Free Essay

Philips vs Matsuhita

In: Computers and Technology

Submitted By assarasua
Words 9318
Pages 38
9-910-410
DECEMBER 11, 2009

CHRISTOPHER A. BARTLETT

Philips versus Matsushita:
The Competitive Battle Continues
Throughout their long histories, N.V. Philips (Netherlands) and Matsushita Electric (Japan) had followed very different strategies and emerged with very different organizational capabilities.
Philips built its success on a worldwide portfolio of responsive national organizations while
Matsushita based its global competitiveness on its centralized, highly efficient operations in Japan.
During the first decade of the 21st century, however, both companies experienced major challenges to their historic competitive positions and organizational models. Implementing yet another round of strategic initiatives and organizational restructurings, the CEOs at both companies were taking their respective organizations in very different directions. At the end of the decade, observers wondered how the changes would affect their long-running competitive battle.

Philips: Background
In 1892, Gerard Philips and his father opened a small light-bulb factory in Eindhoven, Holland.
When their venture almost failed, they recruited Gerard’s brother, Anton, an excellent salesman and manager. By 1900, Philips was the third largest light-bulb producer in Europe.

Technological Competence and Geographic Expansion
While larger electrical products companies were racing to diversify, Philips made only light-bulbs.
This one-product focus and Gerard’s technological prowess enabled the company to create significant innovations. Company policy was to scrap old plants and use new machines or factories whenever advances were made in new production technology. Anton wrote down assets rapidly and set aside substantial reserves for replacing outdated equipment. Philips also became a leader in industrial research, creating physics and chemistry labs to address production problems as well as more abstract scientific ones. The labs developed a tungsten metal filament bulb that was a great commercial success and gave Philips the financial strength to compete against its giant rivals.
Holland’s small size soon forced Philips to look aboard for enough volume to mass produce. In
1899, Anton hired the company’s first export manager, and soon the company was selling into such diverse markets as Japan, Australia, Canada, Brazil, and Russia. In 1912, as the electric lamp industry began to show signs of overcapacity, Philips started building sales organizations in the United States,
Canada, and France. All other functions remained highly centralized in Eindhoven. In many foreign countries Philips created local joint ventures to gain market acceptance.
________________________________________________________________________________________________________________
Professor Christopher A. Bartlett prepared the original version of this case, “Philips versus Matsushita: A New Century, A New Round,” HBS
No. 302-049. This version was prepared by the same author and is a continuation of a series of earlier cases by Professor Bartlett including
"Philips versus Matsushita: Preparing for a New Round,” HBS No. 399-102, “Philips and Matsushita: A Portrait of Two Evolving Companies,”
HBS No. 392-156, and “Matsushita Electric Industrial (MEI) in 1987,” HBS No. 388-144. HBS cases are developed solely as the basis for class discussion. Cases are not intended to serve as endorsements, sources of primary data, or illustrations of effective or ineffective management.
Copyright © 2009 President and Fellows of Harvard College. To order copies or request permission to reproduce materials, call 1-800-545-7685, write Harvard Business School Publishing, Boston, MA 02163, or go to www.hbsp.harvard.edu/educators. This publication may not be digitized, photocopied, or otherwise reproduced, posted, or transmitted, without the permission of Harvard Business School.

910-410

Philips versus Matsushita: The Competitive Battle Continues

In 1919, Philips entered into the Principal Agreement with General Electric, giving each company the use of the other’s patents, while simultaneously dividing the world into “three spheres of influence.” After this time, Philips began evolving from a highly centralized company, whose sales were conducted through third parties, to a decentralized sales organization with autonomous marketing companies in 14 European countries, China, Brazil, and Australia.
During this period, the company also broadened its product line significantly. In 1918, it began producing electronic vacuum tubes; eight years later its first radios appeared, capturing a 20% world market share within a decade; and during the 1930s, Philips began producing X-ray tubes. The Great
Depression brought with it trade barriers and high tariffs, and Philips was forced to build local production facilities to protect its foreign sales of these products.

Philips: Organizational Development
One of the earliest traditions at Philips was a shared but competitive leadership by the commercial and technical functions. Gerard, an engineer, and Anton, a businessman, began a subtle competition where Gerard would try to produce more than Anton could sell and vice versa. Nevertheless, the two agreed that strong research was vital to Philips’ survival.
During the late 1930s, in anticipation of the impending war, Philips transferred its overseas assets to two trusts, British Philips and the North American Philips Corporation; it also moved most of its vital research laboratories to Redhill in Surrey, England, and its top management to the United States.
Supported by the assets and resources transferred abroad, and isolated from their parent, the individual country organizations became more independent during the war.
Because waves of Allied and German bombing had pummeled most of Philips’ industrial plant in the Netherlands, the management board decided to build the postwar organization on the strengths of the national organizations (NOs). Their greatly increased self-sufficiency during the war had allowed most to become adept at responding to country-specific market conditions—a capability that became a valuable asset in the postwar era. For example, when international wrangling precluded any agreement on three competing television transmission standards (PAL, SECAM, and NTSC), each nation decided which to adopt. Furthermore, consumer preferences and economic conditions varied: in some countries, rich, furniture-encased TV sets were the norm; in others, sleek, contemporary models dominated the market. In the United Kingdom, the only way to penetrate the market was to establish a rental business; in richer countries, a major marketing challenge was overcoming elitist prejudice against television. In this environment, the independent NOs had a great advantage in being able to sense and respond to the differences.
Eventually, responsiveness extended beyond adaptive marketing. As NOs built their own technical capabilities, product development often became a function of local market conditions. For example, Philips of Canada created the company’s first color TV; Philips of Australia created the first stereo TV; and Philips of the United Kingdom created the first TVs with teletext.
While NOs took major responsibility for financial, legal, and administrative matters, fourteen product divisions (PDs), located in Eindhoven, were formally responsible for development, production, and global distribution. (In reality, the NOs’ control of assets and the PDs’ distance from the operations often undercut this formal role.) The research function remained independent and, with continued strong funding, set up eight separate laboratories in Europe and the United States.
While the formal corporate-level structure was represented as a type of geographic/product matrix, it was clear that NOs had the real power. They reported directly to the management board,

