Dept of Management and Economics SE-581 83 Linköping Sweden

MERGERS, MNES AND INNOVATION – THE NEED FOR NEW RESEARCH APPROACHES

Christian Berggren, Linköping University, Dept. of Management and Economics Paper presented at the ESRC Transnational Communities Conference on Multinational Enterprises, Warwick, September 6-8, 2001.

Mergers, MNEs and innovation - The need for new research approaches Abstract (max 120 words) A merger wave of unprecedented scale has swept through the Western world. Many studies have demonstrated the unimpressive economic outcome of most mergers. This article argues that an important reason for disappointing long-term performance is the under-estimated difficulties of integrating idiosyncratic technologies in unified product platforms. Instead of focusing on new product development, engineers and designers are absorbed by contentious harmonisation issues. An expanded research agenda is proposed, including the role of special interests in driving merger proposals, the consequences of mergers for engineering creativity and innovation, and the identification of industries where trajectories of mergers, scale and size, confront the alternative dynamics of de-mergers, knowledge based networks and research-based start-ups. Keywords : Mergers & acquisitions, merger performance, merger motives, merger makers, cross-border, innovation

2

1. Introduction A merger wave has swept through Europe and North America since the mid 1990s, affecting a wide range of sectors: banking, telecommunications, cars, pulp and paper, oil, and so on. Increasingly, these mergers are cross-border in nature. According to UN estimates, crossborder mergers and acquisitions increased from a level of $86 billion a year in 1991 to $1.1 trillion in 1999 (United Nations 2000). The record figure in 1999 represented a doubling on the previous year’s figure, itself a record. Many of these deals have brought together firms of comparable size, resulting in the combined entity having roots in at least two countries: BPAmoco, Astra-Zeneca, Daimler-Chrysler, Hoechst-Rhone Poulenc and VodafoneMannesmann are all examples. Recurrent waves of mergers have been part of economic life in Western economies since the end of the 19th century, and there exists a rich literature of merger studies. In financial studies the focus is on economic outcome, however defined. In management and organisation science the emphasis tends to be on post-merger implementation processes and problems, such as culture clashes, competitive position games, communication challenges or learning potentials. In much of this management literature, the basic financial rationale and economic imperative justifying the merger is more or less taken for granted (see Haspeslagh & Jemison, 1991, for example). The ex- or implicit agenda is to help managers manage the implementation process in a more effective way. This article argues that the current merger phenomenon needs to be investigated in a fundamentally more critical way in management research. The purpose firstly is to problematize mergers: their economic rationale, behavioural motives, and second-order consequences for innovation and engineering creativity. The aim is to contribute to a further debunking of determinist notions of “economic imperatives”. Secondly, an agenda for future merger research is proposed, comprising two aspects: collaborative approaches to uncover the special interest groups driving merger propositions and cross-disciplinary studies to investigate the long-term effects of mergers on engineering creativity, projects and innovation. The efficiency theory, viewing mergers as effective instruments to reap benefits of scale and scope, is still a widespread basis for merger studies, sometimes taken for granted without any qualifications (for example Seth 1990). Efficiency theory is at the heart of the prescriptive handbooks on merger strategies, and the basis for public relations exercises accompanying merger announcements. Therefore the article, in the second section presents a detailed review of the performance literature. A distinctive mark of the current merger wave is the prevalence of horizontal amalgamations, where firms in the same industry are being combined. Compared to conglomerate structures these amalgamations are very difficult to dissolve; essentially such mergers constitute irreversible processes. This emphasises the importance of analysing motives and probable outcomes. As is well-known to the informed reader, most outcome studies present a bleak picture of the financial post-merger performance. Much less discussed are the effects of mergers on innovative abilities and engineering creativity. Studies of organisational learning and cognition tend to emphasise the positive potentials of mergers for shared learning. One example is Leroy & Ramanantsoa (1997) who studied collaborative problem-solving workshops during an eight month implementation period. The conclusion stresses the merger implementation as a complex learning process with “each firm being simultaneously a learner and a teacher”, and advises managers to “improve the implementation process” by encouraging “learning between the merging firms” (p 889-90) This interest in the

1

possibilities of shared learning is well received, but tends to overlook the more long-term energy absorbing processes of technological integration. Building on case studies of engineering management in merging companies, the third section highlights the specific difficulties of unifying product platforms in technology-intensive companies, the opportunity costs, and the tendency of redirecting R&D personnel from creative performance to engineering routines. With some important exceptions (Ridderstråle 1996, Bresman & Birkinshaw 1996) these difficulties are not captured by the literature on post-merger implementation problems; one reason being its overly general orientation and short-time focus, another its lack of interest in the interaction between organisation, behaviour and technology, or the choice of firms where technology integration is not a critical issue. The fourth section of the article discusses the “merger enigma”. Given the evidence that most mergers produce disappointing results, why is the phenomenon so persistent? This question is not new. After a brief review of previous studies of managerial merger motives the article discusses recent explanations derived from game theory, which purport to give managerial action a new type of “rational explanation”. So far these approaches lack empirical support. The same seems to be true for a popular explanation in the business press of repeated merger behaviour: the assumption of a “M&A learning curve”. The role of calculation asymmetries in the merger process is emphasised: detailed pre-calculations at the time of the announcement, but a virtual absense of ex-post calculations in the post-merger phase. The fifth section develops an argument for an expanded agenda of merger studies, with an emphasis on two areas: (1) Merger motives and merger drivers, in particular the role of external financial promoters and the corporate finance industry. (2) Mergers and innovation. Here two types of studies are suggested: micro level studies of technology integration efforts and post merger performance of previously innovative R&D units; and studies of industries where contrasting logics and strategies are operating, economies of scale and size as well as economies of innovation and flexible networking, processes of “fusion” (mergers and acquisitions) as well as processes of “fission” (de-mergers and new firms generation).

