9-805-124
REV: JUNE 7, 2006
STEPHEN P. BRADLEY, THOMAS R. EISENMANN
MASAKO EGAWA, AKIKO KANNO
NTT DoCoMo, Inc.: Mobile FeliCa
With more than 49 million customers and a market share of 56% in Japan, NTT DoCoMo, Inc. was the world’s second-largest and Japan’s top mobile communications company. Nevertheless, DoCoMo faced several challenges in its core business in December 2004. First, with nearly two-thirds of the
Japanese population already owning a mobile phone, the market was becoming saturated. Second,
DoCoMo’s growth in data services would be limited after it matched an offer by its largest rival,
KDDI, of flat-rate rather than traffic-based pricing. Finally, rivalry in the mobile phone industry would escalate following two expected regulatory changes: the licensing of one or two new entrants and the introduction of number portability.
With DoCoMo’s first wave of growth (mobile voice) subsiding and its second wave (mobile data) cresting, management believed mobile FeliCa could unleash the third wave of DoCoMo’s evolution:
“lifestyle infrastructure” services. FeliCa, developed by Sony, transferred information by radio over short distances—about 10 centimeters—between a contactless integrated circuit (IC) and a “reader.”
By building FeliCa into its handsets, DoCoMo could move beyond telecommunications into many new applications. For example, customers could hold mobile FeliCa phones close to a reader to quickly and easily make payments in stores, ride commuter trains, or secure entry to buildings.
DoCoMo’s first FeliCa phones were introduced in July 2004. Five months later, Masao Nakamura, appointed CEO in June, was meeting with Kei-ichi Enoki, executive vice president and managing director, products and services division, to discuss how DoCoMo should drive mobile FeliCa adoption. Enoki had headed the very successful launch of DoCoMo’s i-mode data service.
Enoki and Nakamura were debating the relative importance of three potential sources of value from mobile FeliCa. The first was improved mobile phone churn and subscriber acquisition rates. The second was profit from FeliCa Networks, a joint venture between Sony and DoCoMo that would license FeliCa technology to DoCoMo’s mobile phone rivals and provide services to application providers. The third was revenue DoCoMo could earn directly by providing mobile FeliCa services.
Financial services such as “eMoney” and credit card payments looked particularly promising, but
DoCoMo management had to decide how aggressively to pursue them. The opportunity could be enormous, but DoCoMo lacked experience in this arena and would face fierce competition. Should
DoCoMo launch its own payment services or seek to ally itself with established players such as JCB and Visa?
________________________________________________________________________________________________________________
Professors Stephen P. Bradley and Thomas R. Eisenmann prepared this case with assistance from Masako Egawa, Executive Director of the Japan
Research Office, and Akiko Kanno, Research Associate at the HBS Japan Research Office. Some portions of this case draw heavily on “NTT
DoCoMo (A): The Future of Wireless Internet?” HBS No. 701-013, by Professor Stephen P. Bradley, and “NTT DoCoMo: Marketing i-mode,” HBS
No. 502-031, by Professor Youngme Moon. HBS cases are developed solely as the basis for class discussion. Cases are not intended to serve as endorsements, sources of primary data, or illustrations of effective or ineffective management.
Copyright © 2005 President and Fellows of Harvard College. To order copies or request permission to reproduce materials, call 1-800-545-7685, write Harvard Business School Publishing, Boston, MA 02163, or go to http://www.hbsp.harvard.edu. No part of this publication may be reproduced, stored in a retrieval system, used in a spreadsheet, or transmitted in any form or by any means—electronic, mechanical, photocopying, recording, or otherwise—without the permission of Harvard Business School.
805-124
NTT DoCoMo, Inc.: Mobile FeliCa
DoCoMo Overview
NTT DoCoMo was established in 1992 when it was spun off by Nippon Telegraph and Telephone
(NTT). NTT had been privatized in 1986, but the Japanese government still held 46% of its equity in
2004. NTT in turn owned 58% of DoCoMo’s equity, which had been publicly traded since 1998.
i-mode launch. In launching “i-mode” in 1999, DoCoMo’s managers pursued a different strategy for exploiting the wireless Internet than most European and U.S. mobile carriers.
First, i-mode utilized an “always on” packet-switching technology that avoided the need to repeatedly dial up/log in and allowed users to receive incoming calls while surfing the wireless Web.
Also, i-mode users paid for the volume of data packets received or sent rather than for the time they spent online. To encourage adoption, DoCoMo charged subscribers a basic monthly fee of only 300 yen (¥) (approximately $3) plus ¥0.3 for every data packet transmitted. By contrast, the wireless application protocol (WAP) services adopted by many European carriers required mobile Internet users to dial up every time they wished to connect, charged per minute logged in, and could not simultaneously handle incoming calls and browsing.
Second, i-mode was targeted initially to trendy young consumers rather than businesspeople, even though most overseas carriers believed that business users would be less price sensitive and more likely to use premium information and transaction services.
Third, DoCoMo formatted i-mode content with compact HTML, a variant of the HTML language which was already used to encode nearly every Web site accessible using a PC browser. European and U.S. mobile carriers relied on WAP, which transferred data more quickly but required content providers to reformat their information for mobile phone users.
Fourth, while most mobile carriers in Europe and the U.S. were developing or purchasing proprietary content, DoCoMo relied on third-party providers. Content providers could charge consumers directly using DoCoMo’s billing system. DoCoMo collected 9% of the revenue that content providers earned from subscribers.
Fifth, DoCoMo was able to quickly offer many i-mode handset models. Japanese carriers commissioned handset designs from manufacturers and bore inventory risk, which gave the carriers leverage to dictate features and specifications, whereas in the U.S. and Europe, handsets were designed and marketed independently by manufacturers with powerful brands, like Nokia.
Within six months of its launch, i-mode had attracted over 2 million subscribers. By late 2001, it had over 30 million subscribers. At that time, more than 3,000 content partners, having met
DoCoMo’s rigid quality standards, were listed as official content providers (CPs) on i-mode’s menu.
