Hikari Tsushin: The Phoenix of Japanese Capitalism & The Ultimate Cash-Flow Machine

I. Introduction & Episode Roadmap (0:00 – 0:12)

On February 15, 2000, a company that sold mobile phone contracts out of storefronts reached a valuation of ¥7.46 trillion.

That was the day 株式会社光通信 Hikari Tsushin, Inc. touched ¥241,000 a share on the 東京証券取引所 Tokyo Stock Exchange — a market value larger than most of the industrial giants that had defined post-war Japan.¹ Its founder, 重田康光 Yasumitsu Shigeta, was 34 years old, ten days short of his 35th birthday. Forbes estimated his paper fortune at $42 billion that month, up from $1 billion a year earlier and $25 billion in December 1999 — one of the fastest accumulations of nominal wealth in recorded financial history.² The stock was trading around $2,300 a share on roughly 720 times earnings.²

Six weeks later, the stock began a steep decline. Starting March 31, 2000, Hikari Tsushin shares closed limit-down — the maximum daily decline permitted by the exchange — for twenty consecutive trading sessions.¹ For twenty straight trading days, sellers queued while no transactions could clear. By July 2002, the stock changed hands at ¥895, representing a 99.6% drop from its peak.¹ According to the company's reported total-shareholder-return figures, the fiscal year to August 1999 delivered a 1,241.6% return, followed by declines of −90.4% and −74.4% in the subsequent two years.³

Most companies do not recover from a collapse of that magnitude. Hikari Tsushin did. The primary analytical question is not whether it recovered — which is recorded on the tape — but what it became, and whether its operational model is as durable as reported figures indicate.

The second act. In the fiscal year ended March 31, 2026, Hikari Tsushin reported revenue of ¥734.8 billion, operating profit of ¥116.7 billion, and net profit attributable to owners of ¥151.0 billion — all of them company records.⁴ Shareholders' equity crossed ¥1 trillion in September 2025 and closed the fiscal year at ¥1.19 trillion.⁵⁴ The group also held a portfolio of listed equities carried at a market value of ¥1.48 trillion against an acquisition cost of ¥852.3 billion — a holding larger than the operating businesses that funded it.³ In early August 2026, the shares traded near ¥39,000, giving the company a market capitalization of roughly ¥1.73 trillion.⁶

Since its 1996 market debut, Hikari Tsushin has generated a compound annual total shareholder return of 9.7%, compared with 4.6% for the TOPIX including dividends — representing a cumulative return of 1,597% versus 380%.³ That track record includes the 99% drawdown, reflecting a long-term return profile that few original shareholders actually experienced.

What this story is really about. Strip away the market drama, and Hikari Tsushin operates as a distribution engine: it sells third-party electricity, water, insurance, and broadband services to Japanese small businesses and households, converts the resulting monthly cash flows into a proprietary metric called ストック利益 ("stock profit"), and reallocates the surplus into a leveraged portfolio of Japanese small-cap equities. Management does not treat operations and investments as separate activities. Asked by an overseas investor in May 2026 what outsiders most misunderstand, President 和田英明 Hideaki Wada stated that no wall exists between the operating businesses and the equity book, describing the strategy as simply allocating capital where it earns the highest return.⁷

That framing is elegant, but it remains a premise to test.

The company's corporate structure is complex. Hikari Tsushin is a pure holding company overseeing 170 consolidated subsidiaries and 108 equity-method affiliates across seven operating segments, with forty-two of those affiliates being separately listed companies in which it holds voting stakes between 20% and 43%.⁵³ It lacks an obvious peer group in global markets. It is not a traditional utility, though it sells more electricity than most new market entrants in Japan. It is not an insurer, though it owns a licensed insurance business. Nor is it a conventional asset manager, even though its equity portfolio generates higher earnings than many Japanese asset managers. Index providers classify the company under capital goods, while the market continues to debate its valuation framework — a reality management has publicly acknowledged.⁷