2. Research on merger performance. “If merger incentives are taken into consideration at all, a group of firms is typically said to have incentives to merge if the profits of the merged entity in the new equilibrium is higher than the combined profits of the merging firms before the merger, this is the traditional criterion for merger incentives in the industrial organisation literature.” (Horn and Person 1999:2, italics added). The non-discussed problem here is: how do firms “know” that the combined profit will be higher? Maybe they think it will be, but what if they are wrong? Since the early 1970s, a plethora of studies have tried to find out the actual financial consequences of mergers. Three research approaches stand out: + Stock market, or event, studies, i.e. studies focusing on the stock market reaction to merger announcement (the “event”); + accounting studies, i.e. studies of corporate performance as reflected in quarterly or annual reports, comparing pre-merger with post-merger performance, or comparing the performance of merging companies with a control sample of non-merging companies; + and interview and survey studies enquiring into the results of mergers and possible reasons for failures or successes.

2

Less commonly, there have also been attempts to explore the effects of mergers & acquisitions on other economic indicators, such as market share. Comparing the three main approaches - performance, market and interview studies - interview studies tend to be most positive. This may be a result of self-selection bias; companies with a dismal post-merger performance would be reluctant to discuss this with external researchers. In interview studies, there may also be a problem of defining “success” in rigorous terms. The disappointing results of stock market and accounting studies Bild (1998) provides an extensive overview of the accounting and market studies from the 1920s to the mid 1990s. A cautious conclusion of his survey of performance studies, covering a total of 2,600 mergers, is that “companies engaging in merger activity do not earn a postmerger return that is different than the average of their industry, or any other chosen benchmark”. If there is a tendency it is negative (Bild 1998: 159). In a well-known overview Scherer and Ross were more outspoken, after finding that half of the acquired businesses in a comprehensive sample of US manufacturing corporations had experienced “disastrous performance declines” after the merger (1990:172-173): “The picture that emerges is a pessimistic one: widespread failure, considerable mediocrity and only occasional success”. To evaluate market studies it is important to distinguish between effects on the acquirers´ and the targets´ stock value, and between long and short-term effects. If there is a positive effect on share prices at the time of a merger announcement, the gains accrue to the target’s shareholders. Most acquisitions have either no value consequences or negative consequences for the stockholders of the acquiring firms. Several event studies, comparing the stock market value of the involved firms argue that the combined stock market gains are positive, however (Fridolfsson & Stennek 1999:2). This result is heavily influenced by the measurement period, a few weeks or days before and after the event. During this short period the stock market is strongly exposed to the orchestrated public relations efforts of the merger proponents, top managers and their financial promoters and media channels. More critical information and analyses need more time to influence the discussion and evaluation. Unsurprisingly, the upward effect on share prices tends to be short-lived. A recent case is the much-publicised Daimler-Chrysler merger. When the deal was announced, it was presented by Business Week as a ”marriage in automotive heaven” : “if ever a merger had the potential for that elusive quality – synergy – this could be the one” and investors applauded by pushing Chrysler´s shares up almost 20% after the announcement (Vlasic, et al 1998). Since then, share price and financial results have been on a long downhill slide. This is line with market studies adopting a longer measurement period for the acquiring firms’ stock performance, which have demonstrated significant deterioration in 1-3 years after the merger (Scherer & Ross 1990: 169-170). It may be that the capital market reaction at the announcement date reflects positive expectations. Supported by massive public relations exercises from the actors involved in the deal - which are subsequently followed by disappointments as espoused ‘synergies’ do not materialise. Also Rau and Vermaelen (1998) have found that merger firms tend to underperform on the stock market in the three years after the merger. Their explanation is that the market and management over-extrapolates the past performance of successful managers, so–called “glamour firms”. Most of these studies cover the merger waves of the 1960s and 1980s. So far , few scholarly investigations of the many horizontal M&As of the 1990s have been published. Reports by consultants indicate, however, that the problems of post-merger under-performance persist.

3

According to a study by AT Kearner of 135 large-scale international mergers (one billion dollar plus) in 1993 – 1996, 58% underperformed in terms of shareholder value two years after the merger. Moreover, the costs of failed mergers were higher than the gains of successful deals (Hedberg 1998). A similar result was obtained by an interview study carried out by an international consultant firm in 1999 (KPMG 1999). Mergers and market share – adding to the bleak picture Studies of mergers and market share are less common but tend to support the pessimistic conclusions of most accounting studies. In a study covering mergers of the 1950s and 1960s it was found that the market share of acquired business lines deteriorated significantly relative to control samples of non-merging businesses, both for conglomerate and horizontal mergers (Mueller 1985). In an analysis of the pharmaceutical industry, market share changes from 1993 to 1997 of merging companies was compared with companies growing mainly organically. All the companies involved in major mergers Smithkline/Beecham, Bristol&Myers/ Squibb, American Home/American Cynamamid, Roche /Syntex, Glaxo/Wellcome, Pharmacia/Upjohn, Hoechst Roussel /Marion Merrel Dow - had lost market share (AFV 1999). The “organic growers”, from Pfizer to Astra, had all expanded their market share during the same period. More recently, however, several firms in this group have also joined the merger camp. The risk of losing market share was, incidentally, a major reason why the management of Swedish truck producer Scania so vigorously opposed Volvo´s take-over attempt 1999-2000; Scania was finally salvaged by the veto of the European competition authorities. A study of the European heavy truck market 1975 – 1997, cited by Scania managers, demonstrated that the Swedish truck makers when going alone had enjoyed very respectable market share growth. Volvo had expanded its market share in Europe from 10.2% to 15.2%, and Scania had increased its share from 8.8 to 15.1%, a 70% expansion of market share entirely by organic growth. By contrast, several of the merging competitors had suffered disastrous declines. RVI, an amalgamation of Berlier, Saviem and Dodge UK had seen its share fall from 13.7 to 11.5%, whereas Iveco, a combination of Enasa, FIAT-OM, Ford UK, UNIC, Magirus, Seddon-Atkinson and Astra, went from 19.2% to 11.2%, a virtual implosion of market share (Scania 1999:5). On a sector level, mergers in the banking industry have received most attention, spurred by the keen interest of the Federal Reserve in the US. In Larsson (1997) the results of 174 studies are summarised. Taken together these studies do not support any efficiency arguments. There is no clear positive correlation between unit cost and volume, and big banks are on average no more efficient than smaller banks (possibly the cost curve is U-shaped, implying that macro and micro banks are the most inefficient). The same applies for the economy of scope argument; big full-service banks are no more successful than smaller niche banks. In spite of this negative historical track record, bank mergers have been on the increase for several years. The difficult trade-off between cost-cutting and revenue-enhancement One recent interview study (Capron 1999) gives important insights into why the synergies often referred to at the time of a deal are so difficult to realize. The study is based on a comprehensive survey of 200 firms. Cost-cutting, asset divestiture, and streamlining are separated from revenue-enhancing measures using various forms of resource redeployment. The cross-effect between these two are also studied. Further, measures directed towards the

