Post of the Day
August 11, 2000

Board Name:

Posts selected for this feature rarely stand alone. They are usually a part of an ongoing thread, and are out of context when presented here. The material should be read in that light. How are these posts selected? Click here to find out and nominate a post yourself!

Subject:  The New Media-Keiretsu
Author:  Ubermensch1

Since there has been some vocal interest in the media-keiretsu model I posted about many moons ago, I've decided to offer up a sneak peak...a teaser, if you will...of a more comprehensive industry report I'm currently working on. The portion I'm posting deals with what I call the "New Media-Keiretsu" and is available, in all its graphically-enhanced, table-enriched splendor, here:





With the rapid evolution of the Internet into the first new mass medium in over 50 years, the build-out of a new infrastructure for media delivery is in full swing. Global deployment of broadband technology carries with it the promise of a paradigm shift that will forever transform the way we access information and media. There is a land rush underway as we speak, to fill the vacuum created by the gap this shift leaves behind. These unique opportunities springing from the fountainhead of the Internet have served as a catalyst for transition within the media delivery industry, which has undergone a series of its own shifts and transformations to mirror the larger trend. Organizations responsible for creating, delivering, and facilitating consumption of media content have had to quickly redefine their positions within the industry and relative to each other. What we have seen as a result is a pattern of increasingly synergistic relationships forming between media-related companies, the main purpose of which is to protect their existing domain while trying to take advantage of the transition through aggressive expansion. This pattern, along with a trend of accelerated consolidation, has given rise to the formation of tight alliances, both formal and informal, which in turn form the basis of what may be described as an emerging class of "media-keiretsu". This phenomenon was first observed by Stephen Auditore, founder of Zona Research, in 1996. A lot has changed since he first sketched the implicate order that lies beneath the surface, but his central thesis has been born out. There are currently two well-formed and well-differentiated media-keiretsu staking out their respective territories in this new frontier. An examination of the implications is in order.


For those not familiar with the concept of the keiretsu, a hallmark of Japanese business, here is a brief summary:

In the Japanese business world, a keiretsu is a structural formation of business entities who all work together in a very closed, cartel-like fashion. In Japan there are six major keiretsu, each organized around a major bank or financial institution which is the heart of the keiretsu. Those six are the Mitsui Group, the Mitsubishi Group, the Sumitomo Group, the Fuyo Group, the Sanwa Group, and the Dai-Ichi Kangyo Group.

There are many reasons why these keiretsu have formed. One of the original reasons was to thwart foreign takeover of Japanese companies, and this was accomplished through share interlocking - in which keiretsu members held equity shares of each other's companies (*1). A second reason was that banding together stabilized the value of the participant companies, since a significant number of shares were held by parties that valued the shares based on their own financial requirements and those of the company being held, instead of a public equity market (*1). These reasons don't explain the formation of the new media-keiretsu, but nonetheless the resulting effect is much the same, as many member companies in the media-keiretsu own varying stakes in one another.

In Japan, keiretsu come in two forms: horizontal and vertical. The horizontal keiretsu is a cloud of companies that cluster around a center core, which is usually a large bank. One way to think about the arrangement is to liken it to a solar system, with numerous planets orbiting around a sun in the middle. These companies cooperate and tend to compliment each other quite well. The vertical keiretsu traditionally denotes the relationship between a manufacturer, its suppliers, and its distribution chain. The motivation for these keiretsu is to gain more control over both the supply and sell sides of the manufacturing process (*1). The common visual analogy is one of an hourglass, where component/material suppliers on top sell to the manufacturer in a keiretsu. The center firm is responsible for overseeing fabrication of the end product, which is then spread out through the firm's distribution partners on the bottom. The vertical keiretsu arrangement provides price and sales stability to the manufacturing entity at the center of the keiretsu while potentially minimizing customer choices by locking up the distribution channel for similar products (*1).