The analysis unfolds in six moves:

• The rocketship (1988–1999): how a university dropout turned door-to-door telephone sales into Japan's largest mobile-phone distribution agency, and why the economics were structurally fragile even while they looked spectacular.
• The cataclysm (2000): fabricated contracts, a guidance reversal, twenty limit-down days, and a public divorce from ソフトバンク SoftBank.
• The transformation (2001–2019): the abandonment of one-off commissions for recurring "stock profit," and the far less discussed decision to stop employing the salesforce altogether.
• The architecture today: seven reporting segments, one of which quietly does most of the work, and a subsidiary that is more than 10% of group revenue with a regulatory file attached.
• The capital allocation engine: what the equity portfolio actually is, where its profits land in the accounts, and why that placement matters more than the returns.
• The stress test: 7 Powers and Porter applied honestly, the governance file, the risk radar, and the two or three numbers that will tell shareholders whether the machine is still working.

It begins in a rented office in Tokyo, with a young man, a telephone, and a phone book.

---

↑ Back to Top

II. Founding & The 1990s Mobile Telecom Boom (0:12 – 0:35)

The founding asset was a directory.

In February 1988, Yasumitsu Shigeta — 23 years old, a Nihon University student who left without completing his degree — incorporated Hikari Tsushin with ¥1 million in capital to sell and lease office automation equipment and telephone handsets.⁵ The addressable market was not derived from a formal study; it was NTT's Yellow Pages, which listed millions of Japanese small and medium enterprises, each equipped with a telephone and fax machine but lacking a procurement department.⁸ The strategy was straightforward: knock on every door.

Within five months, the company added a far more lucrative business line: reselling long-distance telephone service.⁵ This shift fundamentally altered the nature of its revenue. Selling a fax machine is a single transaction. Signing a business onto a discounted long-distance carrier creates a recurring stream — and under early Japanese telecom deregulation, the agent that enrolled the customer earned a share of that monthly usage stream. Hikari Tsushin discovered that its primary asset was not hardware, but a direct channel to small business owners seeking to simplify their telecommunications.

Then came the mobile phone.

Riding handset deregulation. Through the late 1980s, Japanese cellular service relied on a rental model where carriers owned the handsets and subscribers leased them. That dynamic changed in April 1994, when regulators permitted outright handset sales. Hikari Tsushin moved immediately, integrating handset sales into the line resale business it had begun expanding in June 1993.⁵ The company opened its first dedicated mobile phone store in Shinjuku, Tokyo, in May 1994.⁵ Business phone systems followed in May 1995, followed by PHS handsets and agency contracts that July.⁵

What fueled the explosive growth of the 1990s was carrier economics rather than proprietary technology. 日本電信電話 NTT's mobile unit and its competitors entered a subscriber land grab where the estimated lifetime value of a customer far exceeded the cost of acquisition. Carriers showered agents with large upfront activation incentives and heavily subsidized handsets. An agency with a direct sales force could immediately convert its sales labor into upfront cash, funded by carriers wagering on long-term subscription revenue.

Hikari Tsushin scaled that structural arbitrage. Its franchise storefront network, branded HIT SHOP, expanded from a dozen locations in 1994 to more than 1,500 nationwide by May 1999, according to corporate filings.⁵⁸ Operating alongside the stores was an outbound corporate sales force driven by strict performance metrics: aggressive quotas, public leaderboards, uncapped commissions, and employee turnover that would alarm a traditional Japanese corporation. The sales culture was the product. While any distributor could sell a Motorola handset, few could systematically persuade hundreds of thousands of small business owners each year to sign new telecom contracts.