4

acquiring and the target firms are analytically separated. Post-merger actions to rationalise acquirers’ assets are effective, but rare; instead the majority of actions are directed at divesting the assets of the target companies. These actions, however, have either no significant or negative impact on cost savings; moreover they tend to damage the revenueenhancing resources and capabilities of the target firm. “Overall, these results show the difficulties in capturing benefits of post-acquisition divestiture actions and support the limits … with respect to the effectiveness of acquisitions as a means of rationalising assets.” (Capron 1999: 1007). Regarding revenue-enhancing measures, resources from the target firm may be productively re-deployed, but in this case there is also a negative effect on total acquisition performance. If, on the other hand, resources from the acquiring firm are re-deployed to enhance capabilities, the effects tend to be unambiguously positive. However, this is not the first route taken in the post-merger process. In sum, this study points to the important problems of asymmetries and interaction effects, that the target firm “is very likely to bear the burden of post acquisition asset divestiture”, and that cost-cutting activities often interfere with and damage long-term capabilities. The author remarks: “Cost-based synergies, which have commonly been the focus of attention, are not easily achieved and may require more substantive changes in operating the business than those suggested by the dominant approach based on cost cutting and downsizing gains.” (ibid, p 1009-1010). Thus, irrespective of research approach, be it stock market studies, accounting studies or interview studies, the majority of empirical investigations give no support for merger enthusiasm. If there are any discernible results at all, they tend to be disappointing. Studies of post merger market share confirm this: more often than not merging companies suffer declining market share, sometimes a virtual market implosion. And because of expectations from and promises to the stock market, the acquiring firms tend to focus on cost-cutting measures, at the detriment of future revenue-enhancing capabilities. As the next section shows, this is particularly problematic when affecting the management of product technologies and product platforms.

3. Micro-merger snafus: the under-estimated problem of integrating technologies When mergers in manufacturing industries are announced, the deals are regularly supported by calculations demonstrating the potentials of huge savings in production and purchasing costs. If the amalgamation results in a substantially increase of market it may be possible for the new entity to put a strong pressure on suppliers to reduce their prices. A reduction of actual costs, however, requires integration of idiosyncratic technical solutions into common technical platforms. This may be feasible in a relatively short time in fast moving industries with short product lives, such as the ICT-sector (Cisco stands out as a prominent example, or at least appeared to do so before the stock market tumble of early 2001). It is much more difficult and time-consuming in other types of manufacturing or engineering sectors. In such cases there is seldom a simple set of objective parameters which can be used to select the superior design, be it a power transformer or a diesel engine. Instead there are dozens of relevant parameters, relating to performance, cost, manufacturability, serviceability, customer preferences, user behaviour, etc. Moreover, engineers tend to have strong feelings for their designs, very different from the approach of detached financial analysts. On the road to “integrated designs” there are extended negotiations, tension, conflicts and compromises, and at the end no guarantee that the best solution is chosen. This process, tending to be 5

substantially more protracted than envisaged by top management, is not well captured by general survey studies. Longitudinal, in-depth, case studies are necessary. This is a timeconsuming and therefore rather rare approach. Protracted technical integration problems at ABB A heavily publicised case of international M&A of the 1990s was the electro-technical firm ABB, the result of a merger between Swiss BBC and Sweden’s ASEA, and subsequent acquisitions in the US. Numerous business magazine articles and books have presented the new “global firm” as a resounding success; many of them have been overwhelmingly based on interviews with top management (for example Heimer & Barham 1998.) Two business areas, Automation and Power Transformers, however, have been subjected to more thorough studies of the problems of integrating different technologies, one study involving the author and an international group (Berggren 1996, 1999; Belanger et al. 1999), the other involving two researchers from Stockholm Business School (Bresman & Birkinshaw 1996). Since such studies are not very common, the results will be presented in some detail. The first case is about product integration in the business area Automation. When ASEA and BBC merged there were suddenly two competing systems in the industrial automation field, ASEA's Master and BBC's Procontrol. Which one should survive? The rivalry between the two systems and their proponents triggered hard arguments. When management decided to cease development of ProControl there "was blood on the floor", according to one participant (Berggren 1999:239). In 1990, ABB acquired Combustion Engineering in the U.S including Taylor Instruments whose control and supervision system for industrial processes enjoyed a higher market share in the chemical industry. ABB Automation decided to keep both systems but to merge them successively in future product generations, and eventually market only one system world-wide. A huge international project was set up consisting of twenty subprojects. The intention was to launch a fully integrated system in three years, but in 1996, six years after the start of the integration project, there were still important differences. One reason was the installed base of the existing systems. It was an important asset for the company, since customers rarely switch to a new system supplier. However, it also implied a major obstacle for the development of a unified platform, since new product releases must be compatible "backwards", to all the existing installations. Another reason for the difficulties in integrating the systems was the multi-site development structure, a result of growing by mergers and acquisitions. At the US site, for example, there was a limited interest in eliminating all the specific features of their system in favor of an international platform, since this could pose a threat to their own future (Berggren 1999: 239240). In ABB’s business area Power Transformers the problem of technological fragmentation was even more daunting. After all the mergers and acquisitions in the late 1980s and early 1990s, seven different product technologies were competing internally. A common product program started in 1989 but turned out to be much more complicated than envisaged. In 1996, seven years after the start, a common product protocol was in place, but only 20% of the transformers delivered that year was produced according to this protocol (Björkman 1999: 49). The soul of established machines – competing design philosophies at Volvo and Scania When Volvo Trucks announced its plan to take over Scania in 1999, the CEO at Scania publicly expressed his worries about future market share, and as we have seen above he had

