Company history

Deguchi Haruaki spent thirty-four years inside the industry he would later attack. Hired by Nippon Life in 1972, he handled the negotiations with the Ministry of Finance through the financial-system reforms of the late 1980s and chaired the Life Insurance Association’s financial-planning committee. When Japan’s financial system wobbled around 1997, the international expansion he had been preparing was reversed, the overseas offices were closed one by one, and Deguchi was moved at fifty-five to a building-management subsidiary. From there he wrote down what he thought was wrong with Japanese life insurance: premiums were high, the fee component was never disclosed, and policies were still sold on obligation, sentiment and gifts. Japan was also the one market where the seller was the insurer’s agent rather than the customer’s — and riders had been stacked on top of riders precisely so that products could not be compared, until even employees of the insurers missed their own unpaid claims.

In March 2006, at a Tokyo hotel, the fund manager Taniya Mamoru asked Deguchi whether he would like to build a life insurer, and Deguchi agreed on the spot. Taniya had watched the unpaid-claims scandals break across the majors and read them as an opening; what he wanted was someone who would publish the cost of insurance even knowing the industry would turn on him. Deguchi, then fifty-eight, left Nippon Life the same year, and set one condition for taking the job: a young person who understood the internet. That was Iwase Daisuke, thirty, a Baker Scholar out of Harvard Business School that July. Iwase arrived at their first meeting with a hundred-page plan for a non-life and reinsurance venture, listened to Deguchi sketch a life insurer on a whiteboard for twenty minutes, and threw his own plan away.

In October 2006 they set up a preparatory company, Netlife Kikaku, in Akasaka on ¥100 million of seed capital and a hard two-year deadline: no licence by 2008 and it dissolved. Under the postwar convoy system no independent applicant had been licensed since the 1930s, and in 2006 the idea was widely treated as impossible. Deguchi read five years of Financial Services Agency circulars and the minutes of the Financial System Council’s insurance section, concluded that the regulator actually wanted more competition, and decided the licence would come. He also chose the harder route on capital: taking an incumbent life insurer as a shareholder would have eased the application, but he refused, to keep the company free of another insurer’s colouring. Of eight companies approached one declined and one was voted down by its own board; the rest put in ¥8 billion, and two further ¥1 billion subscriptions at the end of 2007 carried it past the ¥10 billion he had promised the FSA. By March 2008, capital and reserves reached $127.7M (¥13bn) and the company was renamed Lifenet Insurance.

The licence came on 10 April 2008 and the business opened on 18 May — the first independent entrant in seventy-four years and Japan’s forty-fourth life insurer. There were two products, term life and whole-life medical, and not a single rider. There were no salespeople: the whole contract was completed online, which no insurer, domestic or foreign, had done before. For customers in their twenties and thirties the premium could be half what a major charged. Then Lehman failed four months later and almost nothing sold. By the end of January 2009 the company held 3,435 policies; its brand awareness, at 6.7%, was the lowest of any life insurer in Japan, and the two founders handed out postcards introducing the company to canteen staff and to strangers who gave them tissues on the street. In the year to March 2009 premium income was ¥80 million against an ordinary loss of ¥1.37 billion, with forty-eight employees.

The move that changed that was made in November 2008, when Lifenet published its loading rates in full — showing, product by product, how much of a premium goes to claims and how much to the company’s own salaries and costs. No incumbent disclosed this. Under a model of tied agents and nationwide branch networks, some 60% of what a policyholder paid in fees was said to disappear into distribution, and that cost structure pushed the industry toward ever more expensive, rider-laden products; because tied agents carry no comparison, none of it was visible from outside. The industry called Lifenet meddlers, traitors, apostates. Deguchi’s answer was that consumers in the internet age do not trust a company that hides things, so honesty was simply the strongest competitive strategy available. The disclosure made news, and the news filled the hole where the company’s reputation should have been.

The customers who came were the ones the industry had stopped reaching: households in their twenties and thirties on ¥3–5 million a year, at a moment when workplace solicitation was drying up under privacy law and corporate security rules. Asked how they had bought, 46% of policyholders had never held life insurance at all, 18% were adding cover and only 36% were switching from another insurer; about 80% were under forty. Iwase called it a super blue ocean; sceptics noted that the factors Lifenet eliminated were obvious while the factors it created were not, and that first-time buyers might simply be people who had reached the age at which one buys insurance. The numbers still climbed — 40,000 policies by October 2010, 70,000 by May 2011, a first-in-Japan disability product in 2010, a listing on the TSE Mothers market in March 2012, and 60,725 new policies in the year to March 2012. But the ordinary loss ran unbroken through that sixth year.

New policies peaked at just over 60,000 in the years to March 2012 and 2013, then fell by more than half within two years of listing. A medium-term plan published in May 2013 targeted ¥15 billion of ordinary revenue for FY2015; by November 2014 that had been cut to ¥9.5 billion. In May 2014 the company revised its term and medical products and cut premiums for the first time since opening — and the decline did not stop, so neither price nor range was the binding constraint. Deguchi’s explanation was that the pond had not grown while the rods multiplied: online-only rivals went from two in 2008 to eight by 2015, and the majors added web products, walk-in shops and bank counters. But the deeper problem was the company’s own design. A narrow, comprehensible line-up was Lifenet’s lifeline, and rivals pointed out that it was equally a limit on product development, in a market where the right cover depends on income, family and the age of one’s children. Mori Ryosuke later named the internal errors plainly: Lifenet was late moving from the PC to the smartphone, and it strained for profit at the expense of acquiring customers.

