Fujisawa followed unusual and risky financial strategies in the early 1950s in a bid to expand the company rapidly (Sakiya, 1987: 81-89) (stretch). At the same time, in 1952 he created new types of distribution channels for small motorcycles (the Cub) which bypassed traditional motorcycle distributors in favour of retail bicycle outlets (importance of distribution management). 2) In 1954-5 the company experienced a significant financial crisis when product technology and design problems depressed sales (Sakiya, 1987: 94-105) (importance of product strategy for success of high-risk stretch growth strategy). 3) Accordingly, in 1956 Mr Fujisawa asked Mr Honda to design a potential mass-market small motorcycle even as other Japanese competitors focused on larger machines during a period of domestic economic growth: the 1958 Super Cub would be the outcome (Sakiya, 1987: 117) (closeness to customer, industry analysis, new market niches sought). 4) Meanwhile, Honda focused on the internal reorganisation of its production system to cut costs without raising production volumes. Honda’s main rivals were content to reap profits on the basis of high volumes from the mid 1950s economic boom in Japan without cutting production costs. During the recession which started in 1957, Honda was able to reduce its prices to undercut competitors, and on this basis to increase market share (Sakiya, 1987). 5) Following the successful domestic market launch of the Super Cub in 1958, Mr Fujisawa embarked on the high-investment manufacturing strategy described by Abbeglen and Stalk (1985), building the Suzuka motorcycle plant, with its huge capacity of 360,000 motorcycles per year, and which opened in 1960. As the small motorcycle sector grew and as competitors collapsed, suddenly Honda controlled 80 percent of the Japanese motorcycle market. Mr Fujisawa’s goal was not simply to dominate the Japanese market but 43 in so doing to reduce costs to internationally competitive levels (Sakiya, 1987: 173, 175-6) (business processes, industry analysis, leverage). 6) Notwithstanding the advice of subordinate managers to focus initial internationalisation efforts on countries with high potential motorcycle usage, particularly in South East Asia, Mr Fujisawa insisted that meagre foreign exchange allowances be focused on the US market despite low existing motorcycle demand there, on the grounds of a version of trickle-down theory: success in the United States would be followed by success elsewhere (Sakiya, 1987: 1201). Fujisawa also believed that the Super Cub could be successful in the United States (Sakiya, 1987:123) (stretch overrides less ambitious strategy suggested by one type of industry analysis). 7) Because its mission was to explore the market, the Honda team in California was furnished with 25 percent each of four models, including Super Cub. Not unreasonably, they initially sought to sell the company’s larger motorcycles, since larger machines dominated the US motorcycle market (Pascale, 1984) (willingness to learn in unknown market, planned strategy based on industry analysis of US market). 8) The California team was quick to react when the small motorcycles (of which they had imported as many as of the largest motorcycles) proved popular, and the company was quick to react by ramping up production in Japan to meet US demand (Pascale, 1984) (learning, business (production) process). 9) Hence, from 1960 onwards, Honda discovered that there was an unanticipated market niche in the United States for the Super Cub quite similar to the one it had created in Japan from 1958, and applied the same distribution principles (non-conventional retail distribution networks) in the United States as had been applied in Japan (Pascale, 1994; Sakiya, 1987; see also Shook, 1988) (leveraging home market strength, transfer of 'core capabilities'). 10) The early 1960s small motorcycle boom in the United States market did not last, and collapsed in 1966 (Mito, 1990; Sakiya, 1987: 139-40). From the late 1960s onwards, it was the strategy based on larger motorcycles which proved successful (BCG, 1975), and by 1979 it was the larger motorcycles that Honda was building at its new Ohio factory (industry analysis). The weak hypothesis, that what might seem to be contradictory strategy concepts are not necessarily contradictory, can therefore be accepted in this case. Similar analytical work could be pursued to piece together Honda’s general strategy process (strategy and structure, planning and learning), as well as strategy content (positioning/industry analysis and resource-based, competencies and capabilities). Far more could be 44 learned from Honda than has come out of twenty years of polarised debate if the strategy industry could achieve such a change of perspective in a focused and sustained manner. CONCLUSIONS: IMPLICATIONS FOR THE STRATEGY INDUSTRY What conclusions should be drawn from the analysis of how the strategy industry has utilised the Honda case so far? At the simplest level, Honda’s strategic management may have been widely misinterpreted, and hence understandings of Honda among the many