The financial mechanics of this model warrant close examination, as similar principles govern the company today under a different structure. An activation effectively functioned as a bond purchase by the carrier: the carrier paid the agent upfront cash in exchange for an expected multi-year stream of subscription revenue. The agent acted as originator. The underlying credit risk was subscriber churn or default — whether the customer failed to generate traffic, canceled early, or never existed. In the 1990s, agents recognized that upfront cash as revenue on day one while carrying minimal default risk on their balance sheets, since contract clawback provisions were rarely enforced at scale during the expansion phase.

That structure created a powerful growth model while activation volumes rose, but it harbored severe solvency risks if activation growth faltered or contract quality was tested.

The capital markets take notice. The company registered its shares with the 日本証券業協会 Japan Securities Dealers Association in February 1996, making the 31-year-old Shigeta the youngest chief executive of a publicly traded Japanese company at the time.⁵⁸ Hikari Tsushin split its trading unit from 1,000 shares to 100 in January 1997, added a rental server business in September 1998, and listed on the Tokyo Stock Exchange First Section in September 1999.⁵

The market timing was potent. Japan's dot-com boom centered on the thesis that mobile phones would become the nation's primary digital gateway. Shigeta was a leading advocate of this vision, predicting in 1999 that the term "phone" would soon become obsolete as mobile devices added broader functionality.² While his technological foresight proved accurate a decade later, the near-term valuation disconnect was extreme. Shigeta also joined SoftBank's board as an outside director, linking the two most high-profile technology narratives in Japan.⁹

What the underlying metrics revealed. Hikari Tsushin's 1999 financial results projected extraordinary growth, but the underlying business model rested on three vulnerable dependencies:

• Upfront revenue recognition: Revenue was booked at the point of sale rather than over the subscriber's contract life, tying reported earnings entirely to gross new activations rather than retention.
• Carrier concentration: Cash flows depended on a small group of telecom carriers that maintained total control over incentive schedules and held legal rights to claw back payments on non-performing accounts.
• Volume-driven compensation: The commission structure heavily rewarded sales volume over contract durability, incentivizing rapid acquisition regardless of long-term account quality.

Each dependency carried risk, but together they formed a dangerous feedback loop: fund high sales volume, book the upfront commission as immediate profit, leverage the rising earnings to expand market valuation, and rely on higher volume to sustain the cycle. The model lacked an internal stabilizer, and a market valuation exceeding hundreds of times earnings removed external discipline.

The stabilizing force ultimately arrived from the outside — delivered by an investigative magazine.

---

↑ Back to Top

III. The 2000 Cataclysm: The 99% Crash & The Near-Death Experience (0:35 – 1:02)

The exposé landed in March 2000, centered on a single term: 寝かせ nekase — literally "putting to sleep."⁸

Reporting in the Japanese press alleged that contracts were being fabricated and activated across the HIT SHOP network to harvest carrier incentive payments — registering mobile lines to non-existent or uninformed customers, keeping them dormant, and quietly cancelling them once commissions cleared.¹⁸ This was not a subtle accounting dispute; it was an allegation that a material portion of the company's reported growth had been manufactured internally by its own sales incentives.

What followed was faster and more damaging than the allegations themselves. Within weeks, management reversed its earnings guidance. Having previously projected a half-year profit of roughly $52 million, the company disclosed a loss of approximately $112 million for the period — a $164 million swing for a business whose entire valuation rested on predictable forward numbers.⁹ Telecom carriers tightened incentive terms and pursued refunds on non-performing accounts.

Trading in the shares quickly collapsed. Starting March 31, 2000, Hikari Tsushin stock closed limit-down for twenty consecutive trading sessions.¹ In a limit-down market, transactions cannot clear, leaving shareholders unable to exit at any price as the theoretical value of their holdings fell each morning. For twenty straight days, investors in one of Japan's largest listed companies held shares they could not sell.