An additional 50,000 sites—nonofficial CPs—were accessible by typing a URL into the phone’s browser. i-mode was a big financial success (see Exhibit 1 for financial statements). About half of imode users paid extra fees (¥100 to ¥300) to subscribe to at least one premium content service, and imode users spent more on voice calls and were less likely to switch to other carriers.
International expansion. Encouraged by i-mode’s success in Japan, DoCoMo resolved to take i-mode and higher-speed third-generation (3G) data services overseas. Between late 1999 and 2001, the company spent billions of dollars to acquire minority stakes in U.S.-based AT&T Wireless and other overseas mobile carriers (see Exhibit 2 for DoCoMo’s overseas investments). However,
DoCoMo’s international expansion strategy encountered difficulties. 3G licenses required large investments. Also, outside Japan, carriers lacked the power to dictate handset specifications. Finally, many consumers preferred to access the Internet via a PC rather than a two-inch screen. For these
2
NTT DoCoMo, Inc.: Mobile FeliCa
805-124
reasons, many of DoCoMo’s overseas partners did not adopt i-mode. To make matters worse, the value of DoCoMo’s investments plunged with the global collapse of telecommunications stocks.
DoCoMo was forced to write off ¥1.5 trillion of the ¥1.9 trillion ($17 billion) invested overseas.1
In June 2004, Nakamura announced that DoCoMo would no longer aggressively pursue overseas investments but instead would seek technological partnerships. 2 After Cingular announced its merger with AT&T Wireless in October 2004, DoCoMo sold its 16% stake in AT&T Wireless but began working with Cingular to unify specifications for 3G networks. By the end of 2004, DoCoMo had negotiated partnership agreements to provide i-mode technology to 12 carriers (see Exhibit 3 for
DoCoMo’s overseas partnership agreements).
Current Situation
Market maturation. The Japanese mobile phone market was becoming saturated, reaching a
65.7% per capita penetration rate in 2004 (see Exhibit 4 for Japanese mobile subscriber growth). The industry’s average revenue per unit (ARPU) for voice service dropped from ¥6,320 per month in
December 2002 to ¥5,803 at the end of 2003.3 This decline was due to rate reductions, late adopters who used their mobile phones less frequently, and subscribers using mobile e-mail or text messaging instead of calling.4
Data service ARPU was still growing and averaged about ¥2,000 per month. But data use had a skewed distribution. Makio Inui, managing director and senior analyst at UBS investment research, explained: “Most users pay around ¥600 per month, but the top 20% of users bring in 80% of revenue.
This means that the shift to flat-rate data pricing in 2003, led by KDDI, will significantly reduce revenues unless carriers can attract new subscribers, stimulate data traffic among light users, or find new revenue sources.”5
3G services. By early 2002, i-mode subscriber growth was slowing, while data services launched by DoCoMo’s Japanese rivals KDDI and Vodafone were gaining traction. To maintain its lead, DoCoMo introduced next-generation high-speed data services. In October 2001, it became the first carrier in the world to launch 3G services, branded as Freedom of Mobilemultimedia Access
(FOMA). FOMA mirrored i-mode’s business model, adding high-bandwidth content such as video and music streaming. The service employed wideband code division multiple access (W-CDMA), which DoCoMo expected to become one of two global 3G standards. FOMA transmitted data on the downlink (server to handset) at a maximum 384 kilobytes per second (kbps); the uplink speed was 64 kbps. By contrast, personal digital cellular (PDC), the 2G standard used for i-mode, had a downlink speed of 28.8 kbps.
DoCoMo had 89,400 FOMA users in March 2002, short of its 150,000 target. Demand was limited by some technical issues that could eventually be resolved, such as short battery life and poor reception in subways or inside buildings. High handset prices—¥30,000 to ¥50,000 per FOMA handset—also reduced demand among young consumers.
In April 2002, KDDI launched its next-generation service (2.5G) based on Qualcomm’s
CDMA2000-1X standard, which was incompatible with W-CDMA. Although KDDI’s 144 kbps transmission speed was slower than FOMA’s 384 kbps, KDDI’s handset was more compact, less expensive (below ¥20,000, even for advanced camera-equipped units), and had global positioning system (GPS) capabilities. Moreover, KDDI’s next-generation service could use the company’s existing network facilities, which provided better geographic coverage upon launch. DoCoMo, by contrast, had to build new network facilities for FOMA.
3
805-124
NTT DoCoMo, Inc.: Mobile FeliCa
KDDI relied on Qualcomm’s binary runtime environment for wireless (BREW) platform. BREW gave Qualcomm customers like KDDI the ability to download applications into phones and to manage authorization and billing functions; in exchange, Qualcomm earned a share of application revenue. Executive vice president Enoki commented:
KDDI’s fate depends on how well Qualcomm supports BREW, CDMA2000, and its other
3G technologies. Qualcomm sees itself as a mobile version of Wintel [the Windows/Intel PC hegemony] and may be aiming for control not only over chips and standards but also— through BREW—over services as well. DoCoMo is simply not willing to link its fate so closely with that of a single hardware or software vendor.
KDDI’s next-generation service grew rapidly and had 6.8 million subscribers by March 2003, compared to only 330,000 for FOMA and 25,000 for Vodafone’s 3G service, launched in December
2002.6 In November 2003, KDDI pressed its advantage by introducing flat-rate pricing for data services. For the first time ever, from April 2003 through March 2004, KDDI’s net increase in subscribers (gross subscribers added, minus cancellations) exceeded DoCoMo’s net increase; they were 2.9 million and 2.1 million, respectively.7
Alarmed by KDDI’s gains, DoCoMo offered flat-rate pricing for FOMA in June 2004. By
September 2004, DoCoMo’s aggregate ARPU (3G and 2G combined) had declined from ¥7,890 in
March 2004 to ¥7,340.8 However, DoCoMo’s net increase in 3G subscribers between April and
September of 2004 did exceed KDDI’s (see Exhibit 5 for subscriber growth by company).
