Journal of Interactive Advertising, Volume 4, Number 1, Fall 2003

Emerging Broadband Market and the Relevant Policy Agenda in Japan

Yasu Taniwaki

Economic Counselor and Telecommunications Attaché
Embassy of Japan
Washington D.C., USA

Table of Contents


Japan is now ranked as one of the most advanced countries in the deployment of broadband services. Several factors supporting this deployment can be found in policy measures for the promotion of a broadband network infrastructure (e.g. grants by the government as incentives for investment) and promotion of competition in the broadband market (e.g. drastic deregulation to change the competitive regime). In addition, the deployment of the broadband platform and its corresponding newly emerging business models are forcing the regulatory authority to reconsider new competition policy rules; how to cope with vertically integrated business models is one of the key issues to be tackled. In addition, based on the deployment of broadband infrastructure, the stimulation of demand for broadband services becomes the next step to be explored.


This paper outlines the recently emerging broadband market in Japan and to overview the policy agenda corresponding to the fast changing market environment. At the present time, Japan is one of the most advanced countries in broadband services, and is thus forced to address new policy issues caused by the transition from PSTNs (Public Switched Telephone Networks) to IP (Internet Protocol) Networks.

This paper, first of all, presents an overview of the current status of the Japanese broadband market. Based on this current status, policy issues currently being addressed by the Japanese regulatory authority, or Ministry of Public Management, Home Affairs, Posts and Telecommunications (MPHPT), are summarized from the viewpoint of both the promotion of broadband network deployment and the promotion of competition in the broadband market.

Current status of the Japanese broadband market

In Japan, by the end of May 2003, the number of DSL subscribers amounted to 7.91 million, while the number of internet users via CATV networks amounted to 2.14 million (as of April 2003, Figure 1). In addition, fiber optic service with 100Mbps for mass users was launched in metropolitan areas and as of the end of April 2003 the number of subscribers had exceeded 30 thousand.

Price reduction via competition among service providers is well in progress in Japan. The price for DSL service with 12Mbps has decreased to about 20 to 30 dollars per month and fiber optic service is offered for about 40 to 50 dollars per month (Figure 2).

Two main reasons can be found for this rapid progress in the Japanese broadband market. First, it can be found in the present price structure of telephone services. That is, before the development of the DSL market, people had been using dial-up internet access, although the local call charge is time sensitive. Under this circumstance, DSL service was launched on a fixed rate basis and this has accelerated the transition from dial-up internet access to DSL service.

A second and more important reason can be found in the establishment of collocation and unbundling rules for DSL service providers using access networks of the dominant regional telecom carriers, NTT East and NTT West. This interconnection rule, enacted in autumn 2000, has led to a price reduction through lively competition among approximately fifty DSL service providers and eventually to the remarkable increase of DSL subscribers from the year 2001. Thus, interconnection rules to open up the bottleneck facilities still work quite effectively even in the provision of new services such as DSL. There exists the idea that this kind of open policy does not lead to the development of broadband infrastructure. The issue to be tackled, however, is how to set the economically rational access charges or wholesale prices ensuring a fair rate of return for dominant carriers, which is irrelevant to the discussion on effectiveness of the interconnection rules.

A reason why DSL plays a major role in Japanese broadband services can be found in the current status of the regional telecommunications market. That is, NTT East and NTT West have access lines (mainly metal cables) to every household to provide voice telephone service. On the other hand, the penetration rate of CATV services for households still remains at 31.2% (as of the end of March 2003). Thus, regarding access lines for broadband services, the share of NTT East and NTT West remains at 98% in the regional market, leading to the situation where CATV internet services has a relatively weak market position compared with DSL services via access lines owned by NTT East and NTT West.

Major characteristics of the Japanese broadband market can be summarized as "high speed" and "advanced IP based services." Regarding "high speed" features, the speed of DSL in Japan has increased to 12Mbps presently and has recently reached a speed of over 20Mbps. In addition, newly introduced higher speed service has been continuously provided at almost the same price as the existing lower speed service to attract more customers. In this context, the effective downstream speed of DSL service in Japan is envisaged to normally exceed over 1 Mbps.