2

Philips versus Matsushita: The Competitive Battle Continues

910-410

which Philips enlarged from four members to 10 to ensure that top management remained in contact with the highly autonomous NOs. Each NO also regularly sent envoys to Eindhoven to represent its interests. Top management, most of whom had careers that included multiple foreign tours of duty, made frequent overseas visits to the NOs. In 1954, the board established the International Concern
Council to formalize regular meetings with the heads of all major NOs.
Within the NOs, management structure mimicked the legendary joint technical and commercial leadership of the two Philips brothers. Most were led by a technical manager and a commercial manager. In some locations, a finance manager filled out the top management triad that typically reached key decisions collectively. This cross-functional coordination capability was reflected down through the NOs in front-line product teams, product-group-level management teams, and at the senior management committee of the NOs’ top commercial, technical, and financial managers.
The overwhelming importance of foreign operations to Philips, the commensurate status of the
NOs within the corporate hierarchy, and even the cosmopolitan appeal of many of the offshore subsidiaries’ locations encouraged many Philips managers to take extended foreign tours of duty, working in a series of two- or three-year posts. This elite group of expatriate managers identified strongly with each other and with the NOs as a group and had no difficulty representing their strong, country-oriented views to corporate management.

Philips: Attempts at Reorganization
In the late 1960s, the creation of the European Common Market eroded trade barriers and diluted the rationale for independent country subsidiaries. New transistor-based technologies demanded larger production runs than most national plants could justify, and many of Philips’ competitors were moving production of electronics to new facilities in low-wage areas in Asia and South America.
Simultaneously, Philips’ ability to bring its innovative products to market began to falter, and in the 1960s it watched Japanese competitors capture the mass market for audiocassettes and microwave ovens, two technologies it had invented. A decade later, it had to abandon its V2000 videocassette format—superior technically to Sony’s Beta or Matsushita’s VHS—when North American Philips decided to outsource a VHS product which it manufactured under license from Matsushita.
Over the next four decades, seven chairmen experimented with reorganizing the company to deal with its growing problems. Yet, in 2009, Philips’ financial performance remained poor and its global competitiveness was still in question. (See Exhibits 1 and 2.)

Van Reimsdijk and the Yellow Booklet
Concerned about what one magazine described as “continued profitless progress,” newly appointed CEO Hendrick van Riemsdijk created an organization committee to prepare a policy paper on the division of responsibilities between the PDs and the NOs. In 1971, their report, dubbed the
“Yellow Booklet,” outlined the disadvantages of Philips’ matrix organization in 1971: “Without an agreement [defining the relationship between national organizations and product divisions], it is impossible to determine in any given situation which of the two parties is responsible. . . . As operations become increasingly complex, an organizational form of this type will only lower the speed of reaction of an enterprise.”
On the basis of this report, van Reimsdijk proposed rebalancing the managerial relationships between PDs and NOs—”tilting the matrix towards the PDs” in his words—to allow Philips to decrease the number of products marketed, build scale by concentrating production, and increase the

3

910-410

Philips versus Matsushita: The Competitive Battle Continues

product flows across NOs. He proposed closing the least efficient local plants and converting the best into International Production Centers (IPCs), each supplying many NOs. In so doing, van
Reimsdijk hoped that PD managers would gain control over manufacturing operations. Due to the political and organizational difficulty of closing local plants, however, implementation was slow.

Rodenberg and Dekker: “Tilting the Matrix”
In the late 1970s, van Riemsdijk’s successor, Dr. Rodenburg, continued his thrust. Several IPCs were established, but the NOs seemed as powerful and independent as ever. He furthered matrix simplification by replacing the dual commercial and technical leadership with single management at both the corporate and national organizational levels. Yet the power struggles continued.
Upon becoming CEO in 1982, Wisse Dekker outlined a new initiative. Aware of the cost advantage of Philips’ Japanese counterparts, he closed inefficient operations—particularly in Europe where 40 of the company’s more than 200 plants were shut. He focused on core operations by selling peripheral businesses such welding, energy cables, and furniture making, while simultaneously acquiring an interest in Grundig and Westinghouse’s North American lamp activities.
He also continued to “tilt the matrix,” giving PDs product management responsibility, but leaving
NOs responsible for local profits. And he allowed NOs to input into product planning, but gave global PDs the final decision on long-range direction. Still sales declined and profits stagnated.

Van der Klug’s Radical Restructuring
When Cor van der Klugt succeeded Dekker as chairman in 1987, Philips had lost its long-held consumer electronics leadership position to Matsushita, and was one of only two non-Japanese companies in the world’s top ten. Its net profit margins of 1% to 2% not only lagged behind General
Electric’s 9%, but even its highly aggressive Japanese competitors’ slim 4%. Van der Klugt set a profit objective of 3% to 4% and made beating the Japanese companies a top priority.
As van der Klugt reviewed Philips’ strategy, he designated various businesses as core (those that shared related technologies, had strategic importance, or were technical leaders) and non-core (standalone businesses that were not targets for world leadership and could eventually be sold if required).
Of the four businesses defined as core, three were strategically linked: components, consumer electronics, and telecommunications and data systems. The fourth, lighting, was regarded as strategically vital because its cash flow funded development. The non-core businesses included domestic appliances and medical systems which van der Klugt spun off into joint ventures with
Whirlpool and GE, respectively.
In continuing efforts to strengthen the PDs relative to the NOs, van der Klugt restructured Philips around the four core global divisions rather than the former 14 PDs. This allowed him to trim the management board, appointing the displaced board members to a new policy-making Group
Management Committee. Consisting primarily of PD heads and functional chiefs, this body replaced the old NO-dominated International Concern Council. Finally, he sharply reduced the 3,000-strong headquarters staff, reallocating many of them to the PDs.
To link PDs more directly to markets, van der Klugt dispatched many experienced product-line managers to Philips’ most competitive markets. For example, management of the digital audio tape and electric-shaver product lines were relocated to Japan, while the medical technology and domestic appliances lines were moved to the United States.