Regulatory challenges. Regulations allowing Japanese mobile phone subscribers to switch carriers without changing their phone numbers were expected during 2006. Hong Kong and a few other markets that had already introduced mobile phone number portability had witnessed a significant market share impact, but in most markets the impact had been modest.
However, data services were more important in Japan than in other countries, and there were significant differences in data service quality between Japanese carriers. Furthermore, subscribers to mobile data services had carrier-specific e-mail addresses (e.g., xxxxxxx@docomo.ne.jp) that would not be portable.9 Despite these mitigating factors, number portability in Japan was expected to increase marketing costs, reduce prices, and boost subscriber churn. In late 2004, DoCoMo’s monthly churn was about 1%, whereas its rivals averaged 2% to 3% churn.
In the fall of 2004, Japan’s Ministry of Internal Affairs and Communications announced that it would allocate new spectrum for mobile services. DSL providers Softbank and eAccess had expressed interest and were expected to receive new licenses. If Softbank entered the mobile phone business, it seemed likely to follow the same aggressive pricing strategy it had successfully employed with DSL.
DoCoMo’s response to challenges: Lifestyle infrastructure. Faced with these challenges,
DoCoMo needed to find new revenue sources and to bolster subscriber retention. Enoki explained:
A mobile phone is something that is always with the user. Some people here say that the mobile phone can become “lifestyle infrastructure.” I like to think of the phone like a TV remote control for all the transactions in our daily lives—paying a taxi fare, boarding a plane, pulling up personal files on a conference room PC when you enter—you name it. We are adopting new technologies such as infrared, QR code readers, and FeliCa contactless ICs to turn our phones into lifestyle infrastructure.
4
NTT DoCoMo, Inc.: Mobile FeliCa
805-124
Along these lines, DoCoMo and Visa International were jointly developing a mobile phone-based payment system that used infrared data transmission. A downloadable application would store a user’s credit card data and facilitate payment via infrared data transmission to a merchant’s receiver unit. This infrared receiver unit would be substantially less expensive than the readers used for contactless ICs (described below).
Likewise, DoCoMo had embraced “QR codes” that could store several hundred times more data than conventional bar codes in the same amount of space. A range of innovative services could be offered with QR code readers, which added little to the cost of camera-equipped mobile phones. For example, by scanning a QR code printed in a magazine or on a poster, users could obtain coupons, phone numbers, and maps, or access a Web site without having to enter the URL (see Exhibit 6 for an example of QR codes).
FeliCa Overview
Among the new technologies DoCoMo was exploring in order to pursue business opportunities beyond telecommunications, FeliCa (based on the word “felicity”) appeared to be the most promising.
FeliCa was a contactless IC technology developed by Sony and originally built into plastic cards the size of a credit card. FeliCa cards required no battery; they were activated by radio waves sent by a reader/writer device. The user could transfer information by just holding the card close to the reader/writer. Reader/writers would cost between a few hundred dollars and $1,000, depending upon whether features such as screens and input keys were included. The cost of the FeliCa chip itself was not disclosed (see Exhibit 7 for a picture of a FeliCa chip).
FeliCa’s data-transfer speed of 212 kbps was almost twice that of other contactless ICs; the entire process of card detection, authentication, and data transfer could be completed in 0.1 seconds. The IC held individually managed files that could be allocated to different applications. Security was provided by a sophisticated encryption system that generated separate keys for each application.
FeliCa’s fast processing made it ideally suited for use as “electronic money” and for commuter ticketing. Since FeliCa supported multiple applications, a single card or FeliCa-equipped phone could be used as an employee ID for office entry, eMoney, subway fares, and other applications.
FeliCa History
Sony began work on contactless ICs in 1988. In 1995, Hong Kong authorities adopted FeliCa for
“Octopus” cards in their public transportation system. Similar systems were subsequently sold in
China, India, and Singapore. In March 1999, Sony began trial use of the FeliCa card as an employee
ID and “e-wallet” in a Tokyo office and shopping complex. In January 2001, Sony backed an eMoney system called Edy (described below). Despite these initiatives, FeliCa struggled to penetrate the
Japanese market.
FeliCa card usage in Japan finally surged after East Japan Railway (JR East) adopted the technology for its commuter railroad network in the Tokyo area. Within one year of launching its
FeliCa system in November 2001, JR East had distributed over 5 million FeliCa-enabled Suica cards.
(Suica was an abbreviation of “Super Urban Intelligent Card.”)
Encouraged by Suica’s success, Sony, JR East, and DoCoMo discussed the possibility of embedding FeliCa technology in DoCoMo phones. By March 2003, the companies had confirmed that this was technically feasible. However, they held different views about the business model for mobile
5
805-124
NTT DoCoMo, Inc.: Mobile FeliCa
FeliCa. Sony managers were eager to earn near-term returns, having already made significant investments in FeliCa. DoCoMo, on the other hand, did not wish to incur a “first-mover disadvantage” by paying a high price for chips while bearing a pioneer’s technical and business risk.
In that scenario, once mobile FeliCa was established, DoCoMo would be disadvantaged in competition with “free riders”—rivals who could avoid pioneering risks and investments but pay the same price as DoCoMo for FeliCa chips. Recognizing the importance of DoCoMo’s support, Sony promised to keep the chip price low.
This was an important step, but it was still unclear how DoCoMo could profit from FeliCa. With both voice and data services, DoCoMo earned most of its revenue by charging users for traffic. FeliCa use, however, would not significantly increase telecommunications traffic. Takeshi Natsuno, senior manager of multimedia services, who led the DoCoMo team negotiating with Sony, conceived a plan to earn profits from mobile FeliCa through a joint venture between Sony and DoCoMo. In May 2003, the companies agreed to establish a company called FeliCa Networks.
