Competition and the Development of the Internet in Japan by Robert F. Delamar
Academic writers observing the spectacular rise and fall of the Japanese economy since 1945 have attempted to pinpoint the source within the famously opaque Japanese political system that bears the primary responsibility for this economic ascendence and decline. Within the short life-cycle of the Japanese Internet boom of the late 1990's, one can observe a vignette of the broader debate. Chalmers Johnson first identified a tightly controlled economy, "guided" by the Japanese bureaucracy as the source of the Japanese economic miracle. Subsequent writers have argued the opposite, that any credit for the Japanese economic miracle can be attributed to a competitive economy that somehow managed to flourish despite the best efforts of the bureaucracy to suppress it, and futher, that bureaucracy and lack of transparency within the Japanese political system, is primarily responsible for Japan's current economic woes. This paper argues that bureaucratic efforts, mirroring Chalmers Johnson's "developmental state" were partly responsible for Japan lagging far behind its industrial neighbours in economic development associated with the growth of the Internet until 1999. It was only in the latter half of the 1990's, when a concerted effort was mounted to deregulate the telecommunications industry, did the development of the Internet (and the associated economic benefits that flowed therefrom) take off in Japan. Thus, the development of the Internet economy in Japan seems to mirror the arguments of the pro-competition academic writers in the broader debate about the political source of the rise and fall of the Japanese economy. Competition and deregulation helped to spur the development of the Internet in Japan in the latter half of the 1990's. Bureaucracy had inhibited its development until then.
Economic Development: An Academic Framework
The Growth of the Internet and the Internet Economy in Japan
Technology for Social Inclusion
Comparative Context: The United States
Application of Theory to Case Study: Competition Wins the Day
Much academic ink has been spilled debating whether deregulation and competition or a "developmental state" was responsible for Japan's spectacular economic rise and fall since the end of the Second World War. This paper will attempt to examine these competing academic theories within the context of a recent economic development in Japan: the growth of the Internet and the Internet economy. By examining the historical development of the Internet in Japan and the subsequent growth of the Japanese Internet economy, this paper will conclude that competition and deregulation can be credited with stimulating and promoting the overall growth of the Internet and Internet economic development in Japan.
Economic Development: An Academic Framework
An Economic "Miracle"
In 1979, Ezra Vogel's Japan as Number One was published, signalling the beginning of a decade of intense scrutiny of Japan's postwar economic development by popular and academic sources throughout the industrialized world . In the decades preceding the publication of Vogel's book, Japan had risen from the ashes of the Second World War, to hold by the mid-1970's the rank of the world's second largest economy and was invited to become a founding member of the then Group of Seven (G-7) largest industrialized economies . The question foreign writers asked more often than not in the ensuing decade was - which Japanese political actor was largely responsible for Japan's economic miracle? Starting with the analysis of Chalmers Johnson, the answer most often offered in reply was - the Japanese bureaucracy . According to Johnson, taking issue with writers such as Vogel, foreign observers added little to the debate in the 1980's given their inability to look beyond the cultural bias that informed their analysis . For Johnson, a complex interplay between law, government and bureaucracy is the hallmark of the Japanese political system, and the product of this interplay, what he calls the "developmental state" was largely responsible for the Japanese economic "miracle" . How exactly then, had the Japanese bureaucracy (if Johnson's thesis was to be believed) engineered the Japanese miracle?
Johnson's argued that the Ministry of International Trade and Investment ["MITI"]  largely guided and implemented the complex interplay between government and bureaucracy which resulted in Japanese "industrial policy" that was the heart of the Japanese economic miracle . But what exactly was MITI doing? Observers looked at the actions of MITI and saw, not formal regulation as was characteristic of bureaucratic intervention in Western economies, but rather a mix of protectionism and developmental policies, especially "administrative guidance", or non-legal bureaucratic exhortations to industry to comply with the policies of the developmental state . MITI rarely had to resort to formal regulatory action according to Johnson, because administrative guidance was enough to keep Japanese industry following government targets for economic development. MITI's power over the developmental state was so pervasive, non-formal legal mechanisms were enough to keep the economic engine moving . As the 1980's progressed, revisions of Johnson's thesis began to appear.
Hiroshi Iyori responded to Johnson's thesis with an assessment of the power of MITI and its alter ego the Japan Fair Trade Commission ["JFTC"] . The JFTC, according to Iyori, as the guardian of the free market, was largely responsible for Japan's economic miracle after the Second World War, noting that where MITI had intervened to promote what Johnson called the developmental state through administrative guidance, and specifically where MITI encouraged cartelisation, its economic intervention was generally unsuccessful (e.g. textile, petroleum and petrochemical industries) . Further, Iyori argued that it was the system of competitive unregulated industries (e.g. automobiles, motorcycles, cameras and electronic goods) that were the backbone of Japanese economic development, and where the real roots of Japan's economic miracle lay . Thus, according to Iyori, though Japan's economic miracle after the Second World War can partly be attributed to MITI's intervention in the economy; in his view it was primarily competition and less Johnson's developmental state that was the primary engine of Japanese economic development in the postwar years .