4

Philips versus Matsushita: The Competitive Battle Continues

910-410

Such moves, along with continued efforts at globalizing product development and production efforts, required that the parent company gain firmer control over NOs, especially the giant North
American Philips Corp. (NAPC). Although Philips had obtained a majority equity interest after
World War II, it was not always able to make the U.S. company respond to directives from the center, as the V2000 VCR incident showed. To prevent replays of such experiences, in 1987 van der Klugt repurchased publicly owned NAPC shares for $700 million.
Reflecting the growing sentiment among some managers that R&D was not market oriented enough, van der Klugt halved spending on basic research to about 10% of total R&D. To manage what he described as “R&D’s tendency to ponder the fundamental laws of nature,” he made the R&D budget the direct responsibility of the businesses being supported by the research. This required that each research lab become focused on specific business areas (see Exhibit 3).
Finally, van der Klugt continued the effort to build efficient, specialized, multi-market production facilities by closing 75 of the company’s 420 remaining plants worldwide. He also eliminated 38,000 of its 344,000 employees—21,000 through divesting businesses, shaking up the myth of lifetime employment at the company. He anticipated that all these restructurings would lead to a financial recovery by 1990. Unanticipated losses for that year, however—more than 4.5 billion
Dutch guilders ($2.5 billion)—provoked a class-action law suit by angry American investors, who alleged that positive projections by the company had been misleading. In a surprise move, on May
14, 1990, van der Klugt and half of the management board were replaced.

Timmer’s "Operation Centurion”
The new president, Jan Timmer, had spent most of his 35-year Philips career turning around unprofitable businesses. Under the banner “Operation Centurion,” he lost no time in launching an initiative that cut headcount by 68,000 or 22% over the next 18 months, earning Timmer the nickname
“The Butcher of Eindhoven.” Because European laws required substantial compensation for layoffs—Eindhoven workers received 15 months’ pay, for example—the first round of 10,000 layoffs alone cost Philips $700 million. To spread the burden around the globe, Timmer asked his PD managers to negotiate cuts with NO managers. But according to one report, country managers were
“digging in their heels to save local jobs.” Nonetheless, cuts came—many from overseas operations.
To focus resources further, Timmer sold off various businesses including integrated circuits to
Matsushita, minicomputers to Digital, defense electronics to Thomson and the remaining 53% of appliances to Whirlpool. Yet profitability was still well below the modest 4% on sales he promised.
In particular, consumer electronics lagged with slow growth in a price-competitive market. The core problem was identified by a 1994 McKinsey study that estimated that value added per hour in
Japanese consumer electronic factories was still 68% above that of European plants.
After three years of cost-cutting, in early 1994 Timmer finally presented a new growth strategy to the board. His plan was to expand software, services, and multimedia to become 40% of revenues by
2000. He was betting on Philips’ legendary innovative capability to restart the growth engines. He hired Hewlett-Packard’s director of research and encouraged him to focus on developing 15 core technologies. The list, which included interactive compact disc (CD-i), digital compact cassettes
(DCC), high definition television (HDTV), and multimedia software, was soon dubbed “the president’s projects.” Over the next few years, Philips invested over $2.5 billion in these technologies.
But the earlier divestment of some of the company’s truly high-tech businesses and a 37% cut in R&D personnel left it with few who understood the technology of the new priority businesses.
By 1996, it was clear that Philips’ analog HDTV technology would not become industry standard, that its DCC gamble had lost out to Sony’s Minidisc, and that CD-i was a marketing failure. And
5

910-410

Philips versus Matsushita: The Competitive Battle Continues

while costs in Philips were lower, so too was morale, particularly among middle management.
Critics claimed that the company’s drive for cost-cutting and standardization had led it to ignore new worldwide market demands for more segmented products and higher consumer service.

Boonstra’s Reorganization
When Timmer stepped down in October 1996, the board replaced him with a radical choice for
Philips—an outsider whose expertise was in marketing and Asia rather than technology and Europe.
Cor Boonstra was a 58-year-old Dutchman whose years as CEO of Sara Lee, the U.S. consumer products firm, had earned him a reputation as a hard-driving marketing genius. Joining Philips in
1994, he headed the Asia Pacific region and the lighting division before being tapped as CEO.
Unencumbered by tradition, he announced strategic sweeping changes designed to reach his goal of increasing return on net assets from 17% to 24% by 1999. “There are no taboos, no sacred cows,” he said. “The bleeders must be turned around, sold, or closed.” Within three years, he had sold off
40 of Philips’ 120 major businesses—including such well known units as Polygram and Grundig.
Promising to transform a structure he described as “a plate of spaghetti” into “a neat row of asparagus,” he then initiated a major worldwide restructuring. “How can we compete with the
Koreans?” he asked. “They don’t have 350 companies all over the world. Their factory in Ireland covers Europe and their manufacturing facility in Mexico serves North America. We need a more structured and simpler manufacturing and marketing organization to achieve a cost pattern in line with those who do not have our heritage. This is still one of the biggest issues facing Philips.”
Within a year, 3,100 jobs were eliminated in North America and 3,000 employees were added in
Asia Pacific, emphasizing Boonstra’s determination to shift production to low-wage countries and his broader commitment to Asia. And after three years, he had closed 100 of the company’s 356 factories worldwide. At the same time, he replaced the company’s 21 PDs with 7 divisions, but shifted day-today operating responsibility to 100 business units, each responsible for its profits worldwide. It was a move designed to finally eliminate the old PD/NO matrix. Finally, in a move that shocked most employees, he announced that the 100-year-old Eindhoven headquarters would be relocated to
Amsterdam with only 400 of the 3000 corporate positions remaining.
By early 1998, he was ready to announce his new strategy. Despite early speculation that he might abandon consumer electronics, he proclaimed it as the center of Philips’ future. Betting on the
“digital revolution,” he planned to focus on established technologies such as cellular phones (through a joint venture with Lucent), digital TV, digital videodisc, and web TV. Furthermore, he committed major resources to marketing, including a 40% increase in advertising to raise awareness and image of the Philips brand and de-emphasize most of the 150 other brands it supported worldwide--from
Magnavox TVs to Norelco shavers to Marantz stereos.
While not everything succeeded (the Lucent cell phone JV collapsed after nine months, for example), overall performance improved significantly in the late 1990s. By 2000, Boonstra was able to announce that he had achieved his objective of a 24% return on net assets.

Kleisterlee’s Refocusing
By the time the Boonstra stepped down in May 2001, however, a global "tech wreck" recession had begun, resulting in what Fortune described as "a tidal wave of red ink" to greet the new CEO, Gerard
Kleisterlee, a 54-year-old career Philips man. With the share price in free fall from $60 in $2001 to $13 in 2002, Kleisterlee faced what he described as "the biggest losses in the history of the company."