As Auditore first pointed out, the Internet infrastructure has allowed for the rise of a new strain of global keiretsu - the media-keiretsu - global, in that they are comprised of organizations from all over the world and they conduct business on a global scale. Auditore notes that when information movers partner with content creators and consumption enablers, a grouping of vested interests based on media delivery emerges. When the relationships become based on integration of content, technology, marketing, and often co-investment, a media-keiretsu emerges. The same thing can be accomplished through consolidation, also. The most recent such development is the merger of AOL and Time Warner, and the next is likely to be between Vivendi and Seagram. These are profoundly important developments in content and technology integration, however these two examples do not represent entire media-keiretsu in and of themselves. A true media-keiretsu consists of a group of companies who all share a common central "core" just like their traditional Japanese counterparts. The core company, the one that occupies the center spot in the keiretsu, must be a critical enabler of its fellow members. Just as the banks provide financial and managerial support for member companies in traditional Japanese keiretsu, so must the core company of a media-keiretsu provide support for the overarching goals shared by its allies.

Auditore describes the media-keiretsu, in the simplest terms, as consisting of organizations involved in the flow of information and content. This process begins with information/content creation, moves through delivery, and ends with consumption. This is the basic three-part model of media delivery, as well as the structure of the media-keiretsu: Content/Delivery/Consumption. Member companies fall into one (or perhaps more) of these three facets of media delivery. In addition to (or as a function of) being an enabler for companies in each area, the core company is the one that ties all other members together and facilitates the pursuit of the common goal. So where should we look for core companies?

Auditore argues that the delivery area will provide the core of the media keiretsu:

(1) One key attribute of the [media]-keiretsu is that they operate in multiple regulated regions. The content mover adds a harmonized platform for delivery, regardless of the local regulatory environment, which creates more potential consumers for content creators, expanding markets beyond political and geographic borders and beyond regulation.

(2) The [media] deliverers have had to invest in the physical facilities needed for information delivery. This includes switching technology and transmission technology (e.g., wiring, fiber, satellites, etc.). Deliverers have the least flexible investments of any keiretsu members.

The situation faced by content delivery companies is akin to that faced by the first railroad companies that were building lines through unpopulated areas of the U.S. When railroads were first constructed across the U.S., railroad companies needed traffic to make the investment in the railroad worthwhile. To create traffic, they spurred development along their right-of-way. Central to creating this demand for traffic was a concept called recolonization. The railroad companies created recolonizing departments whose function it was to recruit immigrants to relocate and colonize the land along the railroad routes.

The railroad companies' strategy to create traffic is not significantly different than the current day situation in which info/content movers (bandwidth providers) are stimulating the development of content and consumption, either by themselves, but more likely with partners, in an effort to increase bandwidth use. This need to stimulate the use of bandwidth has been a driving force in the formation of [media]-keiretsu.

But delivery companies have not colonized the market for higher bandwidth by themselves. The simple "If you build it they will come" scenario does not translate into particularly effective strategy. In fact more effective have been the pioneering efforts of consumption enablers, who have managed to convince content providers that a potential market exists and that compelling content will draw consumers. Simply put, each area's role in fleshing out the market is inseparably intertwined with the others. The need for delivery and distribution is driven by the availability of content, which in turn is driven by consumption-enabling technology, and consumption-enablers in turn are dependent upon reliable, adequate delivery and the content provider's willingness to do their part. Considering the interwoven nature of all three areas then, it would seem the most likely candidate for core company in a given media-keiretsu is the one whose sphere of influence extends furthest. So we can say that a fairly reliable identifying feature of a core company is its presence in all three facets of the media delivery model.


In 1996, Auditore described the existence of five notable media-keiretsu. Since that time, three of those have subsequently gravitated toward and merged with one of the other two. The two giants that now remain are vying for control of the industry.

These two amalgamated titans are detailed below. You will note that there are many other media and network-oriented companies that have not been listed. This is not meant to imply that they are not significant players within the industry�simply that it is impossible to provide an exhaustive list.

Media-keiretsu #1

Core Company:
> RealNetworks

> (Content) Time Warner, Seagram/Universal, Disney/ABC, Viacom, Sony, CBS, Associated Press, Webglide, DoubleClick
> (Delivery) AT&T, Sprint, GTE, Deutsche Telekom, France Telecom, HongKong Telecom, Enron, Madge, Sun Microsystems, Inktomi, @Home, Loral Space/Communications, SES/Astra, PanAmSat
> (Consumption) AOL, RealNetworks, IBM, Excite, Earthlink, Nokia, Psion, Motorola, Ericsson, Matsushita, Liberate, Iomega


This is undeniably the biggest and strongest of the two keiretsu at this point in time. The group is well established globally; they currently dominate the North American and European continents and are poised to open up China and Southeast Asia.