John Owen Haley built on the same thesis as Hiroshi Iyori. In an article entitled "Administrative Guidance versus Formal Regulation: Resolving the Paradox of Industrial Policy"  and in a later book entitled Authority Without Power,  Haley attacked Johnson's formulation that saw the bureaucracy as primarily responsible for Japanese economic success after 1945. According to Haley, the Japanese bureaucracy had to resort to administrative guidance because it was much less powerful than its Western counterparts . Without the ability to resort to formal regulatory and legal instruments, the Japanese bureaucracy could guide but could not lead. At the heart of the Japanese economic miracle therefore was the lack of regulation and competition. The Japanese economic miracle was due to the lack of power possessed by the Japanese economic ministries, and the ensuing regulatory vacuum .
Case Study: The Internet Economy
From 1990 until the present day Japan has been mired in recession . This paper will look at one area of relative strength in the Japanese economy, the development of the Internet economy throughout this period of recession and contrast its paradoxical success and failure to lead Japan out of economic oblivion within the two conflicting economic visions presented by Johnson and Iyori/Haley. As Iyori notes, "Whether Japan's remarkably high growth rate after World War II can be attributed to the maintenance of competition by means of antimonopoly policy or to skilful control and guidance through the use of industrial policy is subject to unending debate" . Within the short economic lifecycle of the Japanese Internet economy one can see both theories' relative merits and faults. This paper then will add one more opinion to the debate.
The Growth of the Internet and the Internet Economy in Japan
The Growth of the Internet
The growth of Internet access in Japan mirrors that of the rest of the industrialized world, with some notable exceptions . In Japan, Internet access is generally facilitated via wireless media, especially hand-held telephones. This is an important fact for reasons which we will explore later.
The Internet is a global communications network comprised of no-fault data transmission (meaning that if a path to a destination address is blocked, the transmission will seek an alternate path to that destination along as many alternate paths as exist) between distributed interconnected nodes . Because of its distributed nature (access to the physical infrastructure of one Internet Service Provider ["ISP"] allows for access to the data on all computers attached to the networks that the ISP is attached to) the Internet enables communication across widely distributed systems .
The predecessor to what we call the Internet today, the ARPANET, was developed in the 1960's and financed with funds from the United States Department of Defence . Japan was relatively late to the game; the first Japanese based UNIX  network JUNET (Japan UNIX Network) came online in 1984 . Like the North American and European networks that preceded it, JUNET and its contemporary WIDE were academic and research networks, allowing for institutional data sharing within a closed network . The first commercial Internet exchange in Japan was originally conceived of in December of 1992, and went into service in 1993 as Internet Initiative Japan (IIJ), now one of Japan's largest ISPs . By 1997 there were more than 140,000 unique interconnected networks which comprised the Internet [see Figure 1].
Figure 1: Unique Networks Comprising the Internet 
In a January 2001 national comparative Internet access survey (calculated as the percentage of a given country's population that have Internet access) Japan did not make the top 10 [See Figure 2].
Figure 2: Internet User Penetration by Country 
But calculated differently, by the spring of 2001, Japan as an economic region, had the second highest percentage of persons with Internet access in the industrialized world at 23 percent . Yet, by some estimates, even these early numbers were already out of date. The rapid pace of Internet development in Japan in 2000 and 2001 was foreshadowed a year earlier by Yoshiko Motoyama, an analyst in the Tokyo offices of Morgan Stanley as a "quiet revolution in the works" . Ms. Motoyama's forecast proved correct. By March 2001, more than 31 million Japanese had Internet access  double the number the year before in March 2000, when the number was only 16.5 million, a growth rate of 88 percent in one year  (compared to growth in the United States which was "in the teens") . What exactly was happening in the Japanese economy to explain such a dramatic rise? Japan had been clearly behind the rest of the industrialized world in the development of the Internet economy. As Ms. Motoyama noted in 2000, "Japan's relatively low PC and Internet penetration rates were cited as obstacles to the diffusion of the Internet in Japan. Now, these PC numbers and Internet-subscriber numbers are reaching new heights, rising by double digits year over year, while mobile Internet subscribers have grown from virtually zero to 7 million in just the last 16 months" . Why the sudden change? Let's examine in parallel, the impact the sudden development in Internet access had on the rest of the Japanese economy in 1999 to 2001.
The Growth of the Internet Economy
In 1998, the latest year for which we have accurate government figures, shows that the Japanese total Information-Technology ["IT"] Industry (including economic output from ISPs, E-Commerce, Internet Advertising, Hardware, Software, Information Services, etc.) grew steadily through the economic downturn in the late 1980's and throughout the 1990's, amounting to 112.9 trillion yen in 1998 (or 12.5 percent of Japanese real domestic total output) or an average annual growth rate of 7.3 percent from 1980 to 1998 . Analysis from the private sector suggests that the growth rate for the entire IT industry increased rapidly from 1999 to 2001 as Internet access grew, as Yoshiko Motoyama's analysis predicted .