6

Philips versus Matsushita: The Competitive Battle Continues

910-410

Moving quickly, the new CEO began restructuring the company, announcing the outsourcing of
Philips mobile phone production to CEC of China, and the production of VCRs to Japan's Funai
Electric. But it was not sufficient to prevent a 2001 loss of €2.6 billion compared to a €9.6 billion profit in 2000. So, over the next few years, he continued to outsource production of TVs, CD players, and components, while simultaneously moving the remaining in-house production to low-cost countries like China, Poland, or Mexico. He also sold off several businesses including most components, mobile phones, audio, and even the core semiconductor business. Within four years he had removed more than one in four Philips employees, reducing headcount by 60,000.
But Kleisterlee felt he faced a bigger challenge. "Phillips never really had any kind of strategy," he conceded. "If our engineers could make it, we would try to sell it... In 2001, we were still focused on a broad range of volatile, high-volume products like consumer electronics, semiconductors, and components. Now we are trying to create a company that generates value in a more predictable way with a portfolio that is less volatile."
The shape of that new portfolio soon became clear. Using funds generated by selling businesses,
Kleisterlee began acquiring companies in the high-growth medical and lighting segments, and began referring to Philips as "a lifestyle company” centered on health and well-being. "We came to the conclusion that the thing that holds everything together is not the fact that we made our own components and semiconductors. It's the fact that we have a common mission," he said.
A business once at the center of Philips portfolio now had a new role. "Consumer electronics is a very, very small leftover part in our lifestyle portfolio," he explained. "That business is too big a battle to fight now. We plan to be the Dell of consumer electronics, making less and marketing more. That means that we will be focused on product development, brand, and channel management."
So while Phillips continued to create innovations for its TVs, it focused them on its high-definition plasma and LCD sets with breakthroughs like Pixel Plus 2, a digital technology that refined the incoming signal to produce sharper pictures with more vivid color. But in addition to technological breakthroughs, R&D was also focused on more basic products for developing markets – hand-crank radios, high powered mixers designed for exotic foods, and irons with dust tolerant thermostats.
Phillips approach to marketing was also changing. In the developed world, it slashed the number of retail chains it serviced from 600 to 200, focusing particularly on seven giants like Wal-Mart, Tesco, and Carrefour. But in developing countries, it took a different approach. For example in India, its strategy was to sell its adapted low-end products through 35,000 village stores.
Kleisterlee explained how his adaptive product-market strategy worked: "In India, we have vans with diagnostic and lab equipment, equipped with a satellite video link to a top hospital. Instead of making long trips to a city hospital, people can now get cheaper, more convenient treatment where they live." With 700 million people in rural India, the company felt it had a great opportunity.
By 2008, Kleisterlee was ready to confirm his new focused strategy in the organization structure.
Having earlier cut the number of divisions to five (there have been 14 as recently as 1995), in early
2008, he defined just three -- healthcare, lighting, and consumer lifestyle. "We have to organize around markets," he said. "We're going to organize from the outside in."
But competition in consumer electronics remained brutal, especially in a growing global recession.
In late 2008, Philips licensed Funai to make and market TVs under the Philips name in North
America. A few months later, it extended that license to cover other markets as well as products such as DVDs, home theater, Blu-Ray, and other products. "We spent the 1980s and 1990s restructuring

7

910-410

Philips versus Matsushita: The Competitive Battle Continues

and trying to find our way," Kleisterlee said. "My goal is to leave behind a company on a successful part to steady, profitable growth." Some wondered whether he had found that path.

Matsushita: Background
In 1918, 23 year old Konosuke Matsushita (or “KM” as he was affectionately known), invested
¥100 to start production of double-ended sockets in his modest home. The company grew rapidly, expanding into battery-powered lamps, electric irons, and radios. On May 5, 1932, Matsushita’s 14th anniversary, KM announced to his 162 employees a 250-year corporate plan, broken into 25-year sections, each to be carried out by successive generations. His plan was codified in a company creed and in the “Seven Spirits of Matsushita” (see Exhibit 4), which provided the basis of the “cultural and spiritual training” all new employees received on joining the company.
In the post-war boom, Matsushita introduced a flood of new products: TV sets in 1952; transistor radios in 1958; color TVs, dishwashers, and electric ovens in 1960. Capitalizing on its broad line of
5,000 products, the company opened 25,000 domestic retail outlets -- 40% of appliance stores in Japan in the late 1960s. These not only assured sales volume, but also gave the company direct access to market trends. When post-war growth slowed, however, product line expansion and an excellent distribution system no longer insured growth, and the company looked to export markets.

The Organization’s Foundation: Divisional Structure
Plagued by ill health, KM began to delegate more than was typical in Japanese companies. In
1933, Matsushita became the first Japanese company to adopt a divisional structure. In addition to creating a “small business” environment, the structure generated internal competition that spurred each business to drive growth by leveraging its technology to develop new products. But after the innovating division had earned substantial profits on its new product, the “one-product-onedivision” policy was to spin it off as a new division to maintain the “hungry spirit.”
Management provided each division with funds to establish largely self-sufficient development, production, and marketing capabilities. Corporate treasury operated like a commercial bank, reviewing divisions’ loan requests for which it charged slightly higher-than-market interest, and accepting interest-bearing deposits on their excess funds. Divisional profitability was determined after deductions for central services and interest on internal borrowings. Each division paid 60% of earnings to headquarters and financed working capital and fixed asset needs from the retained 40%.
Transfer prices were based on the market and settled through the treasury on normal commercial terms. KM expected uniform performance across the company’s 36 divisions, and division managers whose operating profits fell below 4% of sales for two successive years were replaced.
While basic technology was developed in a central research laboratory (CRL), product development and engineering occurred in each of the product divisions. Matsushita intentionally under-funded the CRL, forcing it to compete for additional funding from the divisions. Annually, the CRL publicized its major research projects to the product divisions, which then provided funding for CRL to develop technology for marketable applications. Rarely the innovator, Matsushita was usually very fast to market—earning it the nickname “Manishita,” or copycat.

Matsushita: Internationalization
Although the establishment of overseas markets was a major thrust of the second 25 years in the
250-year plan, in an overseas trip in 1951 KM had been unable to find any American company willing

8

Philips versus Matsushita: The Competitive Battle Continues

910-410

to collaborate with Matsushita. The best he could do was a technology exchange and licensing agreement with Philips. Nonetheless, the push to internationalize continued.