practising managers who may have been influenced by the outputs of the strategy industry are likely to be incorrect. Accordingly, any practical implications they have drawn from their knowledge may not in fact have a basis in Honda’s actual strategy. At root, these problems are caused by a basic meta-theoretical approach widespread in Anglo-Saxon management thinking and apparently deeply rooted in the theory-building process of the strategy industry. It remains to be asked quite why the strategy industry has consistently adopted the one-sided reductionist approach even in the case of a company for which the evidence suggests it to be particularly inappropriate.3 Recent critical analyses of thinking and theory in the strategy industry have begun to draw attention to some of the inner workings of the industry, particular the way ideas are developed in quick succession as ‘fads’ and ‘fashions’ (Abrahamson, 1996; Barry and Elmes, 1997; Kieser, 1997). These and other analyses can be utilised to formulate hypotheses about the strategy industry consistent with the strategy industry’s utilisation of the Honda case, although to test them properly is beyond the scope of this paper. Hypothesis 1: Buyer and seller collude around ‘strategy-bites’ At least some participants in the strategy industry seek relatively simple explanatory concepts to sell to their clients (clients are 45 broadly defined, including not only organisation leaders but students and practising managers). Concepts are simple not only so that they may more easily be implemented, but equally so that they can be more easily digested as ‘strategy-bites’ too. Here there is collusion between the strategy thinker and the clients to simplify and reduce explanations. Hypothesis 2: Product differentiation leads to polar opposite ‘strategybites’ The functioning of the competitive process in the strategy industry, whether the direct competition of the strategy consultancy companies, or the less direct competition of the academics and other ‘gurus’, requires continuous product innovation and differentiation. In the endeavour to differentiate strategy concepts as far as possible in a manner consistent with Hypothesis 1, participants may opt to develop ‘revolutionary’ ideas which overturn the existing ruling ideas yet are equally simple. It makes strong sense to propose simple polar opposite 4 approaches. Hypothesis 3: Functional chimneys exacerbate polarisation This process is driven further by functional differentiation and ‘chimneys’ within the industry: in particular the disciplinary boundary between those trained in economics and those trained in sociology in which explanations are delimited and reduced by disciplinary selfdefinition, and probably by the perhaps more porous, but still fractious, boundary between consultancy and academia. Hypothesis 4: Mysticism becomes an alternative to simplicity An alternative to simple concepts is a certain mysticism, which suggests complexity while not really helping the client deal with it (thus preserving a role for the strategy analyst as ‘witch doctor’). Mysticism may be communicated by suggesting the apparent ability of certain great corporate leaders to divine the future, or it may be the mysticism of paradox and uncertainty which both recognises complexity and proposes the need for synthesis, but takes the analysis no further. 46 Hypothesis 5: Company exemplars communicate and legitimate the strategybites ‘Success story’ case studies are a vital vehicle for the communication of the strategy concepts. They perform two functions. First, they embody and therefore help communicate abstract new ideas, functioning as ‘exemplars’, as argued by Thomas Kuhn. Second, they legitimate the new ideas because the latter are said to, or more often implied to, explain the success. Well known successful companies are very frequently used by the strategy industry to legitimate and communicate different management theories. As Silbiger (1994:340-1) notes, ‘history is often rewritten to suit the theories of strategic planners’ (Silbiger, 1994:340-1). The implication of these five hypotheses is that, far from being close to management practice on the input side, the strategy industry may be too close to management practice (the practice of its clients) on the output side: too often tailoring the products for the customer’s existing situation and (intellectual/organisational) capabilities rather than attempting to improve the latter. The hypotheses could be further tested by comparing in-depth research analysis with the way other company case studies are used in the management literature. The explanations for the way Honda has been analysed by the strategy industry suggested by the above hypotheses lie within the field of the sociology of knowledge (Barry Barnes), or rather the business of knowledge; in this case the modus operendi of the strategy industry. More broadly, there are more complex linkages to explore in terms of a link between the strategy industry’s propensity towards one-sided reductionism and wider social and philosophical thinking in the West around the philosophical concept of ‘dualism’: the general tendency to think in terms of polar categorical opposites, such as mind/body (the Cartesian dualism), man/nature, and the like. Recognising but setting aside these broader contextual issues, how might the theory-building process in the strategy industry be improved upon? We end this paper by briefly examining the approach taken to business 47 case studies, the epistemology of this ‘pragmatic discipline’, and possible change drivers. 