The reason for this is relatively straightforward. As people use the Internet, the utility of their computing device increases (point-to-point communication applications, especially e-mail, are the reason many people use the Internet in the first place) . As more people become Internet users, they purchase PCs, wireless phones, PDAs and other access devices to increase the availability of their preferred Internet applications (e-mail, Instant Messaging, etc.). The boom in devices then triggers software development, which is meant to increase the utility of the access devices, and the cycle begins. Further, as more users attempt to access Internet content, it becomes necessary to build more fibre-optic backbones to support the increased demand. These backbones require the purchase of expensive network hardware and fibre-optic circuits. The net result is a boom in IT spending through all of its economic segments, from consumers to telecommunications companies. One especially clear indicator of this interrelated phenomenon is the link between PC sales and Internet access. As JETRO notes in its 2001 report on the state of the IT industry in Japan, "Internet access and diffusion of personal computers are closely related to each other" . Yet, as Yoshiko Motoyama indicated earlier, prior to 1999 the relatively low percentage of persons with PCs in Japan was seen as a barrier to the diffusion of the Internet in Japan. What was the barrier? Why did the Internet economy stall before 1999? The technology that kicked off the boom and circumvented the PC dilemma can be described with one word: wireless.
By mid-2000 between 25,000 and 30,000 new subscribers per day were signing up to the various wireless services offered by leading Japanese telecommunications companies . This pace continued at a break-neck speed through 2001 . According to Hiromi Abe, an analyst at Morgan Stanley in Tokyo, 82.4 percent of subscribers to NTT DoCoMo i-Mode service cited e-mail as the major reason why they used the service. Add the availability of Internet content (a reason for purchasing i-Mode service cited by 28 percent of users) and the relative utility of having both a cellular telephone and Internet access device bundled into one package (at only a 30 percent premium over the price for a regular cellular phone) and the roots of i-Mode success become clear . Though PC ownership, by global standards remained low in Japan (PCs being the primary means by which users in North America access the Internet), the substitution of PCs for handheld wireless access devices meant that Japan could finally catch up. By mid-2000, Yoshiko Motoyama estimated that Japan was two years behind the United States in the development of the Internet, but was expected to see convergence within the next couple of years .
Regulation and Growth: Is there a Connection?
How then can we explain Japan's relatively slow Internet development, both in terms of Internet access, and the Internet economy as whole compared to the rest of the industrialized world before 1999? How can we explain the sudden and inexplicable rise in just over two years from technological outsider, to world leader? For the answer we must examine the bureaucratic and regulatory environment in Japan within which the Internet and the Internet economy developed.
The regulatory body that has had the greatest impact on the development of the Internet in Japan is the Ministry of Public Management, Home Affairs, Posts and Telecommunications (formerly the Ministry of Posts and Telecommunications) ["MPMHAPT"]. The Internet economy really began in Japan in 1991, when the MPMHAPT gave approval to telecommunications carriers to begin offering commercial ISP services . As was noted earlier, the first commercial ISP was IIJ, which began offering Internet access in 1993 . The first Internet connections in Japan were extremely expensive, and because most of the commercial ISPs relied on the infrastructure of the then-monopoly Nippon Telephone and Telegraph ["NTT"] from which they leased circuits in order to provide their service, it was an uncompetitive market . Further, a number of regulatory hurdles exist in Japan (and some remain to this day) for those who would potentially provide competitive service to the market.
The MPMHAPT classifies ISPs in three categories: Access Providers, Presence Providers and Content Providers . Presence and Content Providers are not required to notify the MPMHAPT or register with them, before starting their business. However, Presence and Content Providers do not actually provide Internet access, but are rather content networks such as Yahoo! Japan that provide services such as search engines and Web mail . To access the services of these companies a consumer or business must already have an Internet connection. Access Providers on the other hand were highly regulated from 1991 to 1998. Two types of Access Providers exist under the Telecommunications Business Law. Type I providers (NTT, KDDI) own their own circuits (physical infrastructure) and sell services to customers over that infrastructure. Type II providers (Japan Telecom) provide services to customers over lines leased from Type I providers . Permission is required from the MPMHAPT to operate as either a Type I or Type II provider, and the MPMHAPT requires a company to file a business plan before they will approve an application . Until 1998 foreign companies were not allowed to possess Type I licenses or own a majority interest in a Japanese company with a Type I license . This effectively shut the world's largest ISPs out of the Japanese market . When NTT received a Type I license in 1996, it quickly grabbed the largest share of dedicated circuit business market adding to its stranglehold on the Japanese backbone market . The net result of this highly regulated system was little competition and high-costs, further stunting the development of the Internet in Japan.
Between April of 1995 and December of 1997 the MPMHAPT began to deregulate the Japanese ISP market. Prior to deregulation companies that owned primarily voice circuits could not offer Internet service. Thus, until 1996, NTT was effectively shut out of the dedicated circuit business market, though it had the largest backbone in Japan . Most significantly, restrictions on foreign ownership of Japanese ISPs were lifted in February, 1998 . Foreign carriers quickly jumped into the Japanese market with now-defunct PSINet buying a number of Type I carriers in 1998 to gain quick market traction, and Worldcom - the world's largest ISP - opening a Japanese subsidiary in 1999 . Internet access charges for dedicated circuits quickly began to fall .
But deregulation didn't stop there. By 2000, numerous legislative and bureaucratic initiatives were set in motion with the goal being the full-scale deregulation of the Japanese telecommunications industry. As 2001 guidelines released by the Japan Fair Trade Commission notes, "the field is in a transition from a monopoly to competitive market" . Legislative efforts culminated in the passage of the IT Basic Law in January of 2001. The purpose of the law was to "promote ... fair competition among telecommunications business operators and for taking other measures necessary for encouraging the creation of a world-class advanced information and telecommunications network which is accessible by the public at a low cost" . The link between the creation of a robust IT industry and competition became the law in Japan.