FeliCa Networks
FeliCa Networks, Inc. was established with ¥5.8 billion in capital and a 60%/40% equity ownership split between Sony and DoCoMo. DoCoMo contributed 20 of the new company’s 80 employees, while Sony supplied intellectual property and the balance of the team. In June 2004, JR
East acquired a 5% stake in the company for ¥1.05 billion, reducing the ownership of Sony and
DoCoMo to 57% and 38%, respectively.
The plan was for FeliCa Networks to have three sources of revenue. First, it would collect license fees from carriers that purchased mobile FeliCa chips. DoCoMo was exempt from such fees, which would offer a competitive advantage if rivals chose to offer FeliCa-equipped phones. Second, FeliCa
Networks would provide platform management services. Applications were not preinstalled on mobile FeliCa chips; rather, users downloaded them into a 5-kilobyte memory area that had room for about 5 to 10 applications. In exchange for managing this memory area and for serving the encryption keys needed to activate applications, FeliCa Networks received a fee every time a user downloaded an application. Third, FeliCa Networks could provide application providers with a range of hosted services such as managing the servers used to download applications or to authenticate users. For most such services, FeliCa Networks would collect transaction fees. However, large application providers might prefer to rely on in-house units rather than FeliCa Networks for these functions.
Mobile FeliCa Adoption Prospects
Between December 2003 and June 2004, DoCoMo conducted a market trial using 5,000 FeliCaequipped phones. The following month, DoCoMo released its first FeliCa-equipped phones for the general public. By December 2004, DoCoMo had sold 1.3 million mobile FeliCa phones.
Value proposition. FeliCa phones offered several advantages over FeliCa cards. Soichi
Kawachi, president of FeliCa Networks, explained:
Mobile phones offer several functions that are missing on a contactless IC card, including a screen, a keyboard, and a network connection. FeliCa phones can display one-to-one marketing messages sent from the reader/writer. For example, if there are unsold seats for a concert, a last-minute promotional message could be sent to FeliCa mobile phones through reader/writers when users make eMoney payments at a nearby Starbucks.
6
NTT DoCoMo, Inc.: Mobile FeliCa
805-124
Some of these functions could be matched with card-based technology by incorporating screens, keyboards, and network connectivity directly into reader/writers. However, in a co-marketing program offered with Asano Supermarket, FeliCa phones proved to be more effective. Kiyoyuki
Tsujimura, DoCoMo’s executive vice president of corporate strategy and planning, explained:
Asano Supermarket has a loyalty program that offers a maximum 5% discount to customers who make cashless payments using FeliCa. The discount is determined based on each customer’s monthly spending. Revenue has increased because customers are willing to buy an extra radish in order to reach the threshold for a bigger discount. Asano had initially used
FeliCa cards but added support for FeliCa phones. People always carry their mobile phones. It will certainly be more convenient for the customers if they can see on a FeliCa phone how much more they need to buy to qualify for a bigger discount while they are still shopping, before they get to the checkout line.
Security concerns. Security concerns posed a potential barrier to the widespread adoption of mobile FeliCa. When FeliCa phones were equipped with eMoney, building IDs, and other applications, losing one’s phone would be like losing one’s wallet. Takashi Sakamoto, DoCoMo’s executive vice president and managing director, marketing, expressed his concern: “When people lose a wallet with ¥50,000 in cash, they don’t expect that money to be returned. But if they lost a
FeliCa phone, would they have the same expectation? DoCoMo’s reputation could be at stake if people hold us responsible for their financial loss.”
DoCoMo had adopted Edy for its e-wallet function. The maximum value that could be stored in
Edy was ¥50,000. To increase transaction speed, eMoney did not employ personal identification numbers (PINs) or communications networks to authenticate users and authorize payments. This meant that a smart card loaded with eMoney could not be disabled if it was lost or stolen. DoCoMo was able to avoid this problem by equipping FeliCa handsets with a remote lockup feature. In case of theft or loss, users could make a call to DoCoMo customer service to lock their phone’s functions.
Competing technologies.
FeliCa competed with “Type A” and “Type B” contactless IC
Type A cards were used mostly in Europe. The Japanese government had decided to use Type B technology for its residency registration program and was planning to apply the technology for driver’s licenses and passports. Compared to FeliCa technology, Type B had better security but lower transaction speed (see Exhibit 8 for description of competing contactless IC technologies). technologies.10
In November 2003, Sony, Visa, and Infineon announced an alliance to develop a single chip that would integrate FeliCa’s contactless IC with the contact IC technologies widely deployed by credit card issuers. The alliance would leverage standards from GlobalPlatform—an international group of more than 50 card issuers, manufacturers, and vendors that jointly developed specifications for smart cards. 11
Strategic Rationale for DoCoMo FeliCa
It was not clear how much DoCoMo would ultimately need to invest to ensure mobile FeliCa’s success. Natsuno noted: “To date, DoCoMo’s total investment in FeliCa technology has not been large because the basic technology was already developed by Sony.” However, DoCoMo had set aside ¥20 billion for subsidizing small merchants’ installation of reader/writers. Merchants received the subsidized reader/writers in exchange for a small fee for each FeliCa transaction.
7
805-124
NTT DoCoMo, Inc.: Mobile FeliCa
In addition to improving churn and subscriber acquisition rates in DoCoMo’s core business, mobile FeliCa would deliver profits from DoCoMo’s equity in FeliCa Networks and from new revenue streams. Natsuno explained the strategic significance of new revenue streams: “Should we seek to grow substantially, or strive in our maturing market to simply maintain our current share? If we want to grow, we must expand beyond telecommunications. Inevitably, that raises questions about whether we have the resources and capabilities required to enter new markets and what types of risks we should be willing to take.”
Churn Reduction/Subscriber Acquisition
Increased switching costs. Mobile FeliCa could help DoCoMo control churn; this was a priority, given the expected introduction of number portability. If customers wished to switch to a new carrier, they would have to first deplete any stored eMoney on their old phone and then reinstall all their FeliCa applications to their new phone. However, the magnitude of switching costs was not clear. Kawachi of FeliCa Networks pointed out: “Personal information will be managed by the application providers, who should be able to automate the reinstallation process and transfer any personal information or customized settings into the new phone. Of course, users will still need to download the applications, but doing so should not require a big time investment.”