6

reasons to be worried. In a less public way, there were also widespread concerns within Scania´s product development departments regarding problems with technological integration and standardisation. Volvo and Scania both build trucks for the heavy market segment. For financial analysts this means that there are obvious cost synergies in consolidation of product platforms and harmonisation of component designs. As explained by designers, product and process engineers in a training program ran by the author in 1999 - 2000 (“Advanced project management at Scania”), the engineer´s perspective is completely different. In spite of external similarities, design philosophies of Volvo and Scania are remarkably incompatible. This is illustrated by their ways of designing six-cylinder diesel engines, the most expensive component in a heavy truck chassis. Whereas Volvo offers a “modern” design with overhead cam shafts and one solid cylinder head, Scania prefers a “conventional” approach with individual cylinder heads and no overhead cam shaft, justified by reference to the low rotation speed of truck diesels compared to gasoline engines. By merger architects these differences may be seen as a mundane technical issue, easily resolved after an unbiased analysis. Such an assumption, however, is the root cause of many micro-merger problems. As pointed out by interviewed Scania engineers, diesel engines can be evaluated on many criteria: manufacturing cost, life cycle cost, reliability, power, emissions, fuel consumption, driving smoothness and convenience, versatility, serviceability, compatibility with auxiliary equipment, and so on. Adding to the complexity, all these criteria can be weighted differently. The choice of one design will not only affect suppliers and plant tooling, but also spare parts systems and thousands of service shops. After a Volvo-Scania merger, a unified design would be necessary to achieve announced cost rationalisation. But in the absence of simple and agreed criteria, such design decisions could easily be disputed as political. Whereas engineers in the pre merger situation concentrated on doing their best to develop their own solutions and let the marked decide, the post merger process could easily be marked by inward arguments between competing designs, resulting in a loss of tempo and customer focus. Nothing of this is factored into the calculations of cost savings regularly presented at merger announcements. Opportunity costs and distraction of innovative energies The difficulty of integrating different technologies in a common platform does not only mean that the expected cost synergies will take longer time to realise than anticipated. There is also a substantial opportunity cost. After the merger is legally finalised, designers, product and process engineers tend to be absorbed by co-ordination and harmonisation issues for several years instead of focusing on innovation and development of new products. Engineering is discipline of creative problem-solving, about creating and implementing new solutions. In an amalgamated company preoccupied with establishing “common platforms” this role changes. To realise the economies of scale, which have justified the merger, engineers and project managers are obliged to prioritise standardisation and formalisation: establish common systems, common design rules, common, purchasing policies, etc. Creative position tend to be transformed into implementers, standardises and engineering bureaucrats. This change is resisted by many; some of them leave, perhaps to competitors. Others continue to fight for their own solutions and ideas, which is one reason the technological integration process is so time-consuming. Another consequence of large-scale mergers is that innovative individuals and groups in design and development departments become surrounded by and have to report to several new organisational and hierarchical layers. This exhausts scarce human capacities, and strains or

7

severs the crucial direct links between innovators, manufacturing experts and advanced users. Such an increase of organisational layers and managerial reporting requirements was experienced by the legendary innovator in engine technology Per Gillbrand at Swedish Saab, after the company was acquired by GM. Instead of focusing on early introduction of advanced combustion and ignition technologies, using the advantage of their small-scale, interdisciplinary environment, Gillbrand and his colleagues had to spend much of their energy trying to persuade various committees in the central development office of GM, Detroit with its 23,000 heavily departmentalised engineers. (Interview by author, Jan 10, 2001). Innovative projects need a strong culture and practice of cross-functional and interdisciplinary information exchange and problem-solving. In the mega-enterprises resulting from the cross-border mergers of the 1990s, hierarchical structures are strengthened in order to reap the hoped for synergies and economies of scale. Global product managers or functional managers get the upper hand at the expense of cross-functional project management. This may leverage the sales of and refinement of existing products but erode the capacity for future innovation.

4. The conundrum: so poor performance, so many deals “Perhaps merger booms and stock market trading are behavioral phenomena – human being, like some animals are more active when the weather is sunny.” (Brealey & Myers 1996: 942.) When mergers are announced, synergies, cost-rationalisation, and elimination of overlap are favourite arguments advanced to support the deal. It is also commonly argued that mergers are a way of correcting managerial failure: that is, inefficient managers are weeded out by an active market in corporate control. The evidence does not support the view that seeking out greater efficiency is the main motive for mergers. Already in Scherer & Ross it was shown that target firms significantly outperformed comparative non-targets in terms of profitability. (Scherer & Ross 1990:170). This patterns has more recently been confirmed by Franks and Meyer (1998), who demonstrate that firms subject to take-overs tend to have outperformed the market in the previous five years, the opposite of what the efficiency arguments predict. The general “weeding-out” hypothesis thus enjoys weak support. A positive merger learning curve? During recent years, there has been a great interest in organisational learning in management research. Transposed to the M&A field this implies the proposition of an “acquisitions learning curve”, and the assumption that ”the experienced acquirer” would be more successful than the less experienced. Empirical studies have failed to find support for this assumption. Studies by Ravenscraft and Scherer 1987 found active acquirers to be less profitable than the US industry average, and less successful in terms of long-term stock performance. An organisational behaviour-explanation to this outcome would be that although executives ride the learning curve of acquisitive assessments and negotiations, the managerial ranks tend to be worn out, and their commitment increasingly eroded by every new acquisition, restructuring and re-staffing exercise. Experienced investors have suggested a performance curve shaped as an inverted U: First the acquirer experiences a positive learning curve, as its executives learn to assess targets and direct post-merger integration more effectively. After a series of successful acquisitions, however, top management tends to over-extend, trying to digest too large objects and be less attentive to details, and so their merger performance starts

