Kei(tai)retsu

Back to Contents of Issue: January 2002

by Richard Meyer

ONE PARTICIPANT DESCRIBES THE meetings as "underground": informal, face-to-face, and secret, bringing together executives and researchers from two of Japan's largest corporations. In public, the firms are fierce competitors, engaging in ruthless and expensive marketing campaigns, clawing customers from each other, and battling for sales at every turn. But at these meetings the rivals are good friends. They have known each other for a solid decade and share a bond developed over years of working to tackle similar technical problems.

Cooperation among Japanese competitors is not usually particularly noteworthy. It has been happening for ages and will no doubt continue to occur. But this example of collaboration is of particular significance. The participants in the cabal are NTT DoCoMo and J-Phone. These are not old industrial enterprises steeped in the ways of traditional Japan, manufacturing tatami (straw) mats, steel, or vegetable oil. They are new companies making the products of the future. The meetings do not involve hardened company employees drawn together by a long tradition of collusion. They involve executives from Japan's most progressive and promising industry. DoCoMo and J-Phone are leading the country forward both technologically and economically, but they have not completely abandoned the practices of the past.

The secret meetings between J-Phone and DoCoMo are just one example of cooperation in the field of wireless communications in Japan. The operators of the networks not only work closely with each other, they also work very closely with the manufacturers of the phones they sell. They set specifications and guide the handset makers, acting much as a parent company would toward a wholly-owned subsidiary, and are involved to a great extent in the product development process. The manufacturers themselves work with each other, agreeing on standards, co-developing technologies, co-investing in companies, and forming formal alliances.

"Japan is different from any other place in the world, and that has a very serious telecom aspect," says Johan Hjelm, senior specialist at Nippon Ericsson. "DoCoMo sets the direction. And the manufacturers work together."

Unlike the "underground" meetings, most of the cooperation in the mobile phone industry takes place in the open. Little secret is made of the way companies in the industry work together, and much of what happens between manufacturers, especially the alliances and investments, is a matter of public record. For many, unity is considered a strength and is something about which they eagerly boast. J-Phone, for example, issues glossy marketing materials to explain how its close association with the makers helps it develop new products.

"The network operators and handset makers discuss design, function, and what sort of services are required," says Keiji Takao, deputy general manager of J-Phone's 3G Mobile Products Group. "This is different from America and Europe."

Alone, none of these instances of corporate closeness is particularly dramatic. But when seen together, a complex and large system, integrated both vertically and horizontally, emerges. If lines are drawn between the companies in the industry -- dotted, dashed, and solid, depending on the strength of the association -- it quickly becomes clear that the wireless industry in Japan is not much different from Japan's more traditional industries. It has the potential to move very much as one, driving relentlessly, with the power of many, in a chosen direction.

Technology, we have been told, is supposed to bring change. Faster and better communications, increased data storage, longer battery life, and all the other wonders emerging from the laboratories of the world should by all accounts have significantly altered the way society and the economy work. In Japan, this should especially have been the case. The entities that have long dominated the economic landscape of the country -- the cartels, the keiretsu, and the reincarnated zaibatsu -- were supposed to have been undermined as the Internet and wireless communications rendered these anachronisms ineffective. The ease with which transactions could be made was seen transforming connections and cross-shareholdings from assets into burdens. In the time it takes to go through nemawashi (generating consensus in every layer of a company group) and deal with associated companies, an Amazon-like competitor can capture a market and go onto the next. Technology, in the minds of many, should have undermined Japan's rigid corporate hierarchy and torn corporate alliances apart.

In many respects, the ways of corporate Japan did change. The group structures have been weakened, and some would argue that they have, for all practical purposes, become meaningless. Most still exist in name, but they are no longer the economic engines of Japan. The words Mitsubishi, Mitsui, and Sumitomo no longer inspire awe; in fact, they barely generate any reaction these days. At the same time, new entities have formed, which are held together by technology, not threatened by it.



Today, according to Shingo Kasahara, an official in the International Affairs division of the Fair Trade Commission, knowledge is more important than capital. Industry is not based so much on factories and heavy equipment, but on services, software, and intellectual property. As a result, groups are now bound together by patents, joint research agreements, and standards bodies. Links are formed with little more than contracts and conversation. Gone are the days when companies swapped directors and shares to benefit from each others' strengths. No longer do they have to engage in large public transactions to enjoy the power of other corporations. Nothing needs to change hands but words and ideas.