So the company began undoing the things that had defined it. In December 2014 it signed an agency agreement with Hoken no Madoguchi, the largest walk-in insurance-shop chain — face-to-face selling, the very item it had stripped out at launch, rented back through someone else’s stores. Then in April 2015 the board approved a capital and business alliance with KDDI: 8 million new shares for $25.1M (¥3bn) in May, giving the telecoms group 15.95% of the votes, the largest holding, a seat on the board and status as a related party. Deguchi described the purpose not as money but as supplementing credibility — a life policy runs for decades, and for a young company the question “will you still be here?” is itself the objection. From FY2016 the plan rested on three pillars, online direct, KDDI as a tied agent, and face-to-face agencies, aiming at ¥13.5 billion of revenue and an operating profit by FY2018. The au channel underdelivered: by 2018 the FY2018 forecast had been cut to ¥12 billion. Deguchi’s own conclusion was that an insurer’s purpose is to deliver good cover cheaply and comprehensibly, and that there had never been any need to define itself as internet-only.

The founders handed the company on quickly. Iwase became president in 2013 and representative director in 2016; Deguchi left the chair in 2017, saying that at sixty-nine he should step aside for two directors in their thirties whose ages together roughly equalled his own, and noting that the majors probably had no directors that young at all. Mori Ryosuke — who had joined from Goldman Sachs in 2012, from outside the insurance industry entirely — became president in 2018. His signature product, the cancer policy Double Ale, came out of claims data rather than market surveys: six in ten disability claims began with cancer, and people who kept working through treatment lost about 20% of their income, so the cover was designed to replace earnings as well as pay for treatment. It sold above the prior year every month after its 2017 launch. The losses continued regardless — ¥5.3 billion of acquisition costs deferred over the first five years began amortising from FY2013 — and Lifenet posted net losses for eight straight years after listing. Mori called it a good deficit, the natural shape of a business that pays to acquire customers now and collects for decades. At that burn rate, though, equity would have lasted about five more years.

The change that finally moved the numbers was not a product but a way of gathering customers. In August 2022 Lifenet agreed to underwrite group credit life insurance for au Jibun Bank, taking over from Crédit Agricole Life in July 2023. Group credit cover is attached to a mortgage and settles the outstanding loan if the borrower dies or is severely disabled; the borrower does not choose the insurer, so the work of assembling the insured is done by the bank’s mortgage salesforce, not by advertising bought one policy at a time. Lifenet priced the first year too cheaply and lost ¥600 million on claims, but the contract allows annual rate resets, and after the July 2024 revision the book turned: annualised premium on group credit reached ¥7.64 billion by March 2025, up 229.7%, and ¥8.57 billion a year later. Joint cover for two-borrower loans followed in 2025, and a tie-up with Kyoto Shinkin Bank was signed in November 2025 for delivery from July 2026. Alongside the banks came a widening bench of partners — Seven Financial Services, Money Forward, Eisai, Sumitomo Mitsui Financial Group and Sumitomo Mitsui Card, Advance Create — and loyalty-point tie-ins with Ponta and V Point, until selling through a partner’s app was a channel in its own right.

From the year to March 2024 the group also adopted IFRS voluntarily, restating the prior year for comparison, and the two sets of books tell opposite stories. On IFRS, consolidated profit was ¥3.56 billion in FY2023 and rose through ¥5.73 billion and ¥5.99 billion to ¥8.04 billion in FY2026, with insurance revenue climbing from ¥20.7 billion to ¥34.4 billion. On Japanese GAAP the parent company was still losing money — ¥5.10 billion, ¥4.72 billion, ¥3.05 billion — over exactly the same years. The gap is entirely in how insurance liabilities are measured: under IFRS the unearned profit attached to future service sits on the balance sheet as a contractual service margin and is released as the service is delivered, which smooths a business whose costs land years before its revenue. Lifenet went on to make that logic its own yardstick, setting a medium-term target in comprehensive equity — consolidated equity plus tax-adjusted CSM and the value of the group credit book.

Then, in the year to March 2026, the Japanese-GAAP accounts turned as well: ordinary profit of ¥2.86 billion and net profit of $21.5M (¥3bn), the first profit in the eighteen years since the business opened in May 2008 — and thirteen years later than the 2009 forecast that Lifenet would break even within five. Premium income had doubled in four years, from ¥25.4 billion to ¥51.2 billion, individual policies in force passed 600,000, and headcount had gone from 73 to 242. Ownership changed as well: repeated share issues diluted the KDDI side from a 25.02% peak back to 18.33%, and in April 2026 au Financial Holdings agreed to sell its entire stake to Japan Airlines, ending the alliance struck in 2015 and beginning another. Yokozawa Junpei, an engineer who had joined in 2008 and run the KDDI account, became president in June 2025 — the first leader from neither the founding pair nor finance — and moved the listing to the TSE Prime market that July. The open question is the one the founders started from: individual new policies had been stuck in the seventy-thousands for two years, and the company that set out to sell insurance directly and cheaply now grows mostly on other companies’ customers.