1) Case studies This paper raises important questions about the use of business case studies in the management literature. An intensive and reflective inquiry into the role of the case study in management research, teaching and theorising seems long overdue. Certainly the predominant management, particularly business school, approach could usefully be compared with more traditional social science case methods, both in terms of quality of empirical material and the way that conclusions are suggested by the way the case studies are ‘set up’ and presented, sometimes by simply shearing away all apparently extraneous materials leaving inevitably one-dimensional implications to be drawn (the problem of ‘theory-laden’ empirical research). An initial list of possible issues might be as follows. One line of inquiry could be the tendency to focus on ‘success stories’ which tend to legitimate the theory supposedly exemplified; within this is the collusion - or at least silence - of the ‘successful company’ itself even when in some cases its employees find it hard to recognised the picture painted. A second line could be the repetition by academic management theorists and educators of the practice-led work of the consultancy wing of the strategy industry as if it were based on reliable case study research, whereas as we have suggested it may have other motivations. Third, there is the tendency to generalise what is apparently known about one aspect of a company or one event in a company’s history to the whole company, which is seen to encapsulate a certain view of or approach to strategic management. Fourth, there are of course various types of business case studies, from the full complex book through to a few lines of text via the dossiers or short documents which appear to dominate business school pedagogy, and the variety of these needs to be recognised. Fifth, there is the question of how much in-depth research actually takes place and how much is accepted on faith by researchers grateful for any type of access to organisational 48 leaders. This in turn is related to the highly problematic issue in strategic management research of anonymity; it would be interesting to compare how critical is the stance taken by case studies of named companies (many of which sometimes appear largely based on a collection of press clippings) versus anonymous cases. Finally, other social science disciplines, such as anthropology, sociology and geography, for instance, have conducted significant debates on the relationship between the particular case study, the generality of knowledge, and the transferability of lessons from case studies to other contexts, and management could surely learn from them rather than reinvent this wheel. What do we really know about GE, or 3M, or other companies which, as Rumelt (1996) argues (his list includes General Motors, IBM, GE and Honda), are implied to ‘represent’ particular management methods? What might result from in-depth case studies of other widely accepted ‘famous’ business examples? 2) Broader epistemology of management studies To broaden the points just made regarding business case studies, there are implications for the epistemology of management studies as a discipline. Ever the pragmatic discipline, management sorely needs to promote studies of its own sociology of knowledge and philosophy (epistemology and ontology), as a means to provide a certain rigour to keep debates on track. Again, drawing more closely upon the work that has been undertaken in other disciplines would make this task less onerous. On the other hand, the particular closeness of management to organisation practice raises specific issues related to the validation of knowledge in management; is it ‘scientifically tested’, or does ‘sales volume make right’? Here we return to the ‘business of knowledge’. It does seem possible that there is a dangerous tendency within this discipline to sacrifice proximity to practice as an input to the knowledge-creating process in favour of proximity to practice as an output of knowledge, as suggested by the five hypotheses developed above. But is the sacrifice of empirical depth as necessary as some 49 participants appear to believe for them to operate close to practice on the output side, with its attendant focus on success stories, reductionism, one-sidedness and mysticism? Has management theory therefore been sullied by proximity to organisation practice, as some of management’s academic critics might believe? In fact, far from theoretical problems being caused by closeness to organisation practice on the input side, it may be the case that the strategy industry has strayed too far away from practice and developed its theories to suit the output side: the business of knowledge again. The strategy industry may need to draw closer to organisation practice in the sense that practitioners need to be more questioning and critical of the ideas presented to them and apparently derived or legitimated in the business case study. 