"The type of concentration is different between now and 10 or 20 years ago," says Kasahara. "A few decades ago it was based on financial relationships, and you had structures like the zaibatsu. The relationships are different now. Cooperation has changed from financial to intellectual."

On a white board in the offices of Jupiter Media Matrix, a telecommunications and Internet consultancy, senior analyst Fumitaka Okumura sketches a diagram of Japan's mobile phone industry. On the left he writes DoCoMo, J-Phone, Tu-ka, and Au for the network operators. Moving right, he adds the two major content brokers, Cybird and Index, which act as intermediaries between the operators and the companies that offer what is displayed on i-mode and i-mode-like phones. Further right lie the content producers themselves and then the advertising agencies. Off to the side are the handset makers. Lines and arrows crisscross Okumura's sketch, linking the companies represented there.

Okumura dubs the structure a "wireless platform business model" and believes that it is very powerful. He agrees, however, that the moniker he has chosen is rather tortured and that the system represented on his white board should be known as something else. He insists that calling it a keiretsu is not strictly accurate, as the traditional keiretsu -- which literally means "line of succession" -- are held together by financial relationships, but he cannot offer anything more appropriate. He invites a visiting foreign journalist to take a shot. After some thought and at a later date, the word comes to the journalist: keitairetsu, combining the Japanese word for mobile phone and the last character of keiretsu.

The potential power of such a grouping is immense. DoCoMo alone is massive. As of March 31, 2001, it was the largest company in Asia in market capitalization and the 11th largest in the world. It was just a few billion dollars less valuable than IBM and worth about $10 billion more than Intel. The other operators and the Japanese electronic companies -- such as $42 billion-in-sales NEC, $65 billion-in-sales Matsushita, and $62 billion-in-sales Sony -- add significant intellectual, marketing, and management might to the keitairetsu. Working together, developing standards and designs, sharing resources, and harmonizing strategies, the group members have the power to advance technology farther and expand the market faster than any single company.

DoCoMo clearly lies at the center of the grouping. It is not only large in an absolute sense. It is also large in a relative way. The company controls 60 percent of Japan's wireless market, giving it considerable power and influence within the industry. Its direction and strategy, backed up by a huge balance sheet and a large customer base, are almost impossible to resist. The handset makers and the other operators are compelled to follow DoCoMo's chosen path.

"DoCoMo leads and controls the industry," says Okumura, a former internal consultant for the NTT Group. "There are many players under its umbrella."

J-Phone claims numerous victories over DoCoMo, some significant. It says it was the first mobile operator in Japan to offer short messaging and Internet email services (1997), Web browsing (1998), and Java downloading (2001). But advances by others only seem to help DoCoMo. It can always match and beat competitors by dint of its economic power, assimilating and implementing new technologies or product ideas with disproportionate success. J-Phone, for example, beat DoCoMo to market with an i-mode-type phone. DoCoMo just followed and won. Its i-mode not only attracted more customers than the J-Phone equivalent and became the standard, but it also become the generic name by which such phones are known. Challenges are consumed by DoCoMo like vitamins, making it more powerful and placing it back in the lead.

"The main point is that DoCoMo has dominant power. You only need to look at DoCoMo to understand the mobile phone market," Okumura says. He calls DoCoMo a yokozuna, or Sumo grand champion. "J-Phone is a follower," he adds. "Polite people would say it is a good follower."



The power of the keitairetsu is obvious even to the naive observer. When visitors arrive in Japan, they are at first impressed by the phones that the grouping has been able to design and produce. The handsets are smaller and seem generally better than phones outside of the country. They have capabilities that are not even contemplated in the US and Europe. The industry's technological and manufacturing strengths translate into clear business success as well. Nokia, for example, may be able to control a third of the mobile phone market worldwide, but it sells very few phones in Japan.

After a while, visitors may also notice the negative side of economic concentration and cooperation. Many are struck by the lack of diversity in the styles and functions available. Regardless of the manufacturer or the operator, new features seem to emerge as if the same company is making the phone. Differentiation is trivial. Most will also soon realize that while the consumers appear to benefit from the industry's advanced technology, they are paying dearly for it as well, being charged for services and enduring high usage fees. Some may even notice the lack of style in Japanese phones. A clunky old Nokia 3210 handset looks downright artistic when compared with the local models in Japan, which seem like toys. They are packed with features, but appear to be loaded up with more enthusiasm than thought.