8

to deteriorate. The evolution of Electrolux from the 1960s to the 1980s is cited as a case in point (L Låftman, senior pension fund manager, interviewed by author Febr. 8, 2001). One reason for the survival of efficiency claims is the lack of follow-up studies, both within companies and in the business press. Quantitative analysis of expected benefits are often used ex-ante, but very seldom ex-post. In an in-depth financial analysis of seven merger cases, it as found that very few companies conducted any post-audits of merger performance; if they did it was at the explicit request of the board. Stated by a financial director: “Ex-post evaluations are seldom made. Restructurings protect the management from evaluations” (quoted in Bild 1998:95). The lack of follow-ups and critical scrutiny of track record is evident also in cases where relevant information is easily accessible. In 1995, for example, ABB announced the merger of its traction division (trains and railway systems) with German AEG, owned by Daimler-Benz. At the press conference, the CEO of ABB presented a very positive future for the new entity, called Adtranz: The expected synergy effects would quickly result in new orders and increased profitability, and he was convinced that “we will improve earnings every year” (SvD 1995). In that year, the ABB traction division was still a profitable business, reporting earnings of 207 MUSD. The next year the results of the amalgamated entity had slipped into negative (- 2 MUSD), and in 1997 the economic performance had gone from bad to worse ( - 111 MUSD). In 1998, ABB sold out the entire business. This very year the same CEO, but now in a different position, was espousing the virtues of merging Swedish pharmaceutical company Astra with British Zeneca. This was an industry he had no experience of; nevertheless he was never confronted with the grossly misleading ex ante calculations of the Adtranz merger, an industry where he had 15 years of experience. Corporate disinterest in ex post calculations seems to be supported, at least in small countries, by media deference to executive merger promoters. An explanation building on game theory The conundrum is not resolved. So many studies show the economic disappointments of mergers, and yet the volume of merger deals keep increasing. In a survey of existing studies of merger motives (Trautwein 1990), a list of possible explanations was suggested, such as the bounded rationality of decision processes, managerial self-interest in increased compensation and power, and ‘empire-building’. Executive hubris is another hypothesis (Roll 1986). Several studies support such alternative explanations, but so far the empirical evidence seems to be limited and a bit dated. The World Investment Report by UN 2000 asked basically the same question: why so many mergers, when results are so discouraging? The answer in brief is that fear is a more important driver than greed (but on a personal management level the two incentives can be combined). A dubious deal is preferred to no deal. Recently economists have sought to build a “rational” explanation along the same line, using game theory to formalise the argument (Fridolfsson & Stennek 1999:3-4). According to this theory of ”preemptive or defensive merger motives”, mergers are performance-reducing propositions for all actors in an oligopolistic industry. But for company A its is preferable to take the initiative and merge with B, instead of waiting for a merger between B and C and then become an outsider: “even if a merger reduces the profit flow compared to the initial situation, it may increase the profit flow compared to the relevant alternative – another merger…. In particular, the event studies can be interpreted to show that there exists an industry-wide anticipation of a merger and that the relevant information content of the merger announcement is which firms are insiders and which are outsiders.” (3-4, also 17). This might

9

sound plausible, but two questions remain. First, are “merger insiders” actually more successful than outsiders? So far, the hypothesis is not supported by empirical evidence, showing that outsiders´ stock value suffer, or that outsiders are less profitable than the merging insiders (Fridolfsson & Stennek 1999:23-24). Second, how and by whom is the “industry-wide anticipation of a merger” created and sustained? This leads us to the role of the financial promoters of mergers. The increasing importance of the corporate finance industry. Most merger studies focus on the motives of acquiring firms and their financial justification. A different approach is taken by those looking specifically into the role of match makers and financials promoters. In an early study of the first American merger waves, Markham concluded, as summarised by Scherer & Ross (1990:161) “the quest for promotional profits was the most important single motive for merger during the frenzied 1897-1899 and 1926-29 periods”. The hectic and dubious activities of financial match makers, rushing into the market for corporate control in the early 20th century was also observed by the Swedish Commission of Anti-Trust Legislation in 1921 (Rydén 1972). In the 1960s, the deal makers seem to have played a less prominent role. But in the 1980s, corporate finance firms and management strategy consultants, heavily involved in M&A businesses, started to grow explosively. ”We were in the era of megabid mania. The City was awash with money, One was getting approaches from banks all over the world offering all sorts of propositions for any sort of deal that one could think. “ (Ernest Saunders, former CEO of Guinness, quoted in O´Shea & Madigan 1997:236). The prize for participation in mega deals is difficult to resist. When, for example, Guinness took control of Distillers in 1986, the largest British corporate acquisition of the time, the transaction generated more than 250 MUSD in fees to brokers, consultant, bankers and others (ibid, p 241). In the late 1990s, American firms were at the forefront of the merger wave in Europe, while the relative weight of German and other mainland European banks was declining. According to an estimate in Business Week (Nov 1, 1999), Goldman Sachs was involved in European mergers and acquisitions totalling 430.9 billion dollars during the first 10 months of 1999 alone. Morgan Stanley´s European M&A business totalled deals of 421.3 billion dollars for the same period. These two firms were the No.1 and No.2 in the market for mergers and acquisitions in Europe, by far outdistancing any local rivals. Corporate finance advisors and consultants are not neutral service providers. Since mergers present such opportunities to earn large commissions, they are actively contributing to new deals: ”I would rather say that they /the investment banks/ are accelerating corporate restructuring. It´s their job to present analyses and proposals for transactions all the time.” (Senior advisor at Morgan Stanley, AFV 2000:34 ) Previous research has uncovered the importance of managerial self-interest. However, apart from anecdotal evidence there are very few studies of the role of ‘financial advisors’ and dealmakers in the recent merger booms. Consulting firms are themselves going through a phase of concentrations and mergers, and many now constitute a key component of the current international business environment, alongside governments and international organizations. Moreover, in the case of the so-called ‘Big Five’, a potential conflict of interest exists between their roles as advisors and as auditors (Perks 1993). More broadly, the issue is about the role of these special interests in shaping of the M&A agenda. What role do the financial mediators and “advisors” play in shaping the “industry-wide anticipation of a merger”,