Increased differentiation vis-à-vis mobile rivals. DoCoMo could capture market share if it offered FeliCa as an exclusive feature. At the same time, DoCoMo managers recognized that, with strong network effects, FeliCa applications and reader locations were more likely to proliferate if
DoCoMo’s rivals also adopted the technology. For this reason, DoCoMo executives decided not to seek exclusive rights for FeliCa. Nakamura said: “We did not inhibit KDDI from participating in
FeliCa. Even if Softbank enters the mobile phone business, we will not create any obstacles.”
Tsujimura added: “If customers switch from DoCoMo to KDDI, they would be able to use the same set of basic applications like Edy, but they would not be able to use DoCoMo-exclusive upper-level services, for example, our loyalty-points program.”
Other carriers were expected to offer contactless IC services. KDDI had announced plans to launch mobile FeliCa during 2005. Vodafone’s intentions were not clear, but in May 2004, it had unveiled a prototype handset that used flash memory cards with embedded contactless ICs.
Profits from FeliCa Networks
DoCoMo’s equity stake in FeliCa Networks was a potential source of both profits and competitive advantage. Nakamura commented:
With support expected from JR East and other rail companies and eventually from all the mobile carriers, it is true that FeliCa could become a de facto standard. Given the rapid replacement rate for mobile phones in Japan, FeliCa could soon be built into 80 million handsets. Since applications in those handsets must be activated through FeliCa Networks, and since our rivals must license technology from FeliCa Networks, our joint venture may have considerable power, and it might become very profitable. We welcome that success, but to protect DoCoMo’s reputation, we must never abuse our position.
As a major shareholder of FeliCa Networks, DoCoMo also had the advantage of learning about new mobile FeliCa applications before competitors. This meant that DoCoMo had launch-timing advantages and conceivably a head start in negotiating for exclusive rights. However, since most
8
NTT DoCoMo, Inc.: Mobile FeliCa
805-124
application providers would want ubiquitous coverage, DoCoMo might not be able to afford the premium required to secure exclusivity.
FeliCa-Based eMoney
Among the many mobile FeliCa applications that could generate profits for DoCoMo, payment services appeared to be especially promising. FeliCa technology was already used for electronic money, that is, digitally stored value that could be used like cash and replenished. With eMoney, users were freed from fumbling for coins, signing receipts, or waiting for change or for online authorization. Quick transactions were appealing for both consumers and merchants, especially in locations where speed was important, such as convenience stores, taxis, supermarkets, and news kiosks. Contactless IC chips also could store customer data, which enabled merchants to offer loyaltypoints programs.
Merchants assumed that accepting eMoney could boost their market share at the expense of rivals who did not do so. Merchants had no liability for fraud with eMoney, whereas with credit cards they were responsible for using the credit card company’s systems to prevent unauthorized transactions.
However, eMoney was not likely to be used for larger transactions, for which speed was less of a concern and for which authorization was imperative due to fraud risk.
The cost of installing eMoney reader/writers in retail shops was a potential hurdle for FeliCabased eMoney. With credit cards, merchants could use a single device to process transactions from all the major cards (e.g., Visa, MasterCard, American Express, and Japan’s JCB). By contrast, as of
November 2004, each eMoney provider required a unique reader/writer. The technical challenge of integrating multiple eMoney providers’ reader/writers into a single device was not great, but eMoney providers might resist integration for strategic reasons.
In addition to the cost of reader/writers, merchants that accepted eMoney incurred small transaction fees of about 2% to 3% of the transaction amount in the case of Edy.12 These fees were lower than credit card fees, which averaged 3% to 5% for small merchants.13
Potential eMoney Providers
In Japan, there were two providers of prepaid eMoney: Edy and Suica. JCB, a large credit card company, also offered a “post-payment” eMoney service called QUICPay.
Edy. In January 2001, 11 Japanese companies, including Sony, DoCoMo, and several major banks and equipment vendors, jointly established an operating company called bitWallet, Inc. to manage Edy (an abbreviation of “Euro, dollar, yen”). By 2004, 57 companies had invested in Edy, with DoCoMo holding 13% of Edy’s equity. Edy grew slowly at first; only 550,000 cards were issued in its first year of operation. Growth picked up in 2003 after a variety of Edy cards and Edy-linked services became available.
Edy cards were sold at am/pm convenience stores for ¥300 (about $3). Value could be replenished using Edy chargers (or, in some cases, cash register reader/writers) in retail stores. Some credit cards, loyalty-program cards, and employee ID cards had Edy built in. For example, All Nippon Airways
(ANA) issued Edy-enabled frequent-flyer cards through which mileage points could be exchanged for Edy credits. By October 2004, 6.5 million cards and FeliCa mobile phones were Edy-enabled and there were about 12,000 member stores where Edy could be used.14
9
805-124
NTT DoCoMo, Inc.: Mobile FeliCa
Suica. In November 2001, Suica service became available in 424 JR East train stations located within 100 kilometers of central Tokyo. Users paid a refundable ¥500 to buy a Suica “commuter pass” or “I/O card” and could load either type with up to ¥20,000. The commuter pass (unlike I/O) held the user’s identity, so eMoney balances on the pass could be recorded each time the user accessed a network-connected fare reader at a JR East station. Balances could be refunded if the user reported the pass lost or stolen.15 Passengers found Suica to be very convenient; a questionnaire showed that
10% said that they were more likely to use JR trains due to Suica.16
By October 2004, there were 10 million Suica cardholders, a large fraction of the 14 million daily JR
East passengers in the Tokyo area. JR East was negotiating with as many as 60 rail, subway, and bus companies in the Tokyo metropolitan area to integrate Suica with their fare cards. This project was eventually expected to yield about 26 million Suica-compatible cards. JR East was also planning to offer Suica on DoCoMo’s FeliCa mobile phones starting in January 2006.