Regarding the feature of "advanced IP based services," the development of IP telephony for mass users, launched by many service providers including common carriers and ISPs (Internet Service Providers) from the year 2003, is critical. Two major reasons can be found for IP telephony market development. Firstly, in many cases, a call between members of each service provider is "free", or included in a fixed charge for DSL service without any extra charge. This strategic price plan is attracting more customers to DSL service. Secondly, through the initiation of allocating prefix identification numbers "050" for IP telephony in autumn 2002, a call from legacy telephone equipment to IP telephony equipment has been made possible. Thus, in Japan, the transition from legacy PSTNs to IP networks has been ongoing at a remarkable speed and voice telephony is rapidly becoming one of the applications provided with data or video services. On the other hand, this does not necessarily mean the full replacement of traditional telephony via PSTNs with IP telephony because emergency calls, for example, cannot be provided via IP telephony at present. Rather, new unified and value added services which combine voice, data, and video using IP technology are expected to emerge.

Under these circumstances, many countries will be tackling the issue of how to allocate possible policy portfolios to deploy broadband service markets. Generally speaking, competition among service providers benefits consumers, but many discussions focus on what kind of competition leads to the maximization of consumers’ welfare. In other words, the relationship between facility-based competition and service-based competition needs to be a trade-off and should be well balanced. In Japan, the deployment of network facilities has been promoted with some assistance from the government, leading to facility-based competition. On the other hand, competition policies represented by interconnection rules have contributed to service-based competition. To establish the cycle of positive interaction between facility-based competition and service-based competition, policy portfolios should be allocated properly to promote both facility-based competition and service-based competition.

In this context, the next section focuses on measures to promote facility-based competition, followed by the section focusing on measures to promote service-based competition. Here, if allocations of these policy measures are improperly chosen, it will result in simply empowering the dominant carriers and generating negative effects in the deployment of broadband services. Thus, these two principal objectives — promotion of facility-based competition and that of service-based competition should be carefully considered.

Promotion of broadband network deployment


Regarding access lines for broadband services, a variety of wired or wireless lines are technologically available at present. Fiber optic networks, however, are recognized as principal access lines for broadband services because symmetric (both downstream and upstream) high speed broadband services with 100Mbps are available only via fiber optic cables. Furthermore, two way broadband services such as P2P (peer to peer) are expected to become main-stream in a couple of years. With that in mind, in November 1999 the MPHPT put as its first priority, the deployment of fiber optic networks as principal access lines for broadband services and has set out to accomplish the nationwide FTTC (Fiber to the Curb) by FY2005 (the end of March 2006). Here, the government is targeting the FTTC instead of the FTTH (Fiber to the Home) because "last one mile" shall be left to a variety of competing paths such as DSL, fiber optics and FWB (Fixed Wireless Broadband).

Looking at the current status of the deployment of FTTC, its installation rate reached 72% by the end of March 2003, which is almost in line with the pace set by the government (Figure 3). The background of this steady development of fiber optic networks can be found in the initiatives taken by the government. Evidently, the installation of fiber networks is the role of the private sector and the role of the government is limited only to ease the financial burden of the private sector and to develop an appropriate environment to increase incentives for investment in fiber optic networks by the private sector.

Measures to promote deployment of broadband infrastructures

The Japanese government has been assisting service providers to install fiber networks through several programs. Three major policies are summarized in the following paragraphs.

First, a loan system with interest rates lower than the market rate is available to any carrier (including CATV operators providing internet access services) with a fiber optic network installation plan (Figure 4). In addition, tax deductions for digitization investments by carriers are available. Furthermore, carriers introducing broadband access networks can apply for a special 6% to 18% redemption in corporate tax and can decrease the tax standards for fixed assets by 20% to 25%.