The keitairetsu manifests itself visibly just south of Tokyo at the Yokosuka Research Park. There, in a valley below an old hilltop research station set up by NTT, DoCoMo's parent, DoCoMo has established research facilities for its new mobile phones. More than 60 other companies have set up offices or laboratories in the park, mainly to be close to DoCoMo. They say that physical contact with the company and informal face-to-face exchanges with its engineers are vital to their work in the Japanese mobile phone market. Those congregating there are not the fringe elements of the industry or even simply a representative sampling from it. The list of tenants reads like a Who's Who of the mobile universe. All major phone makers, with the exception of those from Korea, have been drawn to the research park. Companies that make and sell about 90 percent of the world's mobile phones are camped out there.

Sony, perhaps more than any other company, knows the power of the keitairetsu and the need to be as close to it as possible. It has experienced difficulties being even slightly distant from the center of the conglomeration. For most of the mobile boom, the company was a relative outsider. According to Kanji Ohnishi, deputy general manager of the business promotion department of Sony Ericsson, two kinds of mobile phone manufacturers exist in Japan. First, there are the "Mova" makers: NEC, Matsushita, Fujitsu, Mitsubishi Electric, and, more recently, Sony. These companies are involved in the development of DoCoMo phones. As a result, they know with a great degree of certainty what is needed to make the phones before specifications are formally published. The group still operates today, and these companies still help decide how the phones are to be built. They are part of the inner circle. Sony was long outside the group and has traditionally received the specifications for new phones only when details were made public. "We were the readers of the text," says Ohnishi, "rather than the writers."

Sony was what is known as a "by" maker. When a "by" maker manufactures phones, DoCoMo may purchase them, but the phone will be co-branded with the manufacturer. For example, phones made by Sony were badged "by Sony." And of course, Sony and the other "by" makers would get the specifications for the new phones after the "Mova" makers.

Ohnishi won't say that being a "by" maker has been at the root of Sony's relative failure in the mobile industry, but it is notable that three of the four original "Mova" makers are the leading manufacturers of mobile phones in Japan in terms of market share. Sony recently became a "Mova" maker, and the company believes that its sales will improve from here on out.

"'Mova' is the major leagues," Ohnishi says. "'By' is the minor leagues." The power and dominance of DoCoMo and the unity of the system should not be exaggerated, however. While the structure is massive and some of the relationships are tight, the Japanese mobile communications industry is not a corporation, by any means, pursuing a single identifiable strategy. The companies within it, like companies all over the world, are independent and opportunistic. Just as with the old keiretsu, the keitairetsu is as fragmented, divisive, and very often as competitive as it is cooperative. During the meetings between DoCoMo and J-Phone, for example, the most important elements of company strategy are not revealed. The competitors talk about standards issues, such as how to make their displays more compatible so that content makers have an easier time programming for the operators. But when it comes to innovation, invention, and new developments, the companies keep secrets from each other. When J-Phone began offering its photo-taking and photo-emailing phone (called sha-mail), DoCoMo executives were shocked and expressed surprise during one of the underground meetings. "DoCoMo has no idea what is coming next from us," boasts Takao of J-Phone.

Sony complains that the industry is so diverse that making phones for it is difficult. While the handsets from different companies may look the same to consumers and function in a nearly identical manner, their guts vary greatly from one operator to another. This means that the manufacturers have to design and make three different kinds of phones if they want to serve the entire Japanese market, and the content makers have to design their content to work on three different systems. KDDI has made matters more complicated by choosing a different network technology for its new phones from that chosen by J-Phone and DoCoMo. The diversity is so burdensome that some makers choose not to supply all the operators -- Toshiba manufactures only for KDDI and J-Phone; Sharp produces only for J-Phone.

When formal cooperation does occur, it tends to be fairly narrow and focused. Toshiba and Matsushita have a joint venture related to screens; Sony and DoCoMo are developing systems for music downloading. And cooperation doesn't always go very smoothly. According to one executive in the industry, cooperation between Matsushita and NEC on mobile phones has only been worked out in principle, and the companies are having trouble getting beyond a broad agreement. The executive says the companies have very different personalities, and it will take time to get them working together successfully.