The SoftBank divorce. The relationship that had amplified Hikari Tsushin's market profile during its ascent accelerated its decline. SoftBank's leadership urged Shigeta to step down as a non-executive director and, in November 2000, requested that Hikari Tsushin withdraw from a joint venture the two had backed.⁹ Forbes reported a growing swirl of issues, including transparency complaints, allegations regarding organized-crime ties, and a market that no longer believed the company's statements about its financial condition.⁹

The digital media venture the two companies had backed together became the public token of a private rupture.⁹ In Japanese corporate finance, public disavowal by a primary partner functions as a credit event, signaling heightened risk to lending banks and institutional counterparties.

Survival. Shigeta executed a series of immediate restructurings to stabilize the company.

He pledged roughly $86 million of his own money — about ¥10 billion at contemporary exchange rates — to subscribe for new shares.⁹⁸ The company closed storefronts rapidly: restructuring reports indicate roughly 1,050 outlets shuttered within four months, shrinking the retail network to a fraction of its peak.⁸ Group interest-bearing debt dropped from nearly ¥230.7 billion in April 2000 to roughly ¥37.2 billion by March 2003.⁸ In 2002, the company moved its fiscal year-end from August to March, aligning with standard Japanese corporate convention while resetting the comparative base.⁵ Shigeta also ceased public media appearances, establishing a quiet operational stance he has maintained ever since.

Financial losses continued after the headlines faded. Forbes estimated a net loss of about $264 million for the fiscal year ending August 2001 — a major operational decline following the initial crash, reflecting the cost of unwinding non-viable sales lines.⁹

The equity market's verdict over those two years, reflected in the company's total-return series, was −90.4% followed by −74.4%.³ A shareholder who bought at the peak and held to the bottom lost virtually all invested capital, whereas an investor who bought at the 1996 debut and held for thirty years did extremely well.³ Both statements are factual, and the gap between them highlights how timing and valuation dictate investor experience.

What actually broke, precisely. Attributing the collapse solely to general dot-com speculation misses the specific mechanical failure in the underlying business model.

The company had built a system that paid third parties per signed contract, recognized income before cash arrived, and evaluated performance on gross additions. When incentives prioritize volume and verification is weak, the channel will manufacture volume as its most efficient output.

Hikari Tsushin's modern architecture — the strict separation of recurring "stock profit" from upfront customer acquisition costs, the refusal to expand at inadequate yields, and the capitalization of contract costs as assets subject to rigorous impairment testing — represents a direct operational answer to that failure. Whether that architecture fully eliminates risk remains an open question: Section IX examines regulatory findings concerning what sales agents at the group's largest electricity subsidiary told customers.¹⁰¹¹

But first, the rebuild.

↑ Back to Top

IV. The Pivot: Building the "Stock-Profit" Subscription Engine (2001–2012) (1:02 – 1:32)

The pivot has a definition, and that definition forms the core strategy.

In corporate filings, Hikari Tsushin defines ストック利益 ("stock profit") as income earned from monthly base charges, usage fees, and insurance premiums paid by signed customers, net of servicing and relationship maintenance costs — revenue derived either from telecom carriers and insurers or directly from end users.⁴ Positioned alongside it is 獲得コスト ("acquisition cost"): the upfront commissions paid to sales agents to secure those contracts. Since its fiscal year 2026 presentation, management has stated the core arithmetic explicitly: stock profit minus acquisition cost equals operating profit.³

That single calculation summarizes the company's financial model. In operational terms, the group deploys capital to buy recurring annuities, making reported operating profit equal to existing annuity income minus the acquisition cost of new contracts added during the year. Mechanically, reported operating profit can be increased in any given period simply by curtailing contract acquisitions.

Transitioning to this architecture required nearly two decades and unfolded across three strategic phases.