Second, even with this assistance, the deployment of broadband infrastructure is not expected in rural areas for economic reasons. Although the coverage rate of fiber optic networks in total has reached 72%, the explicit difference in the installation rate of fiber optic networks between metropolitan and rural areas is 89% and 49% respectively (as of the end of March 2003, Figure 5). To overcome this domestic digital divide problem, the government has set up a program to subsidize local governments for the cost of establishing LANs (Local Area Networks) using fiber networks developed for public services such as e-government projects. In return, allowing common carriers to use the surplus of these network capacities on a non-discriminatory basis has been promoted. To ensure simplicity and non-discrimination on the lease of fiber optic network capacities from local governments to common carriers, the MPHPT, in August 2002, stipulated "The general guideline for standard procedures for local governments to lease fiber optic capacities to common carriers."

Third, Right of Ways (ROWs) guidelines should be focused. Public infrastructure such as poles, ducts, and conduits are essential for carriers to install their networks, but this infrastructure has been owned mainly by public utilities such as NTT East, NTT West and electric power companies. Under this situation, how to ensure that these owners open up poles etc. for other carriers on a non-discriminatory basis is one of the key issues to be addressed. With that in mind, in March 2001, the MPHPT stipulated the guidelines on ROWs, which describes standard procedures for lending poles etc. to other carriers and their relevant standard terms and conditions. This guideline has been annually reviewed following and reflecting the improvement of actual procedures between the parties concerned (MPHPT 2001).

Promotion of competition in the broadband market

Change of business environment in the transition from PSTNs to IP networks

Changes in the market structure reflecting the progress of IP technology forced the regulatory authorities to change the competitive regime from the legacy model based on PSTNs to IP networks. The Japanese competitive regime is stipulated in the Telecommunications Business Law (TBL) which took effect in April 1985. The current regime makes distinctions between facility-based carriers and resale-based carriers, where relatively stronger regulations have been posed on facility-based carriers. In addition, tariff regulations are applied by each service category to check the economic rationality of pricing. Moreover, such regulations as technological requirements and interconnection rules are preliminarily based on PSTNs, which are not always applicable to IP networks.

In connection with this change, the current approach taken by the MPHPT to reform the competitive regime has two stages. As the first stage, recognizing the transitional period from PSTNs to IP networks, the fundamental regulatory framework is to be reformed towards ambitious deregulation, allowing a variety of new business models to emerge. This deregulation in the progress of broadband service markets is, however, not fully enough to cope with the rapidly changing market environment. Indeed, the future market, where the transition from PSTNs to IP networks is considerably developed, is causing new problems from the viewpoint of competition policies. Thus, as the second stage, legacy competition policy tools should be reconsidered.

The following sections outline the two stage approach in Japan to cope with change in the market structure caused by the transition from PSTNs to IP networks.

Regulatory reform at the 1st stage — The revision of the TBL

The present Japanese categories of Type-1 and Type-2 telecommunications businesses have been a basic structure of regulatory frameworks in the telecommunications business field. In the current regime, a Type-1 carrier has its own facilities to provide services and a Type-2 carrier provides services on a resale basis. This scheme has been based on the recognition that facility- based carriers are large scale providers which install their own network infrastructure with social and economic importance, whereas Type-2 carriers are recognized as small size carriers providing value added services based on the services purchased from Type-1 carriers. This basic recognition, however, has been rendered out of date due to the development of IP technology. For example, even resale providers without any facility can provide IP telephony at a nationwide level and some Type-2 carriers are larger than some Type-1 carriers. In this context, it cannot be said that the functions of Type-1 carriers are envisaged to be more important socially and/or economically than those of Type-2 carriers, eventually making it difficult to find the rationality that regulations posed on Type-1 carriers should be stricter from a socio-economic point of view.

Corresponding to these recent changes in the market environment, in August 2002, the Telecommunications Council, an advisory body to the Minister of Public Management, Home Affairs, Posts and Telecommunications, concluded that the time is ripe to comprehensively review the present competition frameworks (MPHPT 2002b). Following the Council’s report, the revision of the Telecommunications Business Law (TBL) has been approved by the Diet in July 2003 and is expected to take effect by the year 2004. This reform includes the following deregulatory measures (Figure 6). Three pillars of this reform are summarized as follows.

