Company history

DoCoMo did not begin as a venture. In August 1991 NTT — privatised six years earlier and still under government pressure to open the market to competition — set up a Tokyo vehicle to receive its mobile business, and three months later added eight regional planning companies covering Hokkaido, Tohoku, Tokai, Hokuriku, Kansai, Chugoku, Shikoku and Kyushu. Renamed NTT Mobile Communications Network in April 1992, it took over NTT’s cellular, car-phone, paging, ship and in-flight telephone operations that July. The nine-company structure that emerged by October 1993 was a regulatory artefact, but it also meant that from day one one organisation was responsible for the whole of Japan.

The timing was extraordinary. Demand for mobile phones exploded almost immediately after the separation, and DoCoMo was the party best placed to absorb it. When handset sales were deregulated in 1994 and tariffs were cut repeatedly, subscribers began adding at roughly ten million a year. President Oboshi Koji put the win down to three things: a network built out from Hokkaido to Kyushu before the traffic arrived, an in-house engineering capability that drove handsets smaller and lighter, and a policy of returning profit to customers as price cuts. R&D spending was protected even in loss-making years, and prices were cut ahead of demand rather than in response to it — a discipline his successor, Tachikawa Keiji, credited as the foundation of everything that followed.

By January 1997 DoCoMo was taking 89% of new contracts in the Tokyo area, and its position was effectively settled. A 1996 Fair Trade Commission inspection into price-fixing at dealers narrowed the price gap with rivals, which perversely helped — dealers reported that at the same price more customers simply chose the DoCoMo brand, leaving the company with the odd problem of having to watch its own discounting for antitrust reasons. IDO’s president complained that DoCoMo had inherited the state monopoly’s people, money and technology and could not be met on level ground. In October 1998 the company listed on the First Section of the Tokyo Stock Exchange at an offer price of ¥3.9m a share, raising $16.2B (¥2.13tn) at home and abroad and handing NTT roughly ¥850bn in gains on the shares it sold. The FTC pressed NTT to cut its 95% holding; NTT refused to go below half.

The insight behind i-mode, launched in February 1999, was that a carrier which only rents out lines has nothing to differentiate and ends up competing on price alone. Put a gateway at the junction between the cellular network and the internet, convert protocols there, and the operator owns not just the pipe but the content and the billing that flow through it. It worked: DoCoMo built a recurring revenue layer on top of the network, though the gateway drew enough load to cause repeated outages and the metered, per-packet billing meant the more customers used it the more they paid. The unit that built it was staffed by hires poached from Recruit and internet firms — colleagues called them the foreign legion, and handset makers remarked on how different they felt from DoCoMo’s public-corporation sales side. Matsunaga Mari, who ran planning in the gateway business division, was the one who insisted the company should decide for itself what went on the platform.

Owning both network and service made DoCoMo the centre of gravity of the entire NTT group, and NTT knew it. President Miyazu Junichiro had warned before the 1999 restructuring that the group could yet break DoCoMo up. When NTT reorganised that July into a pure holding company over two regional carriers and NTT Communications, it kept 67% of DoCoMo, placing the subsidiary alongside its former parents. The economics were already inverting: 43.6% of calls from mobile phones went to other mobile phones by fiscal 1997, bypassing NTT’s network and its interconnection revenue, and in the year to March 1999 NTT’s fixed-line and ISDN subscribers fell for the first time, to 62.91m, while mobile reached 41.53m.

With domestic growth visibly finite, DoCoMo went abroad. From 2000 it set out to export the W-CDMA 3G standard it had led and i-mode with it, taking stakes in carriers across Asia, Europe and the United States — Hutchison Telephone in Hong Kong, 15% of KPN Mobile, and in November 2000 about $10.0B (¥1.08tn) for 16% of AT&T Wireless, then America’s third-largest operator. The policy was deliberate: full control where DoCoMo could take full responsibility, otherwise 15–20% — enough to consolidate, not enough to run. Tachikawa defended it as a long horizon — “we are not short-term speculators” — but the telecom bubble burst almost immediately. In the year to March 2002 the operating business was superb, with revenue past ¥5tn and operating profit over ¥1tn, and yet the same accounts carried an ¥812.9bn writedown on affiliate shares (¥505.6bn on AT&T Wireless, ¥262.7bn on KPN Mobile) plus ¥125.9bn of equity-method losses — roughly the ¥930bn raised from the market the year before, gone in little over a year. At home the FOMA 3G service launched in October 2001 into an incomplete coverage area and languished for two years, and by March 2004 au was beating DoCoMo on monthly net additions for the sixth straight month.