Phase one: identify recurring revenue streams. Hikari Tsushin expanded into insurance agency sales in July 2002, entering a business line where agents collect a share of customer premiums for the life of the policy.⁵ The company subsequently added lease-and-maintenance agreements for office copiers and hardware, fixed-line telephone connections, broadband subscriptions, and other recurring commercial services. In April 2009, it began selling proprietary branded merchandise alongside its third-party agency offerings, shifting from pure agency distribution toward higher-margin product ownership.⁵

Phase two: acquire established annuity bases. In June 2013, 株式会社エフティグループ (FT Group) became a consolidated subsidiary.⁵ In February 2015, Hikari Tsushin acquired control of 株式会社プレミアムウォーターホールディングス (Premium Water Holdings), establishing residential and commercial bottled water delivery as a distinct reporting segment.⁵ The water business illustrates the group's subscription model: low- or zero-upfront cost dispenser installations generate multi-year recurring revenue through contracted water bottle deliveries under five-year agreements.

These acquisitions reflect a consistent transaction criteria. Rather than purchasing companies for brand equity or technology, Hikari Tsushin targeted customer accounts and delivery infrastructure to route additional products through existing distribution channels. Balance sheet metrics reflect this disciplined valuation approach: after a decade of acquisitions, consolidated goodwill stood at ¥18.4 billion against ¥2.85 trillion in total assets at the end of fiscal year 2026.⁴

Phase three: restructure sales force overhead. In April 2017, coinciding with the launch of its retail electricity business, the company began transitioning from 直販 ("direct sales") to 代販 ("agency sales").⁵ Hikari Tsushin phased out its salaried internal sales force, shifting customer acquisition entirely to independent third-party agents paid on a commission-per-contract basis.

This structural shift altered the group's operating profile. Average headcount declined from a peak of 12,367 in the fiscal year ended March 2015 to 4,835 in fiscal year 2026 — a 60% reduction over eleven years.³ Over the same period, annual operating profit per employee rose from approximately ¥2 million to ¥24 million, while annual revenue per employee increased from ¥47 million to ¥151 million.³ In May 2026 earnings commentary, President Hideaki Wada indicated that the company intends to pursue further productivity gains per employee through automated systems and artificial intelligence.⁷

Converting fixed payroll into commission payments transforms sales overhead from a fixed operating expense into a variable acquisition cost line item, which stood at ¥69.6 billion in fiscal year 2026.³ While this adjustment does not lower overall customer acquisition spending, it enables management to reduce contract originations during economic downturns without maintaining redundant internal headcount. Conversely, relying on third-party sales networks delegates end-customer interactions to external agents incentivized primarily by volume — returning to a distribution model structural to the group's 1990s operations, albeit under modified financial metrics.

Financial baseline. In the fiscal year ended March 2011, Hikari Tsushin reported revenue of ¥207.1 billion, stock profit of ¥32.1 billion, operating profit of ¥2.5 billion, and a net loss.³ Management identified that period as the earnings trough during its transition from one-off agency commissions to recurring subscription revenue.⁷ Over the subsequent fifteen fiscal years, stock profit expanded 5.7-fold (a 12.4% compound annual rate) and shareholders' equity grew 11.8-fold (a 17.9% compound annual rate), while maintaining continuous dividend distributions.⁷

The fifteen-year operating profit compound annual growth rate of 28.8% reflects comparison against the depressed fiscal year 2011 baseline; measured over a fourteen-year period from fiscal year 2012, operating profit grew at a compound annual rate of 14.1% — a baseline adjustment Wada noted during the May 2026 investor presentation.⁷

Two governance and structural adjustments accompanied this operational reorganization. In June 2017, Hikari Tsushin reorganized into a 監査等委員会設置会社 ("company with an audit and supervisory committee"), placing independent directors on a statutory monitoring body with formal voting rights.⁵ In June 2019, concurrent with Wada's appointment as president, the group's non-life insurance subsidiary secured its underwriting license, shifting the insurance business from acting solely as a commission agency to underwriting and retaining premium risk directly.⁵

One month later, the group began acquiring equity stakes in listed Japanese companies alongside its core customer acquisition operations.