First, the present Type-1 and Type-2 business categories shall be abolished and market entry regulations shall be drastically relaxed, where permission systems pertaining to Type-1 business shall be abolished.

Second, in order to facilitate the smooth deployment of network infrastructure by telecommunications carriers, ROWs have been granted so far to Type-1 carriers automatically when they get business permissions. Reflecting the abolition of Type-1 and Type-2 business categories, ROWs will be granted to carriers based on an application from the carrier after examination of business plans. Application procedures are to be streamlined and shortened.

Third, reflecting the diversity of service provisions, tariff regulations on non-dominant carriers will be abolished allowing those carriers to provide services on an individualized agreement basis. This enables carriers to make contracts with user companies flexibly and no ex-ante regulations will be posed.

In accordance with the above mentioned development, applications of dominant regulations are required to be reconsidered. In Japan, dominant carriers in the local fixed market are defined as carriers with more than 50% of fixed subscriber lines in each prefecture. In reality, both NTT East and NTT West have a 98% share of the local market and these two carriers are designated as dominant carriers, to which dominant regulations, such as prohibition of abuse of dominant position into other business fields, are applied. More specifically, once designated as a dominant carrier with bottleneck facilities, dominant regulations have been applied to every service market however hard this carrier tries to open up its bottleneck facilities.

NTT East and NTT West, however, have only a 40% share of the DSL service market following the effective interconnection rules of collocation and unbundling, and this situation is recognized as both companies remain in non-dominant positions in the DSL market. This means that it is appropriate to consider dominance of network facilities and that of service markets separately. More specifically, if a dominant position in network facilities (bottleneck facilities) is abused in some specific service market, dominant regulations shall be applied to that carrier in that market. However, if that carrier ensures the openness of its networks and enables other competitive carriers to come into this specific service market smoothly, this service market can be recognized as effectively competitive and the rationality to apply dominant regulations to the carrier is lost.

This approach means conceptually the separation of physical network layer regulations and service layer regulations even within the telecommunications services and the application of dominant regulations can be decided based on review of the effective competitiveness of each market. Thus, a periodic competitive review of each market becomes one of the key factors for operation of dominant regulations. Especially under circumstances of service integration based on the development of IP technology, the methodology of how to define each market is an important issue to be addressed. This study has been conducted by the Study Group set up at the MPHPT in September 2002 and the final report on the market review methodology was compiled in July 2003. The result of this study shall be reflected in the policy of how to determine the operation of dominant regulations.

Regarding this competitive review, one additional consideration is how to make distinctions of market definitions used for the Antimonopoly Act (AMA) and for the TBL. In the case of the AMA under the jurisdiction of the FTC (Fair Trade Commission), market definition is to be carried out by each specific case such as a merger case in a recognized static market. However, in the case of market definition for the TBL, the regulatory authority is required to make presumptions because competition regimes are forward-looking and the Japanese regulatory authority, or the MPHPT, deals with current and potential market structures simultaneously to promote competition.

Regulatory Reform at the 2nd Stage — New competition rules in the IP age

Following the first stage of change of regulatory schemes represented by the revision of the TBL, the MPHPT is already standing at the entrance of the second stage to reconsider the current regulatory regime where the development of IP based services is considerably accomplished. In this section, some examples of issues to be explored, which were raised in the report of the study group at the MPHPT on new business models and the grand design of the competitive environment, are summarized in the following section (MPHPT 2002a).

First, more focus should be put on competition policies for emergence of vertically integrated business models. Telecommunications service providers have been providing their services on an end-to-end basis. The progress of IP technology, however, has caused the functional separation of services. For example, the CDN (Content Delivery Network) business is one example born in the functional diversity of services. Newly emerging services, such as customer authentication and billing services for content delivery, while not telecommunications services per se, are closely related with telecommunications services and, in many cases, are provided as one package. In other words, evolution from telecommunications services to broadband services does not fit the analytical framework ensuring effective and fair competition only in the telecommunications services. Rather, emergence of new business models requires new analytical frameworks consisting of different layers or "layered competition models."