10

referred to above by Fridolfsson & Stennek (1999), as a precondition for merger moves which for all involved will result in deteriorated performance relative to a no-merger development?

5. The need for an expanded research agenda including deal makers and the consequences of mergers for innovation. As this article has shown there is a rich literature on merger performance, measured in various ways. The wave of international cross-border mergers since the 1990s calls out for renewed studies of the merger-making process, however, specifically the influence of external financial promoters, deal makers and consultants. Certainly the question to what extent the wave of mergers is driven by models and economic interests perpetuated by deal-makers is an important one. There is a need for case studies to investigate the role of financial advisors and consultants in selected processes from the initial steps, the drafting of a business case and its financial justifications, to participation in negotiations, set-up of financial arrangement, and finally, involvement in post-merger restructuring and integration. To bring this about, systematic efforts by academics to team up with insiders, for example recently retired executives, would be necessary. Cross-disciplinary studies will be needed, involving students in financial economics, organisational behaviour as well as economic sociology. There is also a need for systematic efforts to explore the consequences of mergers on technological development and innovation, both inside firms and within industries. Such studies of mergers and innovation at firm level will benefit from inter-disciplinary case studies, involving researchers in organisational behaviour and engineering management. One issue would be to follow the efforts to harmonise technological platforms and design structures: time needed compared to initial forecasts, departmental relations and friction, opportunity costs, and changes in the overall activities and orientation of R&D staff in the merging firm. Further, it would be important to study how new-product development projects are affected by, or possibly escape from, increasing organisational complexity and hierarchical reporting requirements. Of particular importance is to study highly uncertain development projects, which tend to be dependent on informal networks, cross-departmental problem-solving and project autonomy. All this may be at risk after an international merger, which normally results in more complex management layers and more stringent requirements on formal progress reviews. Thirdly, it is important to identify particular innovative units or in-company networks, for example by looking at historic patenting records, and explore if and to what degree they become surrounded and compartmentalised by added organisational boundaries and intermediaries. This pertains both to the problem of opportunity costs, and management by trust versus formal reviews. If previously innovative units lose their direct contacts to process specialists or advanced customers, this will not only delay problemsolving, but may degrade their innovative abilities as a whole. To capture consequences of mergers on innovation and the orientation of R&D staff within firms, longitudinal, in-depth case studies are important. This is a difficult venue, however. The potential biases affecting multi-firm interview studies are even more relevant for in-depth case studies. Most insider stories are published by journalists. They provide a mass of imaginative details and outspoken opinions, but tend to be highly partisan and lack in comparative perspective; a recent example is the all-American account of the DaimlerChrysler merger, Vlasic & Stertz 2000. Academic case studies on the other hand tend to be cautious and restrained, refraining from interviewing dissidents and external critics, and sometimes not clarifying the role of the researchers: investigator or dialog partner to 11

management? This was a highly controversial issue in the research team behind the ABB study ”Being local world-wide”, in which the author was one key participant (Belanger, et al 1999). In other cases there are great difficulties in getting the stories released for publication. This has been the case for the planned publication ”Acquisitions: the management of social drama” by John Hunt at London Business School, a well-known specialist in merger processes (Hunt et al. 1987). The book was to be released by Oxford University Press in 1998, but in 2000, a definite date was still not settled, for reasons explained in a personal communication to the author (Febr. 16, 2000): “We are have great problems getting the cases released. Of the 10 acquisitions we studied only one has so far been released. The problem, as you know, is that the 'real' story is very personal. It is this story we want to tell. It is not the story corporations want to read in public.” Mergers are strategically important affairs, and reports of failures and stifled innovation will land squarely at the table of top management, sometimes affecting the share price, but almost certainly affecting the relationship of the researchers to management in the future. As MacDonald & Hellgren (1994) have noted “most management researchers who interview in organisations crave access to the executive suite”. This is particularly so in merger studies. The focus on the top levels make the problems of selectivity and subtle adaptation especially relevant: “It is hard to question closely a manager who normally would not tolerate being questioned at all. It is much easier simply to accept what is said,….It is easier still, and much more conducive to reaping the benefits that flow from the satisfaction of those interviewed, to ask the questions managers wish to answer and to ask them in ways managers will find immediately acceptable. Thus, for example, a question on the role the manager has played in corporate success is much more acceptable than a question about his role in corporate failure.” MacDonald & Hellgren (1994: 13,15). Corporate failures, however, abound in merger processes. A way, albeit time-consuming, out of the dilemma may be to go for lower levels of the organisation, which are anyway the most important when investigating consequences for innovation and development projects. A complementary approach which might ameliorate publication problems could be to focus on project, not firm performance,. In addition to rich, longitudinal case studies, another type of meso-oriented approach should to be explored to capture the consequences of mergers for industrial dynamics and innovation. Such an approach would entail the identification of industries where companies pursuing international economies of scale confront firms, or cluster of firms, competing on the different advantage of knowledge networks and innovation. The Swedish industry of pharmaceutical and biotechnical companies, broadly defined, may constitute such an industry case. For many years this industry was dominated by two domestic giants, Pharmacia and Astra. These firms experienced rapid growth on the back of original innovations, but increasingly both of them, first Pharmacia and then Astra, changed orientation to an emphasis on incremental product refinement and international scale. In the 1980s, Pharmacia started out on a route of mergers and acquisitions which was finalised by the cross-border merger with American Upjohn. This merger has been viewed as a particular disastrous deal in the Swedish debate, citing both exorbitant implementation costs and an unending series of series of restructuring and relocation exercises (Frankelius 1999). Astra for a long time professed its belief in organic growth, but in 1998 joined forces with British Zeneca, a descendant of ICI. In the Pharmacia case, corporate control was transferred to the US, in the Astra case corporate control moved to London. By the merger proponents this was seen as a deterministic development because of the new economies of scale and necessities of global presence, but the local consequences were often demoralising.