↑ Back to Top

V. The Modern Business Model: Segment Breakdown & Economic Engine (1:32 – 2:07)

Start with a number that reframes everything: the parent company of Hikari Tsushin employs three people.⁵

It is a pure holding company with an average employee age of 49 and average tenure of nearly 22 years, and average annual compensation of ¥35.2 million.⁵ The operating work happens across 170 consolidated subsidiaries and 108 equity-method affiliates.⁵ Group employment stood at 4,012 at March 31, 2026, plus an average of 823 temporary staff.⁵

For the year then ended, revenue rose 7.0% and operating profit 11.1%, for an operating margin of 15.9%.⁴ But operating profit is less than 60% of the story: pre-tax profit reached ¥199.1 billion, up 32.1%, because financial income of ¥77.6 billion and equity-method income of ¥26.9 billion together dwarfed financial costs of ¥22.4 billion.⁴ Return on equity attributable to owners was 14.4%.⁴ Roughly two-fifths of pre-tax profit came from the investment book rather than from selling anything.

The seven segments break down as follows.

Electricity and Gas — the engine (¥319.6 billion revenue, ¥35.8 billion operating profit). The largest business by revenue at 43% of the group, and the largest by profit at 29% of segment profit before corporate costs.⁴ It sells electricity and gas to households and small businesses, and it does so with 261 employees.⁵ Its principal operating company, 株式会社ハルエネ Haruene, alone generated ¥153.5 billion of revenue in the year — over 10% of the consolidated total, which triggered separate disclosure.⁵

Two features define its economics. First, Hikari Tsushin owns no generation. It procures 100% of its electricity from the wholesale market, the Japan Electric Power Exchange, and Wada confirmed on the May 2026 call that it holds no fixed generation assets, uses no futures or options to hedge procurement, and has no intention of starting.⁷ Second, it sells only market-linked tariffs, which pass wholesale price movements through to the customer in the same month, versus the three-to-five-month lag typical of major utilities' fuel-cost adjustment mechanisms.³

The claim management draws from this is that gross profit is neutral to fuel prices. Mechanically, that is right, and it distinguishes Hikari from the many Japanese new-entrant retailers that were destroyed by the 2021–2022 wholesale price spike. But three caveats belong on the record. Pass-through protects margin while transferring price shock to customers, so the residual exposure is churn rather than cost — which is why the tariffs cap monthly bill spikes by spreading the excess over three months.⁷ Wholesale spikes also strain working capital; asked to size it, Wada put the cash impact at roughly 1% of group cash.⁷ And segment revenue included ¥12.8 billion of government subsidies recognised under IAS 20 — equivalent to more than a third of segment operating profit, and not a permanent feature of the landscape.⁴

The segment's fiscal 2026 result illustrates the stock-profit equation in action. Stock profit grew 12% and acquisition cost grew 28%, so operating profit grew 1%.³ That is not deterioration; it is deliberate investment. The composition is the interesting part: acquisition volumes in low-voltage (household and small-business) contracts ran at 122% of the prior year while high-voltage (larger corporate) volumes fell to 86%.³ Wada was direct about why — price competition in high-voltage had intensified, and the company declined to match discounts, preferring adequate returns over share.⁷ Over 90% of acquisition spending went to low-voltage.⁷

Where does low-voltage volume come from? Not the internet. Wada described the channel as alliances with residential property management and brokerage companies, built on relationships dating to the era when the group sold NTT fibre connections, in some cases formalised through joint ventures — and said the group partners with most of Japan's ten largest property managers.⁷ Monthly billing per contract runs around ¥7,000 for an individual, ¥25,000 for a corporate customer, and about ¥1 million for a high-voltage contract.⁷ The rental-property channel, not the doorstep, is the actual asset here — a point Section VII will return to.