Expanding Through Color TV
In the 1950s and 1960s, trade liberalization and lower shipping allowed Matsushita to build a healthy export business with its black and white TV sets. In 1953, the company opened its first overseas branch office—the Matsushita Electric Corporation of America (MECA). With neither a distribution network nor a strong brand, the company had to resort to selling through mass merchandisers and discounters under their private brands.
During the 1960s, pressure from national governments in developing countries led Matsushita to open plants Southeast Asia and Central and South America. As manufacturing costs in Japan rose, the company shifted more basic production to these low-wage countries, but almost all high-value components and subassemblies remained in its scale-intensive Japanese plants. By the 1970s, political pressure forced Matsushita to establish assembly operations in the Americas and Europe. In 1972, it opened a plant in Canada; in 1974, it bought Motorola’s TV business in the United States; and in 1976, it for him built a plant in Wales to supply the European Common Market.

Building Global Leadership Through VCRs
The birth of the videocassette recorder (VCR) propelled Matsushita into first place in the consumer electronics industry during the 1980s. Recognizing the potential mass-market appeal of the professional broadcast VCR first developed in 1956 by Californian company Ampex, Matsushita began developing the technology. It launched its commercial broadcast video recorder in 1964, and two years later, introduced a consumer version.
Subsequently, a battle over VCR format developed. In 1975, Sony introduced the technically superior “Betamax” format, and in 1976, JVC launched a competing “VHS” format. Under pressure from MITI, Japan’s industrial planning ministry, Matsushita agreed to give up its own format and adopt the VHS standard. During its 20 years of development, Matsushita’s research team lived the
VCR product cycle, moving from CRL to the product division’s development labs, and eventually to the plants producing VCRs.
Between 1977 and 1985, Matsushita increased VCR capacity 33-fold to 6.8 million units, not only to meet its own needs, but also those of OEM customers like GE, RCA, Philips, and Zenith, who decided to forego self-manufacture and outsource to the low-cost Japanese. Increased volume enabled
Matsushita to slash prices 50% within five years of launch. In parallel, the company licensed the
VHS format to other manufacturers, including Hitachi, Sharp, Mitsubishi and, eventually, Philips. By the mid-1980s, VCRs accounted for 30% of Matsushita's sales and 45% of its profits.

Changing Systems and Controls
In the mid-1980s, Matsushita’s growing number of overseas companies reported to the parent in one of two ways: wholly owned, single-product global plants reported directly to the appropriate product division, while overseas sales and marketing subsidiaries and overseas companies producing a broad product line for local markets reported to Matsushita Electric Trading Company (METC), a separate legal entity. (See Exhibit 5 for METC’s organization.)
Throughout the 1970s, product divisions maintained strong operating control over their offshore operations. They had plant and equipment designed by the parent company, followed manufacturing procedures dictated by the center, and used materials from Matsushita’s domestic plants. By the
9

910-410

Philips versus Matsushita: The Competitive Battle Continues

1980s, increased local sourcing gradually weakened the divisions’ direct control, so instead of controlling inputs, they began to monitor output -- quality and productivity levels for example.

Headquarters-Subsidiary Relations
Although METC and the product divisions set detailed sales and profits targets for their overseas subsidiaries, they told local managers they had autonomy on how to achieve them. But as “Mike”
Matsuoko, president of the European source in Cardiff, Wales emphasized, failure forfeited freedom:
“Losses show bad health and invite many doctors from Japan who provide advice and support.”
In the mid-1980s, Matsushita had over 700 expatriate Japanese managers and technicians on foreign assignment for four to eight years, primarily to play a “vital communication role.” Explained one senior executive, “Even if a local manager speaks Japanese, he would not have the long experience that is needed to build relationships and understand our management processes.”
Expatriate managers were located throughout foreign subsidiaries, but there were a few positions that were almost always reserved for them. The most visible were subsidiary general managers whose main role was to translate Matsushita philosophy abroad. Expatriate accounting managers were expected to “mercilessly expose the truth” to corporate headquarters; and Japanese technical managers were sent to transfer product and process technologies and provide headquarters with local market information. These expatriates maintained relationships with senior colleagues in their divisions, who acted as career mentors, evaluated performance (with some input from local managers), and provided expatriates with information about parent company developments.
Subsidiary general managers visited Osaka at least two or three times each year—some as often as every month. Corporate managers reciprocated these visits, and on average, major operations hosted a headquarters manager each day of the year. Face-to-face meetings were considered vital: “Figures are important,” said one manager, “but the meetings are necessary to develop judgment.” Daily faxes and nightly phone calls from headquarters to offshore expatriates were considered vital.

Yamashita’s “Operation Localization”
Although international sales kept rising, growing host country pressures caused concern about the company’s highly centralized operations. In 1982, Matsushita's newly appointed president
Toshihiko Yamashita launched “Operation Localization” to boost offshore production from less than
10% of value-added to 25%, or half of overseas sales, by 1990. To support the target, he set out a program of four localizations—personnel, technology, material, and capital.
Over the next few years, Matsushita increased the number of local nationals in key positions. In the United States, for example, American became presidents of three of six local companies, while in
Taiwan the majority of division heads were replaced by Chinese managers. In each case, however, local national managers were supported by Japanese advisors who maintained direct links with the parent company. To localize technology and materials, the company developed local subsidiaries’ expertise in sourcing equipment locally, modifying designs to local requirements, incorporating local components, and adapting corporate processes and technologies to accommodate the changes. And by the mid-1980s, offshore production subsidiaries were free to buy minor supplies from local vendors, but still had to buy key components from internal sources.
One of the most successful innovations was to give overseas sales subsidiaries more choice over the products they sold. Each year the company held a two-week internal merchandising show and product planning meeting where product divisions exhibited the new lines. Here, foreign subsidiary managers negotiated for changes in product features, quantities, and even prices for products they
10

Philips versus Matsushita: The Competitive Battle Continues

910-410

felt would better meet their local needs. Product division managers, however, could overrule the sales subsidiary if they thought introduction of a particular product was of strategic importance.
President Yamashita’s hope was that Operation Localization would help Matsushita’s overseas companies develop the innovative capability and entrepreneurial initiatives that he had long admired in the NOs of rival Philips.1 Yet despite his four localizations, overseas companies continued to act primarily as the implementation arms of Japanese-based product divisions. Unusually for a Japanese
CEO, Yamashita publicly expressed his unhappiness with the lack of initiative at the TV plant in
Cardiff. Despite the transfer of substantial resources and the delegation of many responsibilities, he felt that the plant remained too dependent on the center.