One example of this layered competition model can be found in the business model of "i-mode" provided by NTT DoCoMo. This mobile phone based internet access service provided by NTT DoCoMo and other mobile phone companies has achieved a remarkable reputation in Japan and the number of users has reached 62.46 million, which is about 83% of the overall number of mobile phone subscribers (as of the end of March 2003). Here, these mobile phone companies:
– sell terminals with their own brand;
– provide telecommunications services including internet access service; and,
– provide platform functions for content delivery including billing and authentication when subscribers access and/or use content and applications. Each content and application has been developed on a platform for each mobile phone company and no inter-accessibility across different carriers has been ensured.

Considering this business model, it could be possible to apply layered competition models (Figure 7) consisting of layers, such as:
– terminal layer;
– network layer, where fundamental transmission of packetized information etc. is carried out;
– platform layer including such functions as authentication and billing to allow content providers to offer their content and application; and,
– content/application layer.

In reality, NTT DoCoMo’s case is the vertical integration of enhancing the business field from the network layer to the content/application layer. Yahoo Japan, which initiated their business at the content/application layer, is another example of vertical integration to have enhanced their business to the network layer through the establishment of a subsidiary broadband service company called BB Technology. In addition, many collaborative projects between telecommunications service providers and other business entities including broadcasting companies have been carried out (Figure 8).

Integrating vertically the functions at the network layer, the platform layer and the content/application layer requires new aspects for competition policies. The current competition policy focuses only on the fair and effective competition environment at the network layer — network facilities and telecommunications services. New business models, however, are vertically integrated across different layers and the functions at each layer are provided as a single function. In addition, this integration makes it difficult to distinguish between telecommunications services and information services. Even if any artificial distinction is made, rapid progress of technology makes this distinction soon obsolete. Thus, instead of this distinction, when considering applicability of appropriate regulations from the viewpoint of promoting competition, "openness" of interfaces between each layer should be ensured. For example, if a platform layer has been monopolized by one dominant carrier in the network layer, delivery of the content and/or applications might be controlled by this dominant carrier and it hinders the development of content and/or applications industry. Actually, it is not appropriate to enhance the coverage of regulations into this upper layer business, but at least the possibility of dominant positions in the network layer abusing upper layers shall be avoided through establishing measures for "open" interfaces.

Second, thanks to IP technology development, the traditional distinction of "service" categories will be lost. For example, distinctions of "distance" in telecom services will be lost. More specifically, legacy voice service has been notionally separated into "local," "long distance," and "international" calls. IP based service, however, makes this distinction meaningless. Furthermore, in IP networks, the distinction of "transmitted content" will be lost because voice, data, or video is transmitted as packets and eventually the unification of services will take place. Service integration makes it more difficult for regulatory authorities to judge if the retail price for each service is truly based on cost structure and does not generate excessive profit. Here, for example, the cost structure of platform functions such as billing and authentication shall be reviewed if necessary to avoid cross subsidization between network layer services and platform layer services in case of being provided simultaneously by dominant carriers.

Third, in line with the progress of the unification of services mentioned above, the concept or the coverage of "universal service" shall be reconsidered. One approach might be to redefine, periodically, the scope of "universal service," but the definition of essential services indispensable to social and economic activities is difficult, and the progress of technology might make this definition obsolete very soon. Instead, as an alternative method, the concept of "universal access" should be considered. The purpose of "universal access" is to ensure an environment where every household, in rural areas as well as metropolitan areas, can be prepared to enjoy broadband services using fiber optic networks or equivalent networks. This concept does not contain the coverage of broadband services. In other words, physical accessibility to broadband services is put as the main point of service universality. More specifically, the FTTC at the nationwide level is one desirable image from the viewpoint of "universal access" and the cost to maintain FTTC in rural areas is envisaged to be financed by the USF (Universal Service Fund).