The JFTC led a parallel effort to ensure that the bureaucracy would not stand in the way of the legislative developments. In November 2001, together with the MPMHAPT, the JFTC issued a series of guidelines designed to "[promote] competition in the telecommunications business in Japan ... [and] to prevent unnecessary confusion and burden that telecommunications carriers would have in the implementation of the two laws" . The two laws referred to are the Antimonopoly Act of 1947 and the Telecommunications Business Laws of 1984.
The guidelines exist as a triumph for those who wanted to see the formerly highly regulated sector become deregulated and competitive. Much of what was formerly regulation (under the Telecommunications Business Laws) became anti-competitive behaviour. For example, the new guidelines specify in great detail, provisions to ensure greater access to rights-of-ways for competing ISPs enabling them to build networks, co-location access in competitor central offices for competing ISPs, and a myriad of technical changes in order to allow greater competition among existing ISPs . Breach of the guidelines would result in the right to complain to the JFTC or to the MPMHAPT, with mediation and conciliation being the preferred means of dispute resolution for the complainant ISPs (administered by both agencies) . Finally, a report of the Telecommunications Council, an administrative arm of the MPMHAPT responsible for setting telecommunications policy, issued a number of recommendations in a preliminary report in 2001, which included the divesting of the Japanese government's ownership stake in NTT and means to free up competition between the various regional arms of NTT among other recommendations .What was driving the deregulation juggernaut within the Japanese government? Two reports from the Telecommunications Council first in 1999, and again in 2000 shed some light on the mystery. In the first report, dated August of 1999 and entitled "Principles of Major Info-communications Policies for Fiscal 2000: Reviving the Japanese Economy through Info-communications" the Telecommunications Council notes that:"Advanced info-communications is the key to socioeconomic development: The development of advanced info-communications networks and technologies has laid the foundation for new growth in the 21st century, and we may expect to see a dramatic surge in the efficiency of socioeconomic systems as a whole. In the United States [emphasis mine], both the public and private sectors are making enormous investments in info-communications areas, stimulating economic growth worldwide" .
In the second report, dated March of 2000, entitled "IT Japan For All" the Telecommunications Council notes that:"The 20th century was by all accounts an era of considerable prosperity for Japan. As a result of the remarkable success of the postwar era, Japan became a model of the industrialized society for the world. Currently, however, the nation has been mired in a prolonged economic slump and, although there have been signs of improvement, no clear direction for future prospects has emerged ... The 1990's saw the standardization of IT serve as the primary impetus driving the economic recovery of the United States [emphasis mine] and create a ripple effect around the globe ... " .
The report goes on to talk about competition and globalization and notes that Japan must be globally competitive in order to survive in the information age . It seems that the American IT boom in the 1990's was seen by the Japanese bureaucracy as a way out of Japan's economic morass. Japan was clearly behind the rest of the industrialized world in the development of the Internet economy in 1999. It seems that the Japanese government and bureaucracy with its new appetite for deregulation, was attempting to replicate American economic success by following recent American legislative and bureaucratic developments in telecommunications law.
Comparative Context: The United States
Before 1982, AT&T as the parent company of the Bell System provided local telephone and long distance services, with a few exceptions, to almost all of the United States . In 1982, responding to a series of legal challenges brought by Bill McGowan, a former President of MCI, together with intervention by the U.S. Justice Department, Federal Judge Harold Greene declared AT&T to be a monopoly and ordered it to divest its regional operating companies (Bell South, Southwestern Bell, etc.), an order known as the Modified Final Judgment . These new companies were allowed to offer for the first time not just telephone service, but they could now enter and compete in an unrestricted manner in other industries such as the computer and information systems fields . The companies that emerged from the break-up of AT&T were not fully free of regulation - portions of their business remained regulated . From 1982 until 1996, the Modified Final Judgment stood as the most important telecommunications law in the United States. However, with the passage of the 1996 Telecommunications Act, Congress usurped the importance of the Modified Final Judgment. As Goldman and Rawles note: "The goal of the Telecommunications Act of 1996 can be summarized in three words: free market economy" . As the key legal instrument governing telecommunications policy in the United States, the goal was competition in the telephone market.
One key area that remained unregulated by the 1996 Telecommunications Act was ISP interconnection . Unlike the long distance telephone system, the backbone services offered by ISPs are basically unregulated. As noted earlier, the Internet is essentially a network of networks. Access to the individual networks that make up the Internet is governed by a system of free-peering (where carriers interconnect their networks at no charge to each other) and paid transit (where large networks charge small networks fees to connect to their networks) . Whether networks interconnect via peering or transit is based on the respective size of their networks . ISPs found themselves in a curious regulation free zone (as noted earlier, telecommunications companies have both regulated and unregulated portions of their business). The Federal Communications Commission ["FCC"], according to Kende, "maintains a policy to 'focus on sustaining competitive communications markets and protecting the public interest where markets fail to do so'" . The FCC has chosen to treat computer-based services from their inception as an unregulated industry, provided that the telecommunications services component remained competitive (an ISP provides both telecommunications and computer services) . Thus as Kende notes:"telecommunications providers are subject to common carrier regulations that ensure nondiscriminatory access to end users; together with antitrust enforcement, these regulations serve to protect against anti-competitive behavior by telecommunications providers with market power. In markets where competition can act in place of regulation as the means to protect consumers from the exercise of market power, the Commission has long chosen to abstain from imposing regulation" .