The problem the next two presidents inherited was not decline but the absence of anywhere left to grow. Nakamura Masao took over in June 2004 with revenue double and net profit five times the level of 1998, and a market that was adding barely 0.5% a month. By April 2006 DoCoMo had 50.65m subscriptions and about 56% of the market — but au, the strong challenger, held only half that, which meant the upside was not in Japan. When mobile number portability began in October 2006 the company started losing the monthly net-add race outright. Yamada Ryuji, who succeeded him in June 2008 after a career at NTT West and the holding company, made a virtue of being new to DoCoMo, and named the real ailment: effort that no longer produced results, a head office grown large enough to optimise its own parts, and a hesitancy about what to change.

Structurally, though, this was the period in which DoCoMo finally became one company. In July 2008 it absorbed the eight regional DoCoMo companies, ending seventeen years of divided national coverage, after a “new DoCoMo declaration” in April that put existing customers first and changed the logo. The PHS business taken over in 1998 was written down by about ¥60.4bn and closed to new subscribers in 2005. The overseas portfolio was unwound in parallel: when Cingular bought AT&T Wireless in October 2004, DoCoMo sold its stake at $15 a share for about $6.495bn, booking a ¥501.8bn gain in fiscal 2004 — the 3G bubble investments cleared out in roughly four years. DoCoMo was also first in Japan to LTE in December 2010, though once the iPhone 5 supported LTE in 2012 and KDDI and SoftBank poured money into base stations, coverage and speed stopped being a visible DoCoMo advantage.

That left the iPhone. For five years DoCoMo alone refused to carry it, unwilling to accept Apple’s sales quotas or the restrictions Apple placed on the carrier’s own services — the same self-sufficiency that had produced i-mode. The price was 55 consecutive months of net subscriber losses to portability and a share that fell from over 50% to the low 40s. A May 2013 “two-top” push, heavily discounting Sony’s Xperia A and Samsung’s Galaxy S4 and abandoning the long-standing policy of treating every handset equally, failed: portability losses ran 135,000 in May and 146,000 in June, and in June alone DoCoMo shed 140,000 subscribers while SoftBank gained 60,000 and KDDI 80,000. In September 2013 Apple announced DoCoMo as a carrier, and sales of the 5s and 5c began on 20 September. Trading the proprietary platform for retention was the moment the i-mode doctrine ended.

NOTTV, the mobile broadcasting service launched in April 2012 on VHF spectrum freed by the digital television switchover, showed how hard the i-mode trick was to repeat. Six years of preparation produced a million subscribers by mid-2013 — and then DoCoMo started selling the iPhone, a global handset with no receiver for it. The operating subsidiary ran up close to ¥90bn of accumulated losses in three years, and management pulled out in November 2015, taking a ¥30.2bn impairment on the broadcasting assets in fiscal 2014 on top of provisions taken earlier. Natsuno Takeshi, who had built i-mode, was struck that the management would so publicly negate its predecessors’ project, and put the failure down not to the size of the customer base but to the absence of a virtuous circle: i-mode had not succeeded by standing still, he noted, but by adding services constantly.

Earnings recovered — the year to March 2018 brought record-level revenue of ¥4.7694tn, operating profit of ¥1.0966tn and net profit of ¥790.8bn — and DoCoMo promptly gave it away. In late 2018, facing an emergency government proposal on mobile pricing and Rakuten’s imminent entry, it announced up to ¥400bn of customer returns, arguing that a 20–40% cut roughly matched where a new entrant would price and closing the gap its own surveys found between what customers paid and what they thought the service was worth. The strategy in exchange was membership: grow the d Point Club from about 65m members to 78m by fiscal 2021 and eventually to 100m — in effect, everyone in Japan. Fiscal 2019 absorbed ¥200bn from the new tariffs, ¥80bn from separating handset and line charges, and ¥55bn of extra investment including 5G, cutting operating profit by ¥180bn.

That was the trap. DoCoMo led on share but had no cheap sub-brand — no UQ mobile, no Y!mobile — with which to fight on price, and as a listed company it could not simply choose earnings-destroying cuts over its minority shareholders’ interests. On 29 September 2020 NTT, holder of about 66%, announced a tender offer to take DoCoMo private for roughly $40.3B (¥4.3tn), the largest tender offer ever made in Japan. President Sawada Jun was direct about the reason — DoCoMo had fallen to third in revenue and profit — and about the mechanism: consolidating the 30%-plus minority interest let the group absorb the profit hit from price cuts. The company separated from NTT in 1992 to promote fair competition was being returned to its parent in order to comply with a government demand for lower prices.

DoCoMo left the market on 25 December 2020, after twenty-two years as a listed company and three straight quarters of subscriber decline. Ii Motoyuki had taken over as president on 1 December and within two days announced ahamo, 20GB for ¥2,980 a month, followed on 18 December by cuts to the existing large-data plans; SoftBank and KDDI matched within weeks. His own verdict on the intervening decades was blunt: DoCoMo’s staff, he said, ought to feel the shame of not having produced a hit since i-mode in 1999.