12

In a highly unintentional process, however, the Swedish pharmaceutical and bio-technical industry was endowed with a new lease of life by a plethora of research-based start-ups, often originating from university projects, but managed by former employees of Astra and Pharmacia. Specialising in narrow niches, these firms tend to cluster in the Uppsala region, the Cambridge of Sweden and previous headquarters of Pharmacia. Here they form close knowledge networks with academics and other dedicated firms (Thorén 2000). The jury is still out regarding the long-term prospect of this breed of small-scale, research-intensive firms. From the perspective of industrial innovation, this process of fission offers an extremely interesting antidote to the large-scale processes of mergers and “fusion”, which for so long has been propagated as the only route for a competitive pharmaceutical industry. A study of this or similar cases would include long-term growth in earnings and high tech job creation, as well as specific company strategies to circumvent the problems of small scale and resource restrictions compared to the accumulated powers of international mega-companies.

6. Summary and conclusion Historically there have been several merger booms in the Western economies. In the US, and in Sweden, the early 20th century as well as the 1920s witnessed major waves of horizontal amalgamations, often with the explicit motives of creating dominant market positions. Tight anti-trust legislation in the US made this more difficult in the decades after the Great Depression. As a result, the merger booms in the 1960s was dominated by the conglomerate type of combinations. In the espoused merger motives of both the 1960s and 1980s, financial risk reduction due to diversification, and cross-industry technological synergies, figured prominently. During the 1980s there was a considerable relaxation of American antitrust legislation. Encouraged by this trend, the new boom of the late 1990s was once again characterised by horizontal mergers, involving firms in the same industry and principal market. In many industries this implied a renewed move towards consolidation and oligopoly. Compared to previous amalgamations, the scale and volume increased with an order of magnitude. This was accompanied by an ideology heralding the benefits of size and economies of scale in spite of recent advances in flexible manufacturing technologies and the increasing importance of customisation. Whereas most of the mergers of the early 20th century, of the 1960s and the 1980s remained domestic in nature, mergers in the 1990s were predominantly of a cross-border character. The diffusion of international mergers was supported by a new “deal-making industry” of consultants, corporate lawyers, investment banks and corporate finance specialists. They seemed to be driving the merger process in a way that exhibited similarities to the role of financial promoters of the early 20th century waves, but in a decidedly more international and aggressive way. This article started with a review of studies of how mergers have performed historically, singling out three types of approaches: stock market, accounting and interview studies. The result of this overview supported the observation by Bild (1998:216): “It is striking that the three approaches have largely presented fairly similar, that is, mostly insignificant, results. Where any significant outcomes have been found, they have been reported to be predominantly negative.“ Available studies of the mergers of the 1990s did not deviate from this bleak picture. After marshalling this evidence of disappointing performance the article delved deeper at one specific reason for post-merger problems: the difficulties of integrating idiosyncratic corporate technologies into common product platforms. Illustration of this problem was given both from implemented and planed mergers. A related argument focused on the opportunity costs of mergers. By concentrating on cost synergies and stream-lining, it 13

was argued, mergers in knowledge-intensive industries tend to distract and impede innovative energies. This is aggravated by the tendency to redirect product and technology units towards issues of standardisation and commonisation rather than innovation and experimentation. After reviewing mergers from the perspective of performance and consequences for innovation, a conundrum was presented: why so many mergers when most of them perform so poorly? Recently suggested “rational propositions” inspired by game theory was found lacking in empirical evidence, and the need for a closer look at the industry of merger makers was emphasised. The problematic influence of these financial intermediaries was noted already in the merger boom of the early 20th century, and again in the analyses of the notorious role played by Drexel Burnham Lambert and Kohlberg, Kravis and Roberts in the U S merger boom of the 1980s ( Stearns & Allan 1996). Now the time is due for renewed studies, involving a co-operation of academic scholars and senior business ex-insiders. Interdisciplinary studies, combining management and engineering disciplines, are also necessary to investigate the “micro-merger” problems of integrating diverse technologies, and more broadly to assess the complex consequences of merger for innovation. In-depth studies in individual companies are important here, but need to be complemented by analyses of industries, where contrasting logics are operating: both large-scale mergers (“fusions”) and small-scale “fission”, i.e. spin-offs and new start-ups initiated or inspired by former big firm-employees and their networks. In many industries, such as the car industry, high capital intensity of production or distribution systems erect prohibitive barriers to new entrants even when mergers result in enormous inefficiencies. In other sectors, scientific and technical breakthroughs may unravel previous barriers. One such case might be the pharmaceuticals /biotechnology sector, in for example Sweden. Here the dominant firms´ long-time pursuit of size and scale ended in international mergers and relocation 1995-99. Since the late 1990s, however, this trajectory is confronted by a new logic of networking and research-based entrepreneurial start-ups, frequently managed by veterans from the merging giants. Studies of such contrasting logics within an industry are important not only for the insights they will yield in factors supporting or impeding innovative processes. Even more important may be their contribution to a deconstruction of the economic determinism, arithmetic modelling and linear thinking, so popular in the era of international mega-mergers.