According to Kende, as long as a competitive market for backbone services exists, the American ISP market should remain unregulated. However, should the recent trend toward industry consolidation and paid transit result in a non-competitive market, the FCC should consider regulating interconnection charges to ensure a free and fair market .
Though it's not necessary in these pages to retell the great IT boom of the late 1990's in the United States, it's clear that a competitive market for ISP services emerged in the United States in the latter half of the 1990's, and the result was cheap Internet access and the residual boom in the Internet economy which resulted. By 2001, 35 percent of the U.S. population had Internet access, and household PC penetration was upwards of 50 percent . Perhaps a more telling statistic is the pattern of global Internet traffic. The United States, home of the world's largest ISPs, acted as an intermediary for more than half of the world's Internet traffic.
Figure 3: U.S. Internet Traffic Map 
According to Caida.org, an Internet research centre based at the Supercomputer Center at the University of California-San Diego in 2000: "The U.S. [was] the major Internet transit intermediary for the rest of the world: 71 percent of traces that neither start nor end in the U.S. still pass through it. In most connections between different countries, the U.S. is the only third party country that also appears in the path" . In other words American networks comprised the heart of the Internet.
Application of Theory to Case Study: Competition Wins the Day
We've seen then, that the ISP market in the United States has been almost unregulated since its inception. Further, we've seen since 1996 a consistent attempt by the American Congress and the bureaucracy, to free telecommunications companies (many of which are both telephone companies and ISPs) to compete throughout the United States for the Internet access market. The FCC has threatened to regulate one aspect of the ISP market, the cost of network interconnection, due to a movement to charge networks access to the large backbones which make up the heart of the Internet, a practise known as "paid transit". However, for all intents and purposes, all aspects of the ISP market have remained unregulated in the United States, and the Congress and bureaucracy have attempted to ensure a competitive market there.
In Japan, the exact opposite scenario is true. The ISP market has been highly regulated since its creation in 1991. Efforts by the then MPT restricted access to the Japanese market for foreign providers (protectionism) and retained the right to license Type I and Type II carriers (barring NTT from the market until 1996). This closely resembles Johnson's thesis that the Japanese bureaucracy leads the complex process that results in the industrial policy that is at the heart of the developmental state. Throughout the 1990's it was the former MPT who played this role in the development of the Japanese Internet economy. However, this lends credence to Iyori's argument that where Japanese industrial policy encouraged cartelisation, economic ruin was the result, as data suggests that the anti-competitive, highly regulated nature of the Japanese ISP market prior to 1996 stunted the development of an Internet economy there.
Data suggests that Japan lagged the rest of the industrialized world until 1999 in at least the Internet access market as measured by the number of Internet users in Japan (an indicator of the overall development of a country's Internet economy). Further, data suggests that when the process of deregulating the ISP market (the primary providers of Internet access) began in 1996, the result was a dramatic increase in the number of ISPs providing Internet access in Japan (from 406 in February 1996, to 1,645 in Feb. 1997 to 2,576 in Feb. 1998 according to the MPT) . This process was accelerated by the removal of regulatory barriers to entry for foreign ISPs, which resulted in a decline in prices for dedicated circuits due to increased competition . In 1999, when NTT DoCoMo unveiled the new wireless access technology i-Mode, competitive products from Japan Telecom and other competitors emerged almost simultaneously. The result was that from the beginning, the wireless Internet access market has been competitive in Japan . The wireless access boom led to dramatic growth in the Internet economy in Japan from 1999. In parallel with the Internet economy boom, we saw the bureaucracy dismantle of much of the telecommunications regulation governing ISPs and the JFTC, along with the successor agency to the MPT, the MPMHAPT showing a strong commitment to deregulation and competition as the future engine of IT economic development in Japan. Clearly then, deregulation and competition in part led to the Japanese IT boom after 1999.
An interesting final note. In 2001 the world began to show signs of an economic slowdown, led by the collapse of numerous "pure-play" Internet companies, and the slowdown in the Internet hardware industry. Just as the Japanese IT boom was beginning, it slowed - dragged down by fears of an American recession. However, analysis continues shows that the IT industry in Japan remained stronger than just about any other segment of the economy .
An analysis of the development of the Internet and the Internet economy in Japan since 1991 suggests that competition and deregulation, primarily championed after 1999 by the Japan Fair Trade Commission, were at least partly responsible for the impressive growth in Internet development and the subsequent Internet economy in Japan after 1999. Prior to 1996, the Japanese bureaucracy acting in a fashion characteristic of Chalmers Johnson's developmental state slowed Japan's Internet development. It was only after the deregulation of the telecommunications industry in Japan beginning in 1996 and the increased competition that resulted, that the Internet economy really gained momentum. This seems to be consistent with Hiroshi Iyori's analysis that deregulation and competition were largely responsible for postwar Japanese economic development. Nevertheless, data does show that the Japanese IT industry as a whole grew faster than the rest of the Japanese economy after 1980, and this steady growth was only augmented by an IT boom in the late 1990's. Further, the theoretical models discussed in this paper may not be able to explain other factors that contributed to the late 1990's Internet boom in Japan. For example, wireless technology was new and relatively cheap, Japan is a highly urbanized, densely populated society where PC ownership primarily for the purposes of Internet access may have less utility in small Japanese flats than a small handheld device, and the IT boom Japan experienced after 1999 was part of a larger global economic upturn. For the purposes of this paper, however, it seems generally clear that deregulation and competition encouraged the development of the Internet in Japan, particularly considering the generally slow (by global standards) pace of Internet development in Japan prior to deregulation.