It is also important to recognize that cooperation in the mobile phone business is not a trait peculiar to the Japanese market. Collaboration occurs everywhere in the development of standards and systems. The international bodies working on Wireless Application Protocol and third generation phone specifications, for example, get companies talking quite openly about their products and plans. Cooperation, say industry executives, also results from the fact that companies in the industry have a great deal of contact with each other in the normal course of business. Most make and sell components or software that are used in competitors' products. Most also hold patents that their competitors need. The industry, by nature, is fairly tight.

"Japanese companies work together," says Heikki Tenhunen, the head of Nokia Mobile Phones Japan. "That is true. On the other hand, Nokia, Ericsson, and Motorola also work together."

Nor does Japanese cooperation appear to harm the prospects of those not cooperating. While it is good to be close to the center of the keitairetsu and participate in the development of it, it is hypothetically possible to be completely removed from it and still do business successfully in Japan. Patents from Japanese companies are available at a uniform rate, procurement practices are open and transparent, and the companies here are eager for new technology.



Foreign companies trying to sell mobile phones in Japan say they have not felt as though they were being discriminated against. They are allowed to set up in the Yokosuka Research Park, and they say that treatment there by the Japanese companies is excellent. Foreign companies have also been able to form alliances with Japanese companies. Indeed, foreign companies appear to have more alliances with Japanese companies than Japanese companies have among themselves.

"We haven't been strong here, but that's not because of problems with cooperation. It is not a sign of bad cooperation locally," says Tenhunen of Nokia.

The keitairetsu does not have clearly defined borders, with companies, technologies, and people either in or out. It is not a set entity with a hard-wired command and control structure and a rigid hierarchy. It is a complex organism, with an evolving core and layers of affinity more than lines of demarcation. DoCoMo does not just send out orders to associated companies, which are then obeyed and enthusiastically implemented. It cannot simply rally the full force and collective resources of the group. It merely draws the industry in a certain direction, influencing some companies more than others and inspiring mixed degrees of commitment from those who follow. Allegiance is variable and selective, the keitairetsu existing only when and to the extent that members want it to. Its effectiveness is thus mixed. At times the companies of the Japanese mobile communications industry march in unison toward a goal and achieve that goal with military efficiency. i-mode and i-mode-like phones, for example, were created, adopted, implemented, and marketed with remarkable speed and effectiveness. The technology has not only generated significant business and been the envy of the telecommunications world, but it has also been a potent weapon against competitors, both mobile and fixed-line. At times, however, the keitairetsu fails and fails big.

Japan's efforts with regard to third generation phones are one example. Rather than storming the future of mobile telecommunications, the companies of the keitairetsu are creeping or stumbling ahead. Their third generation phones are arriving late, and when they have finally arrived they have not only functioned poorly, but demand for them has been weak. To make matters worse, reliability has become a problem in Japan's mobile phone industry. Despite the awesome engineering and manufacturing abilities of the handset makers and the financial might of the operators, the industry faced a series of major recalls after glitches developed in handset software and batteries. Sony was the hardest hit, with 1.1 million defective phones being pulled back since May. Matsushita has also recalled phones. "We don't have enough software engineers," complains Sony's Ohnishi. "Developing software is difficult. Testing is also difficult. There are 100,000 items to check on a 2G phone and 500,000 on a 3G."

Perhaps most devastating for Japan's mobile phone industry has been its failure, despite its size and structure, to lead the world in creating a standard. It has not even been able to hold on to its own. It was first blindsided by the global standard for mobile communications (GSM), when American companies decided to team up with the Europeans in a conscious effort to block Japan's overseas expansion in mobile communications. Then the Japanese industry was leapfrogged when code division multiple access (CDMA) technology proved successful in mobile phone applications. Regardless of the keitairetsu's power, it was not able to beat a government-led initiative from Europe, the origins of GSM, or the lateral-thinking risk-taking scientists at Qualcomm, the company that developed the basic patents for CDMA use in mobile phones.