In March 2003, JR East introduced Suica cards with electronic money functions (the prepaid limit was ¥20,000). In June that year, JR East issued Visa credit cards (branded as “View”) that incorporated Suica train fare and electronic money functions. As of December 2004, about 50% of the
10.5 million Suica cards issued had eMoney functionality. This slightly lagged Edy’s 6.5 million cards, but only nine months had passed since Suica eMoney had been introduced.
QUICPay.
JCB was the largest credit card company in Japan and the world’s fourth-largest proprietary card system after Visa, MasterCard, and American Express. As of 2004, JCB had 52 million cardholders, 3.7 million of whom lived overseas. JCB cards were accepted at 11.7 million service establishments, including 6.1 million locations in 188 countries outside Japan.
Credit card spending in Japan accounted for about 10% of total consumer spending—only half the
U.S. level. Eager to close this gap, JCB established JMS Co., Ltd. in 2000. The new company had a sales force of 100 working to persuade more merchants to accept JCB, American Express, Visa,
MasterCard, and Diners Club credit cards.
JCB also saw an opportunity to boost credit card use by incorporating a quick-payment capability into its cards. In July 2004, JCB allied with AEON Credit Services, an affiliate of a major retailer, to develop a FeliCa-based service called QUICPay (Quick & Useful IC Payment). In contrast to prepaid eMoney such as Edy or Suica, QUICPay was a post-payment service. Users could allocate up to
¥30,000 of their monthly credit card limit to be used for QUICPay payments. QUICPay users avoided waiting for online authorization or signing receipts and could earn loyalty points from JCB partners.
QUICPay charges were added as a lump-sum total to the user’s monthly credit card bill. If a
QUICPay credit card was lost or stolen, JCB could deactivate the service if its systems detected an obvious difference in a user’s spending habits.17
JCB planned to install QUICPay readers in supermarkets, convenience stores, food courts in department stores, and other locations where speed was important and cash payments were dominant. JCB and AEON invited other credit card companies to participate in their venture. Since
QUICPay used FeliCa technology, the application could be downloaded into FeliCa mobile phones, and a full-scale mobile FeliCa launch was expected in the spring of 2005. QUICPay would compete directly with Edy and Suica, but retailers were probably not willing to install three different FeliCa readers at checkout.
10
NTT DoCoMo, Inc.: Mobile FeliCa
805-124
DoCoMo Decisions about Financial Services Strategy
DoCoMo could promote the use of eMoney on FeliCa phones in different ways. One option was to continue to support Edy and the installation of Edy-exclusive readers. To date, all FeliCa phones had
Edy applications preinstalled. A second option was to promote eMoney reader interoperability, encouraging the coexistence of multiple eMoney providers. A third option was to merge efforts with
JR East and/or JCB in order to harness network effects.
In addition to eMoney, DoCoMo could build credit card functionality, specifically “charging” or
“revolving” payment capability, into mobile FeliCa. The balance on a “charge” card such as
American Express or Diners Club had to be paid in full each month. Credit card users “revolved” when they did not pay their monthly bill in full, incurring a loan which they paid off in monthly installments. It was not clear whether mobile phone-based credit—charging or revolving—would provide enough advantages over traditional plastic credit cards to drive consumer and merchant adoption.
Merchants would only make extra investments to accommodate mobile phone-based credit if they believed that doing so would boost transaction volumes. Because credit card payments typically were larger than cash transactions, consumers probably would wish to enter a PIN on their phone’s keyboard to initiate online authorization and protect against fraud. As a result, unlike eMoney, which was easier and faster to use than cash, mobile phone-based credit transactions could not be completed any more quickly than traditional credit card payments. Any benefits from mobile phonebased credit would have to come from using the phone’s screen and network connection to forward promotional messages or other information (e.g., the account’s available credit).
DoCoMo had at least three options for offering mobile phone-based credit. The first was to enter nonexclusive partnerships with incumbent credit card companies to build charging and/or revolving functionality into FeliCa phones. In exchange, DoCoMo could seek a portion of their transaction fees.
Along these lines, in August 2002, DoCoMo had already introduced DoCoMo credit cards in the traditional plastic card format in affiliation with eight credit card companies. Previously, DoCoMo’s loyalty-points program had been based on monthly cellular phone payments, but the DoCoMo credit cards accumulated points for other retail purchases.
DoCoMo saw opportunities to use its points program for co-marketing with partners. For example, smaller retailers or restaurants that could not afford their own loyalty programs could offer DoCoMo points, and DoCoMo could collect a percentage of their increased revenue as a promotion fee.
The second option was to seek an exclusive partnership with a single card issuer to provide phone-based credit. Competition for first-time credit card users was fierce; companies battled to become the first card in a consumer’s wallet, reasoning that this card could capture a disproportionate share of the consumer’s lifelong transaction volume. DoCoMo, with its strong brand among young consumers, could be a valuable ally in this quest.
A final option was to launch phone-based credit without a financial services partner. DoCoMo could leverage the billing system developed for i-mode transactions and simply add credit transactions to a user’s monthly phone bill.
As DoCoMo managers debated their options, they had to weigh their lack of experience in financial services. Tsujimura explained: “Credit card companies’ main profit comes from extending revolving debt to cardholders, rather than transaction fees on payments. It takes skill to determine whether a customer is creditworthy, and DoCoMo does not want to suffer big losses from bad debt because we lack that skill.”
11
805-124
NTT DoCoMo, Inc.: Mobile FeliCa
Conclusion
When DoCoMo launched mobile FeliCa, the company chose to roll out nationally rather than targeting a limited geographic market. CEO Nakamura reflected: “We had an intense internal debate about the merits of conducting more trials or rolling out regionally in stages versus launching FeliCa nationwide. There were good arguments for learning more about how consumers will use this technology and further perfecting our operations.” Sakamoto added:
FeliCa is a complicated and risky service due to the possibility of theft, loss, or malfunction.