About the Author
Robert F. Delamar is co-founder and Managing Editor of the online arts and technology journal spark-online. Robert is currently studying in the Faculty of Law at the University of British Columbia, home of the internationally respected Center for Asian Legal Research, under the direction of Professor Stephan Salzberg. Prior to law school, Robert worked for a Silicon Valley technology start-up company where he was responsible for global market research and business analysis (including Japan).
1. Y. Funabashi, "Japan's Unfinished Success Story" (2001) 48 Japan Quarterly 4 at 10 [hereinafter Unfinished Success].
2. Unfinished Success, ibid. at 10.
3. See C. Johnson, MITI and the Japanese Miracle (Stanford, Calif.: Stanford University Press, 1982) at 17 [hereinafter MITI and the Japanese Miracle].
4. MITI and the Japanese Miracle, ibid. at 7. See also ibid. at 23 where Johnson notes: "Americans are sometimes confused by Japanese economic policy because they pay too much attention to what politicians say and because they do not know much about the bureaucracy ... ."
5. MITI and the Japanese Miracle, ibid. at 17.
6. Now the Ministry of Economy, Trade and Industry ["METI"].
7. MITI and the Japanese Miracle, ibid. at 17-34.
8. MITI and the Japanese Miracle, ibid. at 20-30. Johnson's analysis of the difference between Western and Japanese ideas of bureaucracy is particularly important.
9. MITI and the Japanese Miracle, ibid. at 31-32. Johnson looks to Japanese history to find the roots of the "developmental state" and MITI's role in it, arguing that it was not until the 1960's that anything approaching a "theory" of the developmental state existed in Japan.
10. H. Iyori, "Antitrust and Industrial Policy in Japan: Competition and Cooperation" In: Law and Trade Issues of the Japanese Economy (Seattle: University of Washington Press, 1986) [hereinafter Antitrust and Industrial Policy].
11. Antitrust and Industrial Policy, ibid. at 15.
12. Antitrust and Industrial Policy, ibid. at 15. Iyori notes: "... although competition was a primary cause of for industrial growth in postwar Japan, the development of infant industries was in a limited sense the result of protection by industrial policy. To the extent that industrial policy eliminated or suppressed competition, however, the ultimate consequence was economic failure."
13. Antitrust and Industrial Policy, ibid. at 15.
14. J. Haley, "Administrative Guidance versus Formal Regulation: Resolving the Paradox of Industrial Policy" In: Law and Trade Issues of the Japanese Economy (Seattle: University of Washington Press, 1986) [hereinafter Administrative Guidance].
15. J. Haley, Authority Without Power: Law and the Japanese Paradox (New York: Oxford University Press, 1991) [hereinafter Authority].
16. Administrative Guidance, ibid. Haley notes: "The prevalence of administrative guidance can be explained by two basic features of postwar Japan: the essentially promotional thrust of Japanese economic and social policies and the intrinsic weakness of formal law enforcement in Japan." As Haley further notes, for formal law enforcement to exist as a credible threat, it must be "available and actually used."
17. See generally, Authority, ibid. See also Administrative Guidance, ibid.
18. "The Sadness of Japan" Economist, volume 362, number 8260 (16 February 2002), 11 at 11.
19. Antitrust and Industrial Policy, ibid. at 14.
20. See J.E. Goldman and P.T. Rawles, Applied Data Communications: A Business Oriented Approach, Third ed. (New York: Wiley, 2001) at 2-12 [hereinafter Applied Data]. In this paper where the term "Internet access" is used, it is defined as the growth in the number of media connected to the Internet. It is a medium neutral term that doesn't distinguish between dial-up, broadband (DSL or Cable), or wireless access. Particular growth rates for a single medium (e.g. wireless) will be identified as such.
21. See Applied Data, ibid. at 657. In this paper where the term "Internet" is used, it is defined as the use of the "Internet Suite of Protocols" over a variety of physical and non-physical media enabling geographically distributed groups of computers to communicate. "Internet Suite of Protocols" (Transmission Control Protocol/Internet Protocol or TCP/IP) refers "generally ... to an entire suite of protocols used to provide communication on a variety of layers between widely distributed different types of computers." See also Telegeography, Hubs + Spokes: A Telegeography Internet Reader (Washington D.C.: Telegeography, 2000) at 15-25 and 62-65 [hereinafter Hubs + Spokes].
23. See R.H. Zakon, "Hobbes' Internet Timeline" at http://www.zakon.org/robert/internet/timeline/, accessed 29 March 2002, [hereinafter Zakon]. The first four nodes on the "ARPANET" were UCLA, Stanford Research Institute, UCSB and the University of Utah. Much of the originally academic thinking behind what became the ARPANET was done at MIT and RAND.