3) Potential change drivers Could consumer pressure, from client organisations, students and practising managers be the answer? The overall reputation of the strategy industry, closely related to the value of the ‘management guru’, is currently under concerted attack. Will strategy industry members begin to get their houses in order in response? Certainly, consumers of the industry’s ‘service’, whether fellow participants, students, managers, companies, need to be alert and sceptical in accepting ideas, subjecting them and their champions to rigorous critique. Only through informed consumer pressure - and a break away from the conditions validating the five hypotheses introduced above will the strategy industry improve the quality of the types of products we have analysed in this paper. The paper has pointed to two key areas where improvement appears to be vital: the business case study, and the nature of debate as a means to make progress in theoretical knowledge. 50 REFERENCES Abrahamson, Eric (1996) ‘Management fashion’, Academy of Management Review 21.1: 254-285. Abbeglen, James C. and George Stalk Jr. (1985) Kaisha, The Japanese Corporation. New York: Basic Books. Barry, David and Elmes, Michael (1997) ‘Strategy retold: toward a narrative view of strategic discourse’ Academy of Management Review 22.2: 429-452. Boston Consulting Group (1975) Strategy Alternatives for the British Motorcycle Industry. London: HMSO. de Wit, Bob and Meyer, Ron (1994) Strategy: Process, Content, Context. West Publishing. Doyle, Peter (1994) Marketing Management and Strategy. Prentice Hall. New York: Floyd, Stephen W. and Wooldridge, Bill (1996) The Strategic Middle Manager. San Francisco: Josey Bass. Goold, Michael (1996) ‘Learning, planning, and strategy: extra time’ California Management Review 38.4:100-2. Goold, Michael (1992) ‘Design, learning and planning: a further observation on the design school debate’. Strategic Management Journal 13:169-72. Hamel, Gary and C.K. Prahalad (1994) Competing for the Future. Harvard Business School Press. Boston: Harvey-Jones, John (1988) Making it Happen: Reflections on Leadership. London: Harper Collins. Heller, Robert (1989) The Decision-Makers: The People and Methods of Great Achievement. Sevenoaks, Kent: Hodder and Stoughton. Heller, Robert, and Carling, Will (1995) The Way to Win. Little, Brown Co. London: Kay, John (1993a) ‘The structure of strategy’ Business Strategy Review 4.2:17-37. Kay, John, (1993b) Foundations of Corporate Success: How Business Strategies Add Value. Oxford: Oxford University Press. Kieser, Alfred (1997) ‘Rhetoric and myth in management fashion’ Organisation 4.1:49-74. Kotler, Philip (1994) Marketing Management: Analysis, Planning, Implementation and Control. Englewood Cliffs, NJ: Prentice Hall. edition. 8th Mair, Andrew (1998b) ‘From British Leyland Motor Corporation to Rover Group: the search for a viable British model’. Forthcoming in M. Freyssenet, A. Mair, K. Shimizu and G. Volpato (Eds.), One Best Way? 51 Trajectories and industrial models of the world’s automobile producers, 1970-1995. Oxford: Oxford University Press. Mair, Andrew (1998a) 'The Honda Motor Company, 1967-1995: Globalization of an innovative mass production model’. Forthcoming in M. Freyssenet, A. Mair, K. Shimizu and G. Volpato (Eds.), One Best Way? Trajectories and industrial models of the world’s automobile producers, 1970-1995. Oxford: Oxford University Press. Mair, Andrew (1997) ‘Strategic localisation: the myth of the postnational enterprise’. In K.R. Cox (ed.). Spaces of Globalization: Reasserting the Power of the Local. New York: Guilford, pp. 64-88. Mair, Andrew (1996) ‘Honda Motors: a paradoxical approach to growth’. In C. Baden-Fuller and M. Pitt (eds.) Strategic Innovation: An International Casebook. London: Routledge, pp.435-61. Mair, Andrew (1994b) ‘Honda’s global flexifactory network’, International Journal of Operations and Production Management, 14.3:623. Mair, Andrew (1994a) Honda’s Global Local Corporation. Macmillan. Houndmills: Mair, Andrew, Richard Florida and Martin Kenney (1988) ‘The new geography of automobile production; Japanese automobile transplants in North America’, Economic Geography, 64: 352-373. Mintzberg, Henry (1996b) ‘Reply to Michael Goold’ California Management Review 38.4:96-9. Mintzberg, Henry (1996a) ‘Introduction: CMR Forum: The “Honda Effect” revisited’ California Management Review 38.4:78-79. Mintzberg, Henry (1991) ‘Learning 1, Planning 0: Reply to Igor Ansoff’ Strategic Management Journal 12:464-466. Mintzberg, Henry (1989) Mintzberg on Management. Press. New York: The Free Mintzberg, Henry (1987) ‘Crafting strategy’ Harvard Business Review July-August, pp. 66-75. Mintzberg, Henry, and Quinn, James Brian (1991) The Strategy Concepts, Contexts, Cases. Prentice Hall International (2nd Mintzberg, Henry, and Quinn, James Brian (1996) The Strategy Concepts, Contexts, Cases. Prentice Hall International (3rd Mito, Setsuo (1990) The Honda Book of Management. Process: edition). Process: edition). London: Kogan Page. Nonaka, Ikujiro (1991) ‘The knowledge-creating company’ Harvard Business Review, Nov-Dec, pp 96-104. Nonaka, Ikujiro (1988) Toward Middle-Up-Down Management. Management Review. Sloan Nonkaka, Ikujiro, and Takeuchi, Hirotaka (1995) The Knowledge-Creating Company. New York: Oxford University Press. 