What happens when the consumer loses a phone with ¥50,000 in stored value? How can we protect that value? What if there was a train pass on the device? In the event that a FeliCa phone is lost, which application providers should be contacted by the customer, and which by
DoCoMo? There are a countless number of contingencies, and we had to be ready to handle all of them.
DoCoMo management recognized the importance of security, but management was also concerned about being the first to market with mobile FeliCa. Nakamura explained:
Security is crucial, but most of our prospective partners, such as ANA and am/pm, were eager to offer mobile FeliCa applications nationwide. More importantly, this would prove that
DoCoMo could still be nimble. Market research shows that consumers now perceive KDDI to be more aggressive and innovative than DoCoMo. Our great success has made us defensive and arrogant. That is in the nature of human beings, but it is a big problem for us. We have begun to suffer from what I call “big corporation disease.” Our decision making has slowed, and too few managers here are willing to endorse initiatives that bear high risk but promise high potential returns. We have not been sufficiently customer-centric. For example, in our zeal to develop and launch 3G technologies, we have neglected the needs of our huge base of 2G customers. We need to change our image and reputation. FeliCa gives us an opportunity to prove we can move quickly, take smart risks, and deliver new value to our customers.
12
0.86
(116)
1,001
3,658
212
1,057
3,752
4,809
4,351
3,286
14
886
79
85
458
650
1,102
3,945
5,048
4,488
3,157
153
1,021
70
87
560
In the year ending March 2002, DoCoMo posted substantial impairment losses of ¥625 billion, net of deferred income taxes of ¥453 billion for revaluation of investments in overseas affiliates.
366
1,003
4,169
506
4,659
4,153
2004
Note:
252
777
3,909
5,172
4,107
3,266
2
716
88
35
1,065
2002
Compiled from 2002, 2003, and 2004 DoCoMo annual reports.
205
Net income
546
3,173
353
84
58
1,087
4,686
3,599
3,104
2002
U.S. GAAP Basis
2003
Source:
509
Operating income
2,610
39
80
71
732
0.30
30
115
592
Operating expenses
3,719
2,987
2,797
3,118
2,526
2,380
Total operating revenues
Operating revenues from wireless services
Cellular (2G/mova) services
Cellular (3G/FOMA) services
Packet communications services
PHS services
Other
Operating revenues from other businesses
Japanese GAAP Basis
2000
2001
1999
DoCoMo Financial Statements (billions of yen)
Year Ending March 31
Exhibit 1
805-124 -13-
805-124
NTT DoCoMo, Inc.: Mobile FeliCa
Exhibit 2
DoCoMo’s Overseas Investments (as of March 31, 2002)
Ownership
Percentage Time of Investment
Company Name
Investment
AT&T Wireless Services, Inc.
$9.8 billion
16.01%
January 2001
U.S.
KPN Mobile N.V.
$3.4 billion
15.00%
July 2000
Netherlands,
Belgium, Germany
Hutchison 3G UK Holdings Ltd.
$1.6 billion
20.00%
September 2000
U.K.
KG Telecommunications Co., Ltd.
$516 million
21.42%
November 2000 and
July 2001
Taiwan
Hutchison Telephone Company Ltd.
$440 million
25.37%
December 1999 and
May 2001
Hong Kong
Source:
Region Served
Compiled from 2003 DoCoMo Annual Report and DoCoMo, Form 20-F, July 10, 2002.
Exhibit 3
DoCoMo’s Overseas Partnership Agreements
Company Name
Country
E-Plus Mobilfunk GmbH & Co. KG
Germany
KPN Mobile N.V.
Netherlands
Far EasTone Telecommunications
Taiwan
BASE N.V./S.A.
Belgium
Bouygues Telecom S.A.
France
Telefónica Moviles España
Spain
Wind Telecomunicazioni SpA
Italy
COSMOTE Mobile Telecommunications S.A.
Greece
Telestra Corporation Limited
Australia
Cellcom
Israel
mmO2
U.K., Ireland, Germany
Mobile TeleSystems OJSC
Russia, CIS countries
Source:
14
“NTT DoCoMo and MTS Form Strategic Partnership for i-mode in Russia and other CIS countries,” DoCoMo press release, December 17, 2004.
NTT DoCoMo, Inc.: Mobile FeliCa
Exhibit 4
805-124
Japanese Mobile Subscriber Growth
160%
140%
Growth Rate
120%
100%
80%
60%
40%
20%
0%
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
YEAR
Source:
“Information on Subscribers of Cellular Telephone, Pager and PHS in Japan,” Information and Communications
Policy Bureau, Ministry of Internal Affairs and Communications [Japan].
Note:
Growth rate is year-on-year growth rate at end of each year.
15
805-124
NTT DoCoMo, Inc.: Mobile FeliCa
Exhibit 5
Subscriber Growth by Company
Overall Subscriber Growth
60,000,000
50,000,000
DoCoMo
KDDI
Vodafone
Subscribers
40,000,000
30,000,000
20,000,000
10,000,000
Year
16
Dec-04
Sep-04
Jun-04
Mar-04
Dec-03
Sep-03
Jun-03
Mar-03
Dec-02
Sep-02
Jun-02
Mar-02
Dec-01
Sep-01
Jun-01
Mar-01
Dec-00
Sep-00
Jun-00
Mar-00
Dec-99
Jun-99
Sep-99
Mar-99
0
NTT DoCoMo, Inc.: Mobile FeliCa
805-124
Exhibit 5 (continued)
2G Subscriber Growth
50,000,000
DoCoMo(2G)
45,000,000
KDDI(2G)
Vodafone(2G)
40,000,000
Subscribers
35,000,000
30,000,000
25,000,000
20,000,000
15,000,000
10,000,000
5,000,000
Dec-04
Dec-04
Mar-04
Mar-04
Jun-04
Dec-03
Dec-03
Sep-04
Sep-03
Sep-03
Jun-04
Jun-03
Jun-03
Sep-04
Mar-03
Mar-03
Dec-02
Sep-02
Jun-02
Mar-02
Dec-01
Jun-01
Sep-01
Mar-01
Dec-00
Sep-00
Jun-00
Mar-00
Dec-99
Sep-99
Jun-99
Mar-99
0
Year
2.5/3G Subscriber Growth
20,000,000.00
18,000,000.00
DoCoMO(3G)
KDDI(2.5/3G)
Vodafone(3G)
16,000,000.00
Subscribers
14,000,000.00
12,000,000.00
10,000,000.00
8,000,000.00
6,000,000.00
4,000,000.00
2,000,000.00
Dec-02
Sep-02
Jun-02
Mar-02
Dec-01
Sep-01
Jun-01
Mar-01
Dec-00
Sep-00
Jun-00
Mar-00
Dec-99
Sep-99
Jun-99
Mar-99
0.00
Year
Source:
Adapted from Telecommunications Carriers Association.