24. See Bell Labs, "The Creation of the UNIX Operating System" at http://www.bell-labs.com/history/unix/#, accessed 30 March, 2002. UNIX is the operating system (and its variations) that runs most of the large servers that house most of the content on the Internet. It was invented in the 1960's by researchers at Bell Labs, the famous research and development laboratory of AT&T. See also "The Bell System" at http://www.att.com/history/history3.html, accessed 30 March 2002.
25. Zakon, ibid.
26. See "IIJ Up Close" at http://www.iij.ad.jp/info/index-e.html, accessed 30 March 2002, [hereinafter IIJ].
27. IIJ, ibid.
28. Zakon, ibid.
29. See "Global Internet Usage has Come a Long Way" eMarketer, at http://www.emarketer.com/estatnews/estats/eglobal/20010420_global_usage_itu.html, accessed 1 April 2002, [hereinafter Global Internet Usage].
30. See M. Meeker, "A Look at Global TMT Market Status and Internet User/Usage Propensity" In: Overview Presentation of Highlights from Recent Internet Reports (Tokyo: MSDW, 19 April 2001) at 4. Cites the following Internet access figures, by region: North America, 33 percent; Japan, 23 percent; Europe, 14 percent.
31. See Y. Motoyama, "The Sleeping Giant Awakes" In: Japan Investment Perspectives (Tokyo: MSDW, 21 June 2000 at 1 [hereinafter Sleeping Giant].
32. See "Japan" CIA World Fact Book, at http://www.cia.gov/cia/publications/factbook/, accessed 30 March 2001. The population of Japan is estimated by the World Fact Book at approximately 126 million as of July, 2001. Assuming MSDW's Internet access figures are correct, 31 million persons represents 25 percent of the Japanese population.
33. See Y. Motoyama, Japan Internet: Active Users Gaining Ground (Tokyo: MSDW, 2001) at 6, [hereinafter Japan Internet].
34. Japan Internet, ibid. at 5.
35. Sleeping Giant, ibid. at 1.
36. See The IT Industries' Guide to Entering the Japanese Market (Tokyo: JETRO, February 2001) at 3-4, [hereinafter JETRO IT]. The JETRO publication cites figures from the then Ministry of Posts and Telecommunication ["MPT"] (now the Ministry of Public Management, Home Affairs, Posts and Telecommunications ["MPMHAPT"]) and the then Ministry of International Trade and Industry ["MITI"] (now the Ministry of Economy, Trade and Industry ["METI"]).
37. See Y. Motoyama and H. Abe, "The Internet in Japan" In: Global Internet Primer (Tokyo: MSDW, June 2001) at 611-641, [hereinafter Global Internet Primer].
38. See C. Boyd, "Metcalfe's Law" In: Why Strategy Must Change, at http://www.mgt.smsu.edu/mgt487/mgtissue/newstrat/metcalfe.htm, accessed 12 April 2002. Robert Metcalfe, founder of 3Com, opined one of the three famous "laws" of the Internet. Metcalfe's law states: "the usefulness, or utility, of a network equals the square of the number of users." See also M. Pastore, "E-Mail Killed the Post Office Star" at http://cyberatlas.internet.com/big_picture/traffic_patterns/article/0,,5931_734641,00.html, accessed 12 April 2002.
39. JETRO IT, ibid. at 16.
40. Sleeping Giant, ibid. at 1.
41. See generally Global Internet Primer, ibid.
42. Global Internet Primer, ibid. at 624-625.
43. Sleeping Giant, ibid. at 1.
44. See "Internet Service Providers, Regulations and Practices" (1999) 28 JETRO Japanese Market Report at 1 [hereinafter ISP Regulations].
45. IIJ, ibid. The first Internet access in Japan was in the form of dedicated circuits installed at business and university customer premises. Dial-up service began shortly thereafter.
46. See ISP Regulations, ibid. at 3 and 8. See also Applied Data, ibid. at 1-12 and 301-356. It's important for the purposes of this paper to understand how Internet service providers ["ISPs"] provide Internet access service to customers and the economic structure this service rests uponŠ
Generally, ISPs build a point of presence ["PoP"] in the central office ["CO"] of a major telecommunications company that serves the geographic region (ranging from a city block to local community) in which they plan to offer service. The ISP then must string fiber-optic or copper wire circuits from the PoP to the business they are offering service to, or lease the right to a fiber-optic circuit or copper wire to the business from a telecommunications company that already has infrastructure. If the service offered is dial-up or broadband (xDSL, ISDN, etc.), special equipment is also co-located in the same PoP to receive the customer's data transmission (coming into the PoP over the telephone line). The data transmissions (whether dial-up, broadband, wireless or dedicated circuit) are then aggregated on to the main "backbone" (large data circuits) owned by the major telecommunications companies (such as NTT or KDDI). The ISP is charged by the large telecom company for access to this backbone (unless they have their own), based on the amount of data traffic they put over the backbone. The large telecom company's backbone is connected at peering points to the backbones of other large networks. The aggregate of all these networks is the Internet. This is what a consumer or business is paying for when they purchase Internet access from an ISP.