References AFV 1999. “Den oundvikliga fusionen. ”, Affärsvärlden (“Business World”) No. 1-2, Jan 13, p. 22-27. AFV 2000. “Anders Ljung. Inte förvånande att Telia dröjde”, Affärsvärlden, No. 24, June 15, pp 29-34. Belanger, J., Berggren, C., Björkman, T. & Köhler, C. (eds.) (1999). Being local world-wide. ABB and the challenge of global management. Ithaca: Cornell University Press. Berggren C (1996) "Building a truly global organization? ABB and the problem of integrating a multi-domestic enterprise." Scandinavian Journal of Management. Vol. 1. Berggren, C. (1999). Distributed Development in a Multinational. In Belanger, et al. (eds.) 1999, pp 233-247.

14

Björkman, T (1999). Lean Management in Practice, the Headquarters Perspective. In Belanger, et al. (eds.) 1999, pp. 36-60. Bild, M. (1998).The valuation of take-overs. Stockholm: Stockholm School of Economics. Bresman, H. & Birkinshaw, J. (1996). ABB and Combustion Engineering. Stockholm: IIB. Brealey, R. A. & Myers, S. C. (1996). Principles of corporate finance. N Y: McGraw- Hill. Capron, L .(1999). The long-term performance of horizontal acquisitions, Strategic Management Journal, 20, 987-1019. Dickerson, R. et al. (1995). The Impact of Acquisitions on Company Performance: Evidence from a Large Panel of UK Firms. University of Kent Research Paper, Canterbury. Frankelius, P. (1999). Pharmacia & Upjohn, erfarenheter från ett världsföretags utveckling Värnamo: Liber Ekonomi. Franks, J. & Meyer, C. (1996) Do Hostile Takeovers Improve Performance?, Business Strategy Review, 7(4), 1-6. Fridolfsson, S-O & Stennek, J. (1999). Why Mergers Reduce Profits and Raise Share Prices. Stockholm : IUI. Haspeslagh, P.C. & Jemison, D. D. (1991) Managing acquisitions: Creating value through corporate renewal. N Y : Free Press. Hedberg, C. (1998) Många fusioner misslyckas, Svenska Dagbladet, Dec 14, 1998 Heimer, C. & Barham, K. (1998) ABB The Dancing Giant. London: Financial Times/Pitman Publishing. Horn, H. &Persson, L. (1999). The Equilibrium Ownership of an International Oligopoly. London: Centre for Economic Policy Research. Hunt, J., Lees, S., Grumbar, J.J. & Vivian, P. D.(1987). Acquisitions: the human factor. Oxford University Press. Intika Consulting (1999). 'From Deal to Delivery'. Acquisitions Monthly, August pp. 29-31. KPMG (1999). Unlocking shareholder value: The keys to success. London: KPMG. Larsson, C-G, 1997. Bankfusioner – kan de löna sig? Ekonomisk debatt, 25(4), p 229-239. Leroy, F. & Ramanantsoa, B. (1997). The Cognitive and Behavioural Dimensions of Organisational Learning in a Merger: An Empirical Study, Journal of Management Studies, vol 34 (6), 871-894.

15

MacDonald, S. & Hellgren, B. (1994) The interview in management research. Warwick Business School Working paper Mueller, D. C .(1985). Mergers and market share, Review of Economics and Statistics. 67(2), 259-267. Perks, R. (1993) Accounting and Society. London: Chapman and Hall. O´Shea, J. & Madigan, C. (1997). Dangerous company. London: Nicholas Brealey Publ. Ravenscraft, D. & Scherer, F. (1987). Mergers, Sell-offs and Economic Efficiency. Brookings Institution , Washington DC. Ridderstråle, J. (1996). Global innovation. Managing international innovation projects at ABB and Electrolux. Stockholm: IIB. Roll, (1986) The Hubris Hypothesis of Corporate Takeovers, Journal of Business 59(2), part 1, 197-216. Rydén, B. (1972). Mergers in Swedish Industry. Stockholm: Almqvist & Wiksell. Saxenian, A. (1994). Regional Advantage - Culture and Competition in Silicon Valley and Route 128, Harvard University Press, Cambridge MA. Scania 1999. “Scania Inside”, No. 2, Jan 27. Scherer, F. M. & Ross, D. (1990). Industrial Market Structure and Economic Performance. Boston: Houghton Mifflin Company. Seth, A. (1990). Sources of value creation in acquisitions: an empirical investigation, Strategic Management Journal, Vol 11, 431-446. Stearns, L.B. and Allan, K D (1996). Economic Behavior in Institutional Environments: The Corporate Merger Wave of the 1980s. American Sociological Review, 61, 4, Aug, 699-718 SvD 1995. “ABB gör storaffär med Daimler-Benz”, Svenska Dagbladet, March 17. Thorén, M. (2000). “Det våras i biotech valley”, Veckans affärer, No. 14, April 3, p 39 – 48. Trautwein, F. (1990). Merger motives and merger prescriptions. Strategic Management Journal, 11 (4), 283-295. United Nations (2000) World Investment Report. New York: UN. Vlasic, B., Kerwin, K. &Woodruff, D. (1998). Daimler & Chrysler What the deal would mean, Business Week, May 18. Vlasic, B. & Stertz, B. A. (2000). Taken for a Ride. NY: William Morrow & Co.

16

17