Within this group's stable are several of the world's largest telecommunications companies. Add companies pursuing terrestrial and wireless broadband initiatives, and the result is a whopping agglomeration of delivery infrastructure providers.

Similarly, the group boasts several of the largest and most diversified media companies in the world, though Time Warner and Disney's relationship may be described as tenuous at the moment. Disney has expressed concern over the possibility that AOL/Time Warner will lock up the distribution channels for cable television programming. A dispute between the two earlier this year resulted in Time Warner briefly pulling ABC from its cable systems. It is unclear at this time how the relationship between Disney and Time Warner will unfold, but the two may try to distance themselves from one another. If their differences are not resolved there is a chance that Disney will unexpectedly switch allegiances and move to keiretsu #2, but in all probability RealNetworks' strength as an enabler will be enough to keep them in this group.

If the merger between AOL and Time Warner is approved, they, along with RealNetworks, will have a presence in all three facets of media delivery. They could emerge as a kind of "co-captain" of this keiretsu, sharing a portion of RealNetworks' core position, but are unlikely to displace RealNetworks from the center, as they are not really an enabler of the other group members and thus do not have the same ability to hold them all together in the face of diverging interests.

Company to watch is AT&T. The company has relationships with Keiretsu #2 members British Telecom and Microsoft. They presently show no signs of switching their membership but should be watched closely as future circumstances develop, as this could change. They could also move toward a spot on the Independents list.

Media-keiretsu #2

Core Company:
> Microsoft

> (Content) GE/NBC, News Corp., BMG, EMI
> (Delivery) British Telecom, NTT, Nextel, 3COM
> (Consumption) Yahoo, Compaq, Intel, Samsung, Qualcomm, WebTV


Microsoft is striving to compete with keiretsu #1, and has had some successes. Yahoo (through has a firm relationship with Microsoft, and the web portal company is currently developing a co-branded media player with the software giant.

BT/MCI provides this group with a wide reach of connectivity. NTT has been rather slow to join the group, but has recently consummated its relationship with Microsoft on both the wireless and terrestrial networking fronts. The inclusion of NTT adds more of a global face to what was primarily an Anglo-centric group.

Company to watch is British Telecom. They have strong ties to Microsoft, but also have a relationship with AT&T, and MCI is in the process of merging with Sprint. They show no signs of switching their membership at present, but should be watched closely, as this may change in the future. They may also move toward a spot on the Independents list.

The Independents

While it would be nice and neat if all media-delivery companies fell into one keiretsu or the other, that's not the case. There are several notable independents not yet firmly married to a keiretsu. These companies tend to have leanings toward one keiretsu or the other, but have yet to decidedly join a group. Below is a list of some of the more prominent companies:

Akamai - Provider of edge server caching systems. Has relationships with both keiretsu, but may be leaning slightly toward keiretsu #2. Their products and services are not well differentiated from those of other companies in the same industry (namely Inktomi) and therefore may ultimately be unable to co-exist in the same group.

Apple - Computer and software company. Partially owned by Microsoft, and thus would seem to be beholden to keiretsu #2. However the company has their own media format called QuickTime, which competes with Microsoft's Windows Media format. Apple does not really have the relationships to qualify for their own keiretsu, and instead actually seems to be leaning slightly toward keiretsu #1, having recently announced a supportive deal with RealNetworks. Nonetheless, it is extremely unlikely they will ever officially join that group.

Cisco - Internet network infrastructure provider, and the quintessential independent. The company has relationships with members of both keiretsu, including RealNetworks, Microsoft, Sun, and IBM. The company has given no indication of choosing sides.

Digital Island - Builder of broadband networks. Has relationships with both RealNetworks and Microsoft, and alliances with Sun and Inktomi on one side, and Compaq and Intel on the other, putting them squarely on the fence. Recently accepted substantial financing from Microsoft, Compaq, and Intel to build a network to accommodate Windows Media, which may signal a leaning toward keiretsu #2, but nevertheless they have remained cautiously noncommittal, and will likely be happy to go with whoever eventually wins.