The economic issue for the small ISP is pretty clear. If they do not own their own infrastructure (fiber-optic circuits or the local copper wires) they must purchase the right to use a large telecom company's infrastructure. In addition, they pay for the right to place their PoP in the CO of the telecom company. Further they are charged access fees to the large telecom company's network. All of these costs must be recouped by the small ISP in the form of the monthly service charge their customer pays for Internet access. Because the large telecom companies also compete for the same business as the small ISPs, it becomes prohibitively expensive for the small ISP to compete. Consumer Internet access circumvents the problem somewhat, as the consumer already owns either a telephone line into their home, or into their small business. The ISP then only must provide the dial-up PoP or broadband PoP to process the data stream coming over the pre-existing telephone circuit, and must also pay the large telecom company backbone access fees and they are in business. This service cuts out the cost of the local circuit ["local loop"]. In Japan, however, because local telephone calls cost money (unlike in North America) dial-up is expensive (the consumer must pay the Internet access cost and the local telephone toll). Further, in the lucrative business Internet access market, most companies pay for the dedicated circuit. Given the prohibitive cost of physical infrastructure, only the large telecom carriers who have infrastructure in the ground already can compete for this market.
47. ISP Regulations, ibid. at 1.
48. ISP Regulations, ibid. at 4.
49. ISP Regulations, ibid. at 5.
51. ISP Regulations, ibid. at 6.
52. See Hubs + Spokes, ibid. at 62-64. "The ten largest IISPs control 70 percent of international Internet bandwidth." The largest International Internet Service Providers ["IISPs"] are all U.S. based or have American roots and include (ranked according to size): 1) Worldcom/UUNet; 2) Sprint; 3) Cable & Wireless; 4) At&T; and 5) GTE/Genuity.
53. See ISP Regulations, ibid. at 1 and 8. See also JETRO IT, ibid. at 38.
54. ISP Regulations, ibid. at 6.
55. ISP Regulations, ibid. at 6.
56. JETRO IT, ibid. at 39.
57. JETRO IT, ibid. at 39.
58. See JFTC, "Guidelines for the Promotion of Competition in the Telecommunications Field" (2001) at 6, at http://www.jftc.go.jp/e-page/guideli/011130telecomGL.pdf, accessed 15 April 2002, [hereinafter JFTC Guidelines].
59. JFTC Guidelines, ibid. at 5.
60. JFTC Guidelines, ibid. at 6.
61. JFTC Guidelines, ibid at 14-30.
62. JFTC Guidelines, ibid at 63-66.
63. MPMHAPT, "Key points of the first report on Desirable Pro-competitive Policies in the Telecommunications Business Field for Promoting the IT Revolution compiled" (2001) 11, at http://www.yusei.go.jp/eng/Releases/NewsLetter/Vol11/Vol11_21ref.html, accessed 15 April 2002.
64. MPMHAPT, "Principles of Major Info-communications Policies for Fiscal 2000: Reviving the Japanese Economy through Info-communications" (1999), at http://www.yusei.go.jp/eng/, accessed 16 April 2002.
65. MPMHAPT, "IT Japan For All" (2000) at 3, at http://www.yusei.go.jp/eng/, accessed 16 April 2002, [hereinafter IT Japan].
66. IT Japan, ibid. at 4-8.
67. Applied Data, ibid. at 6. The FCC ruled in 1971 that MCI could compete with AT&T for long distance service. Despite this fact, however, AT&T was for all intents and purposes a monopoly.
68. Applied Data, ibid. at 6.
69. Applied Data, ibid. at 6.
70. Applied Data, ibid. at 7.
71. Applied Data, ibid. at 9.
72. M. Kende, "The Digital Handshake: Connecting Internet Backbones" in OPP Working Paper No. 32 (Washington D.C.: FCC, 2000). at 2 [hereinafter Digital Handshake]. Kende notes: "The Internet is not a monolithic, uniform network; rather, it is a network of networks, owned and operated by different companies, including Internet backbone providers. Internet backbones deliver data traffic to and from their customers; often this traffic comes from, or travels to, customers of another backbone. Currently, there are no domestic or international industry-specific regulations that govern how Internet backbone providers interconnect to exchange traffic, unlike other network services, such as long distance voice services, for which interconnection is regulated."
73. For a brief history of the development of the American system of ISP peering and transit see Digital Handshake, ibid. at 4-8.
74. Digital Handshake, ibid. at 4-8.
75. Digital Handshake, ibid. at 9.
76. Digital Handshake, ibid. at 9.
77. Digital Handshake, ibid. at 12.
78. Digital Handshake, ibid. at 39.
79. See Global Internet Usage, ibid. and JETRO IT, ibid. at 17.
80. See "Telegeography Maps" at Telegeography at http://www.telegeography.com/resources, accessed 15 March 2002.
81. See B. Huffaker et al., "Measurement of the Internet Topology in the Asia-Pacific Region" (2000) at http://www.caida.org/outreach/papers/2000/asia_paper/asia_paper.html, accessed 15 April 2002.
82. ISP Regulations, ibid.at 3.
83. JETRO IT, ibid. at 39.
84. See JETRO IT, ibid. 13-15 and 58-59.
85. See generally, Y. Motoyama, "Active Users Gaining Ground" In: Japan Internet (Tokyo: MSDW, 18 May 2001) at 2.
Paper received 6 June 2002; accepted 14 June 2002.
Copyright ©2002, First Monday
Copyright ©2002, Robert F. Delamar
Competition and the Development of the Internet in Japan by Robert F. Delamar
First Monday, volume 7, number 7 (July 2002),