Tanii’s Integration and Expansion
Yamashita’s successor, Akio Tanii, expanded on his predecessor’s initiatives. In 1986, in an effort to integrate domestic and overseas operations, he brought all foreign subsidiaries under the control of
METC, then merged METC into the parent company. Then, to shift operational control nearer to local markets, he relocated major regional headquarters functions from Japan to North America,
Europe, and Southeast Asia. Yet still he was frustrated that the overseas subsidiary companies acted as little more than the implementing agents of the Osaka-based product divisions.
Through all these changes, however, Matsushita’s worldwide growth continued, generating huge reserves. With $17.5 billion in liquid financial assets at the end of 1989, the company was referred to as the “Matsushita Bank.” Frustrated by their inability to develop innovative overseas companies, top management decided to buy them. To obtain a software source for its hardware businesses, in 1991 the company acquired MCA, the U.S. entertainment giant, for $6.1 billion. Within a year, however,
Japan’s bubble economy had burst, and almost overnight, Tanii had to shift the focus from expansion to cost containment. Despite his best efforts, the problems ran too deep. With 1992 profits less than half their 1991 level, the board took the unusual move of forcing Tanii to resign in February 1993.

Morishita's Challenge and Response
At 56, Yoichi Morishita was the most junior of the company’s executive vice presidents when he was tapped as the new president. In a major strategic reversal, he sold 80% of MCI to Seagram, booking a $1.2 billion loss on the transaction. Over the following 18 months, under the slogan
“simple, small, speedy and strategic,” he then moved 6,000 staff to operating jobs.
Yet the company continued to struggle. Japan’s domestic market for consumer electronics collapsed—from $42 billion in 1989 to $21 billion in 1999. And the rise of new competition— from
Korea, then China—created a global glut, then a price collapse. With a strong yen making its exports uncompetitive, Matsushita’s product divisions shifted production offshore, mostly to low-cost countries like China and Malaysia. By the end of the decade, its 160 factories outside Japan employed
140,000 people—about the same number of employees as in its 133 plants in Japan. Yet management seemed unwilling to close inefficient Japanese plans or lay off staff with the commitment of lifetime employment. Despite Morishita’s promises, internal resistance prevented his implementation of much of the promised radical change.
In the closing years of the decade, Morishita began emphasizing the need to develop technology and innovation offshore. Concerned that only 250 of the company’s 3,000 R&D scientists and
1 Past efforts to develop such capabilities abroad had failed. For example, when Matsushita acquired Motorola’s TV business in the United States, the U.S. company’s highly innovative technology group atrophied as American engineers resigned in response to what they felt to be excessive control from Japan’s highly centralized R&D operations.

11

910-410

Philips versus Matsushita: The Competitive Battle Continues

engineers were located outside Japan, he began investing in R&D partnerships and technical exchanges, particularly in emerging fields. For example, an 1998, he signed a joint R&D agreement with the Chinese Academy of Sciences, China’s leading research organization. Later that year, he announced the establishment of the Panasonic Digital Concepts Center in California. Its mission was to act as a venture fund and an incubation center for the new ideas and technologies emerging in
Silicon Valley. To some it was an indication that Matsushita had given up trying to generate new technology and business initiatives from its own overseas companies.

Nakamura's Transformation
In June 2000, Kunio Nakamura, the 38-year veteran who had headed MEI’s North American operations was named president. Operating profits were 2.2% of sales, with consumer electronics generating only 0.4% due to losses in the TV and VCR divisions. Just as Morishita had promised seven years earlier, the new CEO vowed to raise operating margins to 5% in three years.
By December, Nakamura was ready to announce his first three-year plan dubbed “Value Creation
21” or VC 21. Its main objective was to build a "super manufacturing company" on three foundations: a strong technology-based components business, a flexible and responsive manufacturing capability, and customer-oriented, solutions-based businesses. The new CEO emphasized the need to retain
Matsushita’s fully integrated value chain, justifying it with a "smile curve” which promised high returns in both upstream components and downstream services and solutions to offset lower returns in the highly competitive consumer electronics products in the middle of the value chain.
At the core of VC 21 was a plan to close inefficient scale-driven plants and concentrate production in Manufacturing Centers, facilities transformed from old mass production assembly lines to modern flexible manufacturing cells. The transformation would be implemented first in Japanese mother plants, and then rolled out to the 170 plants it had worldwide.
Furthermore, as part of a plan to replace Matsushita's historic fragmented and compartmentalized structure with a "flat web-based organization", Nakamura separated plants from product divisions which now had to source their products from non-captive and non-exclusive Manufacturing Centers.
Sales and marketing was also stripped from the once powerful product divisions, and absorbed in one of two global marketing organizations, one for appliances and the other for consumer electronics.
"It was a cultural revolution," said one manager.
But the strong financial performance Nakamura had assumed would support his plan, disappeared with the "tech wreck" recession of 2001, resulting in Matsushita’s first quarterly loss in its history. The CEO and immediately announced five emergency measures to reverse the situation.
In one bold move, the company dropped its lifetime employment practice and offered early retirement to 18,000 employees. Over 13,000 accepted, not only reducing costs, but also allowing a new generation of managers to emerge. In total, the domestic workforce was cut by 25,000, and 30 inefficient plants were closed.
Despite these efforts, in March 2002 Matsushita announced an operating loss of ¥199 billion ($1.7 billion), and an even more shocking loss of ¥428 billion ($3.7 billion) including restructuring charges.
Calling the situation "an intolerable social evil", Nakamura committed to delivering profit of ¥100 billion the next year. He told his executives that because implementation of his emergency measures had not been satisfactory, he was launching a management improvement initiative. He challenged them to deliver "a V-shaped recovery" driven by V-Products – innovative, customer-focused products launched rapidly into global markets, at competitive prices. He focused 70% of investments on consumer electronics and semiconductors, urging his managers to move past Matsushita's reputation for slow innovation and imitation. "In the digital age, there is no room for imitators," he said.
12

Philips versus Matsushita: The Competitive Battle Continues

910-410

To eliminate the internal competitiveness he felt that had constrained the turnaround, he grouped all businesses into one of three closely linked domains -- Digital Networks (primarily consumer electronics, mobile phones, and telecom), Home Appliances (including lighting and environmental systems), and Components (with semiconductors, batteries, and motors.)
In March 2004, at the end of the three year "VC 21" plan, the company reported a profit of ¥185 billion ($1.9 billion) on sales of ¥7,500 billion ($72 billion). As impressive as the result was, it was still less than half of the promised 5% operating margin. So Nakamura announced a new three-year plan called "Leap Ahead 21" with a 5% objective for 2007 as an interim step on the way to 10% by 2010.