Oracle - Database management company. Oracle competes head on against media-keiretsu #2, and has no love for Microsoft. Had some relationship with News Corp. at one point, the current status of which is not clear. The company's role in media delivery is not yet well defined. When it becomes more so, the company will likely join keiretsu #1.

Qwest - Telecommunications and broadband network provider. Like Cisco, the company has given no indication of choosing sides.

Auditore notes the tendency for keiretsu monogamy, and suggests that one way to predict future alliances for a given company is to look toward the other keiretsu members and the alliances they have struck. While the tendency may be for keiretsu monogamy, an equally interesting aspect of media-keiretsu is the degree to which the major participants will create tactical relationships with organizations not in their group (*1). There are many examples of this, such as the relationship between AT&T and British Telecom, and Intel's work with RealNetworks to engineer computing solutions optimized for streaming media applications. However these relationships, as Auditore suggests, seem to be purely tactical and do not alter the formal media-keiretsu lineups�so while we may see shifting tactical alliances between various companies, strategic alliances tend to remain relatively pure within a media-keiretsu - unless a company actually chooses to realign itself, strategically moving to another group or the Independents list.


Media-keiretsu have emerged because the Internet architecture has reintroduced a state of hyper-competition to the media delivery industry - especially so in the areas of delivery and consumption, which have been more or less locked up for years under the traditional television architecture. This ramp-up in competition is not for nothing. The Internet is a brand new medium, with wide structural gaps whose filling requires aggressive moves by those who would take advantage of the lucrative opportunities in the space. In this environment, establishing new positions and maximizing potential returns while minimizing risk and protecting current positions is best accomplished by banding together, and companies from each segment have been quick to form mutually beneficial alliances that will also advance the common agenda - which is to push the need for broadband use among consumers and validate the new model of media consumption. This new frontier also requires a key enabler, an entity capable of crossing the boundaries that exist between segments in the media delivery industry, and uniting the goals and operations of individual organizations.

Auditore writes, "While these alliances appear to benefit those participating, the value to the consumer remains ambiguous; in the near future consumers may find themselves in an environment where technology costs will be low, connection costs low, but choices may be limited." With the forces of coalescence and consolidation operating at full tilt, the field of possible emerging media-keiretsu has narrowed to only two. With a conspicuous absence of accepted standards for the delivery and consumption of media formats over the web, the stage is set for a good old-fashioned slug-fest. There are two possible outcomes in this environment:

(1) The two core companies from each keiretsu will recognize that the quickly maturing industry of media delivery will benefit most from the implementation of standards, and will work together to create a set of basic standards that will make the different systems and approaches at least minimally compatible. The adoption of standards would likely accelerate the maturation process and quickly increase the overall size of the online media delivery market, thus growing each side's potential returns substantially without affecting their respective market share. This would drastically soften the edges around any media-keiretsu and possibly even disintegrate them altogether.

(2) The two core companies will engage in an all-or-nothing battle for the lion's share of the media delivery pie. The winner will dominate the industry in near-monopolistic terms. This scenario could possibly slow the maturity of the industry somewhat, though perhaps not by any noticeable measure. As the industry matures and the market expands exponentially, content producers will be pressed to align themselves with whichever core enabler emerges as the clear leader. Following this scenario all the way to the end of the line, it is likely that the losing company will retreat from the enabler position and reposition themselves solely as a solution provider in the delivery area, providing or enhancing the necessary infrastructure for delivery of content to consumption devices.

Of the two scenarios, the second appears to be the most probable. As the only two real contenders left, and considering the lack of a serious threat coming from any other quarter, there is no perceptible strategic advantage to be found in cooperation. Another factor is the level of industry-specific diversification within the two core companies. Speaking strictly within the media-delivery industry, RealNetworks has diversified its revenue streams to a large degree, while Microsoft has not yet been able to do so in comparable fashion. Considering that the first scenario eliminates the primary source of differentiation and makes a virtual commodity of enabling services, we can see that if this were to come to pass, the less diversified company would be at a significant disadvantage to the other. Therefore Microsoft is much more likely to want to compete head to head than they are to capitulate and embrace a shared standard.


*1 -- Auditore, Stephen. "Emerging Info-Keiretsu." Zona Research Inc., 1996.

Read More Posts by This Author
Go To This Post
More Recommended Posts
Get past Posts of the Day in the Archives