17
805-124
NTT DoCoMo, Inc.: Mobile FeliCa
Exhibit 6
QR Code of NTT DoCoMo Web Site
Source: http://www.nttdocomo.com/
Note:
Code sizes varied depending on the amount of information they contained. The most frequently used size was 2–4 cm, measured on the edge.
Exhibit 7
Source:
18
Picture of a FeliCa Chip
NTT DoCoMo company information.
NTT DoCoMo, Inc.: Mobile FeliCa
Exhibit 8
805-124
Competing Contactless IC Technologies
Contactless smart cards were cards embedded with IC chips that could communicate with the card reader through wireless self-powered induction technology. Since 2001, ISO/IEC 14443 (card standard, Type A and B) had become the standard contactless protocol for smart cards, which typically had communication distances up to 10 cm. FeliCa, also known as Type C, had not yet obtained card standard status from ISO. Type A, Type B, and so-called Type C technologies all used
13.56 MHz frequency but were categorized according to their various encryption and modulation methods. In addition to ISO 14443, there was also the ISO 15693 standard with longer communication distances of up to 70 cm; this was most commonly used for logistics.
Type A was used worldwide; the most well-known technology was MIFARE, developed by
Philips in 1994 and used in such diverse applications as electronic money, access control, road tolling, airline ticketing, phone cards, and transportation. Public transportation systems in London, the
Netherlands, Beijing, Seoul, and Moscow had adopted MIFARE. Its data transmission speed was 106 kbps (FeliCa’s transmission speed was 212 kbps). By the end of 2003, MIFARE’s user base included approximately 400 million smart-card ICs and 2 million readers.
Type B was developed by Motorola. Unlike Type A cards, the Type B cards were equipped with central processing units (CPUs), which allowed greater memory capacity. NTT Communications, an
NTT group company engaged in long-distance and international communications businesses, was a
Type B technology leader in Japan. NTT used a combination of public and private key-encryption methods. The public key method required longer processing time, as its algorithm was more complicated; however, it provided greater security than Type A or Type C technologies. Type B technology was applied for residency registration cards in Japan, and the government was planning to apply the technology to other public IDs such as driver’s licenses, health insurance ID cards, and passports. Source:
MIFARE, “About MIFARE,” MIFARE Web site, http://www.mifare.net/about/, accessed March 11, 2005; “Mobile plus contactless ICs?” IT Media: Mobile News, September 18, 2002, http://www.itmedia.co.jp/mobile/0209/18/ n_ic.html, accessed March 11, 2005; “NTT’s challenge against FeliCa—Latest trends in contactless IC cards,” IT
Media: Mobile News, March 5, 2003, http://www.itmedia.co.jp/mobile/0303/05/iccard.html, accessed March 11,
2005.
19
805-124
NTT DoCoMo, Inc.: Mobile FeliCa
Endnotes
1
“DoCoMo Signals a Cultural Shift,” The Financial Times, September 10, 2004, p. 8.
2
“DoCoMo to Review its Overseas Strategy,” Nikkei Sangyo Newspaper, June 22, 2004, p. 5.
3
“Survey of the Current Status and Issues of Networks,” Ministry of Internal Affairs and Communications
[Japan].
4
2004 DoCoMo Annual Report, p. 56.
5
Casewriter interview, Nov. 9, 2004.
6
Data source: Telecommunications Carriers Association.
7
Ibid.
8
DoCoMo, “Investor Relations—Annual/Quarterly Results FY2004 2Q Data,” DoCoMo Company Web site, http://www.nttdocomo.co.jp/english/corporate/investor-relations/business/04qrt2_e.html, accessed January
11, 2005.
9
Makio Inui, “Japanese Wireless: Uncertainty Reigns,” UBS Investment Research Report, June 29, 2004, pp. 69–
70.
10
Type A and Type B cards had obtained ISO 14443 (card standard). FeliCa technology, also known as Type
C, had obtained ISO 18092 (communication standard) in December 2003 but had not yet obtained the card standard. 11
GlobalPlatform was a cross-country forum—consisting of over 50 industry members including card issuers, manufacturers, and vendors—formed to develop, manage, and promote specifications for multiple application smart cards and enabling devices. Source: Gerhard Fasol, “i-mode FeliCa Wallet Phones,”
Eurotechnology Japan Report, December 2, 2004, p. 22.
12
“Emerging e-Money Business—Edy,” Nikkei Sangyo Newspaper, June 29, 2004, p. 3.
13 “Consumer Finance—Playing Field for Competition in the Retail Market,” Nikkei Sangyo Newspaper,
November 15, 2002, p. 15.
14
bitWallet, Inc., press release, December 8, 2004.
15 JR
East, “What Is Suica?” JR East Company Web site, http://www.jreast.co.jp/suica/about/merit/ index.html, accessed March 4, 2005.
16
17
JR East, “Development and Introduction of a Contact-less IC Card System,” November 2004.
“Credit Card Going on Mobiles—Is QUICPay Really Safe?” IT Media Mobile News, July 21, 2004, http://www.itmedia.co.jp/mobile/articles/0407/21/news101.html, accessed March 4, 2005.
20 