Ohtsubo’s Inheritance
In April 2006, after announcing operating profits of $3.6 billion, Nakamura announced that he would step down as CEO in June. Spontaneously, analysts at the presentation gave him a standing ovation, a unique event in reserved Japan. His successor, Eumio Ohtsubo, previously head the consumer electronics business, embraced Nakamura’s commitment to surpass Samsung's 9.4% operating margin by 2010.
Having led effort making Matsushita the world's leading plasma TV maker, Ohtsubo committed to dominating the fast-growing flatscreen market by investing Its $1.3 billion cash balance in focused
R&D and more efficient global production. He wanted to build an ability to develop, manufacture, and launch superior new products twice a year, globally. "We will absolutely not be beaten in the flatpanel TV business," he said.
In January 2008, he surprised many when he announced the change in the company's name from
Matsushita to Panasonic, reflecting the name of the company's best-known brand. It was part of
Ohtsubo’s efforts to grow overseas revenues from less than 50% to 60% by 2010. Still, with Panasonic in 78th place on the Interbrand survey of the world brand recognition, he had a way to go.
But any talk about foreign sales growth evaporated when the global financial crisis struck in 2008.
In the December quarter, company sales slid 20%, while operating profit plunged 84%. Immediately
Ohtsubo initiated a review of the company's 170 overseas plants, vowing to shut down any with operating profit of less than 3%, or declining sales over three years. The review resulted in the closure of 27 plants, and the lay off 15,000 workers. But the global crisis hit hard, and with restructuring charges and write-offs, the company projected a $4.2 billion loss for the full year ending in March
2009. The carefully laid plans for $90 billion in revenues and 10% operating margin by 2010 were vanishing dreams.

13

121

Employees at year-end (in thousands)

$35,564
2,838
9,587
9,078
35,885
20,238

219

1.2

1.6

13.9%

$33,018
1,247
-2,380
-2,510
30,549
6,611

273

1.4

1.6

20.7%

51,595

F0.0

F55,764
2,260
-2,389
-4.3%
F-4,447
-4,526
11,165
-30.2%

1990

3.62
2.15
1.68
0.94
1.40

Annual reports; Standard & Poor’s Compustat®; Moody’s Industrial and International Manuals.

Note: Exchange rate
Guilder/Dollar 1970
1980
1990
Euro/Dollar
2000
2008

Source:

Sales
Operating profit
Pretax income
Net income
Total assets
Shareholders’ equity (common)

$36,868
770
155
(260)
46,169
22,697

1.2

Current ratio

Selected data in millions of dollars:

1.9

Outstanding trade receivables in month’s sales

12.8%

Inventories as a percentage of net sales

38,541

€0.36

€0.7
33,048

€37,862
NA
3,022
8.0%
NA
9,577
15,847
60.4%

€26,385
NA
551
2.1%
NA
(178)
16,267
-1.0%

Total assets

Distribution per common share, par value F10 (in guilders)

2000

2008

$16,993
734
364
153
18,440
6,044

373

1.7

3.0

32.8%

39,647

F1.80

F36,536
1,577
N/A
4.3%
F532
328
12,996
2.7%

1980

1970

$4,163
NA
NA
120
5,273
1,747

359

1.7

2.8

35.2%

19,088

F1.70

1,280
N/A
8.5%
F446
435
6,324
7.3%

F15,070

Philips Group Summary Financial Data, 1970-2008 (Reported in millions of Dutch Guilders (F) to 1996;
Euros (€) after 1997

Net sales
Income from operations (excluding restructuring)
Income from operations (including restructuring)
As a percentage of net sales
Income after taxes
Net income from normal business operations
Stockholders’ equity (common)
Return on stockholders’ equity

Exhibit 1

910-410

-14-

Annual reports.

513

Total

--

16%
33
7
-14%
-21
-15

4,280

668
374
287
-1,915
-169
1,063
-113
-82

€5,051
14,681
2,107
-10,440
-3,030
716
1,831
37,862

2000

100%

16%
9
7
-45
-4
25
-3
-2

13%
39
6
-28
-8
2
5
100%

4,044

983
167
-157
2,233
886
--423
(805)

F 8,353
22,027
-11,562
10,714
9,425
--2,381
64,462

1995

F 7,026
25,400

100%

2,260

419
1,499
-189
-43
---218
-22

18
17
15
--4
100%

24%
4
-4
55
22
--10
(20)

13,059
8,161
---2,118
F55,764

13%
34

1990

100%

18%
66
-8
-2
---10
-1

23
15
---4
100%

13%
46

F 3,074

F 910
34
397
1,484
44
---200
6

F 7,976
16,906
6,664
17,850
11,620
---3,272
F 64,266

1985

100%

30%
1
13
48
1
---7
0

12%
26
10
28
18
---5
100%

Data are not comparable to consolidated financial summary due to restating.

Product sector sales after 1988 are external sales only; therefore, no eliminations are made; sector sales before 1988 include sales to other sectors; therefore, eliminations are made.

Totals may not add due to rounding.

Notes:

591
254
407
--336
-441
--845

€4,634
9,415
2,183
-3,984
-6,138
-4,455
28,627

Lighting
Consumer electronics
Domestic appliances
Professional products/Systems
Components/Semiconductors
Software/Services
Medical systems
Origin
Miscellaneous
Increase not attributable to a sector Operating Income by Sector:

Lighting
Consumer electronics
Domestic appliances
Professional products/Systems
Components/Semiconductors
Software/Services
Medical systems/Health care
Origin
Miscellaneous
Total

Net Sales by Product
Segment:

2003

Philips Group, Sales by Product and Geographic Segment, 1985-2003 (Reported in millions of Dutch Guilders (F) to 1996;
Euros (€) after 1997

Source:

Exhibit 2

910-410

-15-

910-410

Exhibit 3

Philips versus Matsushita: The Competitive Battle Continues

Philips Research Labs by Location and Specialty, 1987

Location
Eindhoven, The Netherlands

Size (staff)
2,000

Specialty
Basic research, electronics, manufacturing technology Redhill, Surrey, England

450

Microelectronics, television, defense

Hamburg, Germany

350

Communications, office equipment, medical imaging Aachen, W. Germany

250

Fiber optics, X-ray systems

Paris, France

350

Microprocessors, chip materials, design

Brussels

50

Artificial intelligence

Briarcliff Manor, New York

35

Optical systems, television, superconductivity, defense Sunnyvale, California

150

Source: Philips, in Business Week, March 21, 1988, p. 156.