Fourth, interconnection rules shall be reconsidered reflecting recent interconnection contracts. Under the PSTNs regime, access charges have been set in a time or volume sensitive manner. The interconnection of IP networks, however, has been "peering" or "transit," where a variety of compensation methods are applied. Under these circumstances, new interconnection rules for dominant carriers in IP networks shall be established to set compensation among carriers. For example, more focus should be put on the introduction of fixed rate access charges instead of current traffic sensitive access charges.

In addition, proliferation of "best effort" services causes the misperception of service quality among end users. For example, DSL service with the nominal speed of 1.5Mbps sometimes ensures the effective speed of 500 kbps at maximum, and it varies by provider or by time zone usage. To ensure the credibility of the quality of new services, an SLA (Service Level Agreement) concept shall be introduced into services for mass users as well.


The development of IP technology is driving the emergence of new broadband business models allowing users to enjoy new broadband services. In Japan, a combination of broadband infrastructure deployment policies and a change in competitive regimes towards deregulation has realized one of the most advanced broadband infrastructures in the world. However, the historical change of network structures from PSTNs to IP networks may challenge the regulatory authority to review policy tools to promote competition in the broadband service market. Particularly, the emergence of vertically integrated business models requires new study on how to ensure a fair and effective competition environment at each layer. In addition, deployment of a broadband infrastructure is not the final destination for service providers, users and policy makers as well. That is, how to stimulate the demand for broadband content and applications, and how to ensure the smooth delivery of attractive content and applications should be explored. In this context, in July 2003, the IT headquarters of the Japanese government put forth "e-Japan strategy II," aimed at more growth in content and application industries corresponding to the broadband infrastructure. Keeping in mind the necessity for minimization of regulations under a rapidly changing broadband market environment, the next steps for the establishment of desirable broadband policies are to be explored in the next several years.


MPHPT (Ministry of Public Management, Home Affairs, Posts and Telecommunications, Japan) (2001), Guidelines for Use of Poles, Ducts, Conduits and Similar Facilities Owned by Public Utilities, <>

MPHPT (2002a, June 8), Study Group on New Business Models and the Grand Design of the Competitive Environments for the New Information and Communications Era, How Competitive Environments in the Telecommunications Business Field Should Be Established in the Broadband Age, <>

MPHPT (2002b, August 7), Telecommunications Council, Final Report on Desirable Pro-Competitive Policies in the Telecommunications Business Field for Promoting the IT Revolution, <>



I am grateful to the members of the Telecommunications Council and the Study Group on New Business Models and the Grand Design of the Competitive Environments for the New Information and Communications Era, on which this paper is based. The views expressed in this paper, however, are those of the author.

About the Authors

Yasu Taniwaki is Economic Counselor and Telecommunications Attache, Embassy of Japan in Washington DC. He joined the Ministry of Posts and Telecommunications (Ministry of Public Management, Home Affairs, Posts and Telecommunications) in 1984. After serving in several positions, including ICCP Division of the OECD (1987-1989), he served as Deputy-Director of the Telecommunications Policy Division (1993-1997), where he dealt with several telecommunications policies such as the reorganization of NTT, and a variety of deregulation programs of the Telecommunications Business Law (TBL). After serving as Secretary to the Minister of Posts and Telecommunications (2000-2001), he served as Director for Telecommunications Policy, where he drafted a report on new Japanese telecom competitive schemes compiled by the Telecommunications Council in August 2002 and a report on layered competition models in the IP age, compiled by the Study Group on New business Models in June 2002. He also contributed to setting up the Telecommunications Dispute Settlement Commission and Japanese Universal Service Fund, as well as to introduce the concept of "dominant regulations" through revision of the TBL in 2001. He has served in his capacity since June 2002. His e-mail address is: [email protected].

Copyright © 2003 Journal of Interactive Advertising