Still, the keitairetsu, like its predecessors, is a powerful structure that only seems to be getting more powerful over time. Challenges do not break the group apart; they drive member companies closer together. Recession, failures, and mistakes motivate them to share more technology, forge more alliances, and work more closely to develop phones. Bonds that may have been weak -- those that would be represented by dotted lines on a chart -- tighten and become dashed or even solid. This is what happened with NEC and Matsushita, which together control an estimated half of the handset market in Japan. The two companies, which before were somewhat distant relatives on the keitairetsu family tree, chose to work together because of the great difficulties they are facing in the development of software. Adversity breeds unity.

"They had it up to here," an industry executive says about NEC and Matsushita, with his hand at neck level.

The greatest risk to the keitairetsu is regulatory. The ties that bind the group are potentially anti-competitive and could run afoul of Japan's anti-monopoly law. While it would be easy to argue that any of the arrows on Okumura's white board represent no more than standard business relationships, in most cases the companies are engaged in more than arm's length transactions. It is clear that what is going on between them is not just buying and selling, customer to company.

The Japan Fair Trade Commission (JFTC) says it is paying very close attention to the telecommunications industry in Japan and has a special unit dedicated to looking for anti-trust infractions within the information technology industry. The government has already considered breaking up companies within the NTT group in order to foster more competition. If DoCoMo were dismantled, the keitairetsu could lose much of its direction. If cooperation were challenged elsewhere in the industry, the keitairetsu could also be threatened.

Plenty of people would like to see DoCoMo humbled and no doubt are secretly cheering its 52 percent plunge in profit during the half-year through September. Some say that the company is arrogant and has too much power. Critics include the parent company, NTT, which has found its subsidiary distant and aloof. Consultants complain that the company does not provide them with information; they are told that everything they need is on the company's Web site. Some journalists find the company unresponsive as well. (It refused to speak officially for this story, for example.)

Particularly frustrating is the fact that DoCoMo has done very little to deserve what it has, other than to exploit its inherited power. According to industry observers, DoCoMo has very little of its own technology, and most of the basic patents for its mobile phones reside elsewhere. It has done little more than coordinate the industry, they say, exploiting the strengths of others. "Success of the system is a result of efforts by these guys," Okumura says as he points to the content and handset makers on his white board. "DoCoMo has no ideas and not so many patents."

He adds, "People are dreaming about how DoCoMo will be broken up."

They are likely to be dreaming for a long time. Even if DoCoMo and the system it leads is violating the anti-monopoly laws of Japan, the company and the structure that has developed around it can survive intact. When it comes to technology, the government weighs damage against benefit. If the harm of an anti-competitive arrangement is seen to be outweighed by the good it brings, then it will be allowed to stand.

"An alliance may lessen the number of competitors and limit selection, that is true," says the JFTC's Kasahara. "On the other hand, it may reduce costs and promote innovation. There is merit and demerit. If the balance goes to the [demerits], it may violate the anti-monopoly law. Judgment is made case by case."

The JFTC does not take aim at the structures present in the industry. Links between companies, even competitors, are not enough to trigger action. The commission only acts when it finds clear cases of harm to consumers. It offers two examples that illustrate how it has viewed various potential anti-trust violations in the industry. In one case, DoCoMo had handset makers agree not to use DoCoMo technology for a six-month period in phones they made for other companies. The commission questioned that restriction, saying it may be anti-competitive. As a result, DoCoMo changed the contract with its suppliers, deleting the clause requiring them to withhold technology. In another case, 19 companies involved in the development of third generation phones created a patent pool, whereby they shared their intellectual property. The commission investigated but found that since the resources of the pool were made available to non-pool members, it was not a violation of the act.

The Japanese attitude toward anti-trust is not unique and matches very closely the prevailing views in other jurisdictions. The US Department of Justice, for example, also believes that technology may require special application of anti-trust laws, so the Japanese acceptance of the keitairetsu will not likely face diplomatic challenges. And it is very difficult under any anti-monopoly statute to punish a company for market domination alone. Damage must invariably be demonstrated.

Like the keiretsu, the keitairetsu can exist comfortably within the limits of the system. The old concentrations made sure they kept their associations legal, staying under shareholding limits, maintaining distance in corporate relationships, and keeping the purpose and aims of their meetings fuzzy. At any single point, the old system was innocuous: a golf outing here, a conference there, and a few shares in the vault. Similarly, the keitairetsu's contracts, standards meetings, and alliances -- while they can result in immense power -- fit neatly into Japan's legal structure and can survive most bureaucratic challenges. @

Note: The function "email this page" is currently not supported for this page.