16

Integrated circuits

Philips versus Matsushita: The Competitive Battle Continues

Exhibit 4

910-410

Matsushita Creed and Philosophy (Excerpts)

Creed
Through our industrial activities, we strive to foster progress, to promote the general welfare of society, and to devote ourselves to furthering the development of world culture.

Seven Spirits of Matsushita
Service through Industry
Fairness
Harmony and Cooperation
Struggle for Progress
Courtesy and Humility
Adjustment and Assimilation
Gratitude

KM’s Business Philosophy (Selected Quotations)
“The purpose of an enterprise is to contribute to society by supplying goods of high quality at low prices in ample quantity.”
“Profit comes in compensation for contribution to society. . . . [It] is a result rather than a goal.” “The responsibility of the manufacturer cannot be relieved until its product is disposed of by the end user.”
“Unsuccessful business employs a wrong management. You should not find its causes in bad fortune, unfavorable surroundings or wrong timing.”
“Business appetite has no self-restraining mechanism. . . . When you notice you have gone too far, you must have the courage to come back.”

Source: Christopher A. Bartlett, “Matsushita Electric Industrial (MEI) in 1987,” HBS No. 388-144 (Boston: Harvard Business
School Publishing, 1988) p. 17.

17

910-410

Exhibit 5

Philips versus Matsushita: The Competitive Battle Continues

Organization of METC, 1985

1985 Organization:

MEI

Matsushita Electric
Trading Co. (METC)

Import
Division

Industrial Goods
Export Division





China
Division

Regional
Operations

MIS and
Administration

Latin America
Division

(8)
Planning Section (2)

Finance and
Accounting

Service

Telecom
Data Processing
Office Automation
Semiconductor

Europe
Division

36 Product Divisions

North America
Division

Planning

Personnel

Asia/Oceania
Division

Middle East/
Africa Division

(50)
Manufacturing Section (5)
Video

Sales Section (40)
Appliances

Audio · · · · ·

Source: Christopher A. Bartlett, “Matsushita Electric Industrial (MEI) in 1987,” HBS No. 388-144 (Boston: Harvard Business
School Publishing, 1988) p. 23.
Note: ( ) = number of people.

18

¥9,069
527
5.8%

Net income
As % of sales

$90,949
8,424
NA
4,263
2,827
74,648
37,530

100

$68,862
4,944
1,501
2,224
941
77,233
35,767

103

290,448
143,773
50%

355
343
526

7,955
3,684

¥14.00

¥100
1.4%

¥7,299
219
3.0%

2000

$78,069
6,250
2,609
2,678
1,017
92,159
36,575

89

265,397
112,314
42%

316
296
378

8,202
3,255

¥13.50

¥90
1.3%

¥6,948
232
3.3%

1995

$37,753
4,343
2,847
3,667
1,482
49,379
20,131

159

198,299
59,216
30%

355
238
346

7,851
3,201

¥10.00

¥236
3.9%

¥6,003
572
9.5%

1990

Source:
Annual reports; Standard & Poor’s Compustat®; Moody’s Industrial and International Manuals. aData prior to 1987 are for the fiscal year ending November 20; data 1988 and after are for the fiscal year ending March 31.

Sales
Operating income before depreciation
Operating income after depreciation
Pretax income
Net income
Total assets
Total equity

In millions of dollars:

Exchange rate (fiscal period end; ¥/$)

305,828
170,265
56%

503
320
554

Capital investment
Depreciation
R&D

Employees (units)
Overseas employees
As % of total employees

7,443
3,742

Total assets
Stockholders’ equity

¥35.00

¥282
3.1%

Sales
Income before tax
As % of sales

In billions of yen and percent:

2008

Matsushita, Summary Financial Data, 1970-2000a

Cash dividends (per share)

Exhibit 6

$24,890
3,682
2,764
3,396
1,214
21,499
10,153

213

175,828
38,380
22%

288
227
248

5,076
2,084

¥ 9.52

¥216
4.1%

¥5,291
723
13.7%

1985

$13,690
1,606
1,301
1,520
584
11,636
5,129

213

107,057
NA
NA

NA
65
102

2,479
1,092

¥7.51

¥125
4.3%

¥2,916
324
11.1%

1980

$4,572
317
224
273
105
4,206
1,890

303

82,869
NA
NA

NA
28
51

1,274
573

¥6.82

¥32
2.3%

¥1,385
83
6.0%

1975

$2,588
NA
NA
408
195
2,042
900

360

78,924
NA
NA

NA
23
NA

735
324

¥6.21

¥70
7.6%

¥932
147
15.8%

1970

910-410

-19-

--

-2,175
817
1,618
-¥7,682
¥3,698
3,601

--1,399
123
¥9,069

¥6,789
5,404
(3,120)

51%
49

28
11
21
-100%

--

--

---

18
--

1,306
--

1,316
--

23%
--

¥1,706
--

Notes: Total may not add due to rounding.

Source: Annual reports.

--

2000

¥4,319
---

2008

¥3,455
3,493

---530
¥6,948

374

1,797

-916

¥1,827
893

--

1995

50%
50

---8
100%

4

26

-13

26%
13

--

¥3,382
2,621

---573
¥6,003

312

1,375

-802

¥2,159
781

--

--

56%
44

---10
100%

5

23

-13

36%
13

FY 1990

Matsushita, Sales by Product and Geographic Segment, 1985-2000 (billion yen)

By Product Segment:
Audio, Video, Communications
Networks
Video and Audio Equipment
Electronic components
Home appliances and household equipment
Home appliances
Communication and industrial equipment Batteries and kitchen-related equipment Information and communications equipment
Industrial equipment
Components
Others
Total
By Geographic Segment:
Domestic
Overseas
Corporate

Exhibit 7

¥2,659
2,632

---372
¥5,291

217

849

-763

¥2,517
573

--

FY 1985

50%
50

---7
100%

4

16

-14

48%
11

--

910-410

-20-

Similar Documents