52 Pascale, Richard T. (1996) ‘Reflections on Honda’ California Management Review 38.4:112-17. Pascale, Richard T. (1990) Managing On the Edge. Penguin. Harmondsworth: Pascale, Richard T. (1984) ‘Perspectives on strategy: the real story behind Honda’s success’ California Management Review 26.3:47-72. Peters, Tom (1988) Thriving on Chaos. London: Pan. Peters, Tom (1992) Liberation Management. London: Pan/Macmillan. Quinn, James Brian (1991) Honda Motor Company. In Mintzberg, Henry, and Quinn, James Brian (1991) The Strategy Process: Concepts, Contexts, Cases. Prentice Hall International (2nd edition); pp. 284-299. Quinn, James Brian (1996) Honda Motor Company 1994. In Mintzberg, Henry, and Quinn, James Brian (1996) The Strategy Process: Concepts, Contexts, Cases. Prentice Hall International (3rd edition); pp. 849863. Rumelt, Richard P. (1996) ‘The many faces of Honda’ California Management Review 38.4:103-111. Sakiya, Tetsuo (1987) Honda Motor: The Men, the Management, the Machines. Tokyo: Kodansha International (2nd edn). Sanders, Sol (1975) Honda: The Man and his Machines. Brown. Boston: Little Shimizu, Koichi (1998) ‘Toyota Motor Company’, forthcoming in M. Freyssenet, A. Mair, K. Shimizu and G. Volpato (Eds.), One Best Way? Trajectories and industrial models of the world’s automobile producers, 1970-1995. Oxford: Oxford University Press. Shook, Robert L. (1988) Honda: An American Success Story. New York: Prentice Hall. Silbiger, Steven (1994) The 10-Day MBA. London: Piatkus. Stacey, Ralph D. (1993) Strategic Management and Organisational Dynamics. London: Pitman. Stalk, George, Evans, Philip,, and Shulman, Lawrence (1992) ‘Competing on capabilities: the new rules of corporate strategy’. Harvard Business Review, April-May, pp.57-69. Stalk, George, Jr. and Hout Thomas M. (1990) Competing Against Time. New York: The Free Press. Wheelen, Thomas L. and David J. Hunger (1995) Strategic Management and Business Policy Reading, Mass.: Addison Wesley, 5th edn. Whiteley, Richard C. (1991) The Customer Driven Company: Moving From Talk to Action. London: Random House. Whittington, Richard (1993) What is Strategy - And Does it Matter? London: Routledge. 53 Womack, James P., Jones, Daniel T., and Roos, Daniel. (1990). Machine that Changed the World. New York: Rawson Associates. The Endnotes 1 Rumelt (1996) only identifies three of these strategy industry participants who have made significant use of Honda -- Boston Consulting Group, Pascale, and Hamel and Prahalad -- but a fuller analysis of the business of knowledge relationship between the strategy industry and Honda can be obtained by broadening the discussion to other participants. Rumelt also focuses only on the debate between ‘emergent’ and ‘planning’ views of strategy, whereas the broader debate also encompasses a series of other strategy dichotomies which, it is argued here, it is important to recognise. Rumelt does make the point, however, that Honda is unusual in being used to represent a variety of different theoretical standpoints. 2 Use of these contentious labels in this paper is provisional. The arguments do not necessarily apply to all ‘Western’ thought. Ironically, they may not apply to the original hearth of AngloSaxondom in continental northern Europe. They certainly apply to the United States and the United Kingdom, however. Yet they may equally apply to certain Latin countries, in particular France. 3 Is it too far fetched to hypothesize that Honda in particular has been the subject of this considerable debate not simply because of its successes in recent decades, but precisely because, with its non-dualist approach to management thinking, the company does offer considerable evidence in support of each side of the conceptual polarities, permitting ‘each side’ of each argument to justify its position? 4 Good advice to the budding management guru might therefore be as follows. Analyse the latest management idea, develop a theory based on its polar opposite, and launch this theory when the latest theory is judged to be about to wane. On this model, the contemporary interest in ‘business process reengineering’ may soon give way to a turn towards ‘The organization’s new structure’, while the current interest in ‘learning organisations’ may give way to ‘The organization’s new intellectual division of labour’ in a tack back towards Taylorism. ...
View Full Document

This note was uploaded on 01/26/2010 for the course ADMS 4900 taught by Professor Jungchinshen during the Spring '10 term at York University.

Ask a homework question - tutors are online