Telecommunications (¥127.5 billion, ¥29.4 billion). The second-largest profit contributor, at 24% of segment profit on 213 employees.⁴⁵ Fixed and mobile line services plus ancillary content subscriptions. Acquisition cost was flat while stock profit grew 7%, delivering 11% profit growth — the signature of a mature book being harvested.³ Line acquisitions ran at 93% of the prior year and content at 109%.³ Wada characterised the line business as having bottomed after a difficult stretch and entered gentle growth, without pretending it would double.⁷

Finance (¥45.5 billion, ¥22.1 billion). The most profitable segment by margin, at 49%, and the one carrying the least examined risk.⁴ Launched overseas in earnest in May 2023, it makes microfinance loans in Cambodia and Malaysia against motorcycles, mobile phones and cars, at rates Wada described as roughly at the statutory ceiling.⁵⁷ The overseas receivable balance stood at approximately ¥160 billion at year-end.⁷ It employs 1,311 people — a third of the group's workforce for 6% of revenue — because underwriting includes physical home and address verification, which Wada contrasted with competitors approving loans online in an hour.⁷

Asked directly by an overseas investor how to think about funding cost, credit risk and cyclicality in a business growing this fast, Wada's answer was a single sentence: trust that the underwriting standards will not be loosened.⁷ For a book of this size lending at ceiling rates in two emerging markets, that is a thin answer, and it is the most conspicuous gap in an otherwise detailed disclosure package.

Beverage (¥85.3 billion, ¥9.7 billion). Premium Water Holdings, 70.1% owned, is the largest employer in the group.⁵ It reported its own fiscal 2026 revenue of ¥80.3 billion and operating profit of ¥12.6 billion, its eighth consecutive year of profit growth, with 1.82 million customers, an industry share of about 32%, and — the number that matters — roughly 90% of new customers signing contracts of five years or longer.¹² That is the strongest contractual lock in the group. Note the correction to the common telling: the share is around a third of the home-and-office delivery market, not half.

Insurance (¥31.5 billion, ¥9.4 billion). Non-life and life insurance and warranty services, materially reshaped by the group obtaining its own non-life licence through さくら損害保険 Sakura Insurance in June 2019.⁵ Stock profit grew 35% while acquisition cost grew 66% — the group's most aggressive investment ratio, and the clearest signal of where management currently sees the best returns.³

Agency Sales (¥98.5 billion, ¥12.8 billion) and Solutions (¥26.9 billion, ¥3.8 billion). These are the legacy and the experiment. Agency sales — distributing other manufacturers' and carriers' products — shrank 9% in revenue and 12% in profit, partly reflecting subsidiary disposals.⁴ Solutions, the customer-management and payment-platform business including EPARK, shrank 4% in revenue while profit grew 48% on restructuring.⁴³ Worth flagging plainly: Solutions is 3.7% of group revenue, not the 20% profit engine it is sometimes described as. Corporate software is not what this company does at scale.

A note on the insurance mechanics, because they change the accounting. Under IFRS 17, the insurance segment recognised insurance revenue of ¥25.2 billion against insurance service expenses of ¥19.4 billion, with a small reinsurance loss.⁴ In plain terms, that portion of the segment is no longer a commission business at all: Hikari collects premiums, holds reserves and pays claims. It is a smaller, more profitable version of the same idea as the microfinance book — the group has moved from being paid a fee for finding a customer to owning the customer's economics, and therefore the customer's risk.

Two things to hold onto. First, group gross margin was 49.5% and one-off business disposal gains of ¥3.9 billion sat inside reported operating profit, so the underlying operating profit growth was 7% rather than 11%.⁴³ Second, the balance sheet carries ¥60.6 billion of capitalised contract costs — the portion of agent commissions deemed recoverable and therefore booked as an asset rather than an expense — and the recoverability of that asset is one of only three critical accounting judgments management identifies.⁴ If churn were to deteriorate materially, the write-down would arrive in the income statement.

Portfolio pruning is part of the discipline. During fiscal 2026 the group sold five businesses — a CRM operation, an insurance services unit, an office-equipment sales business, a medical digitisation venture and an energy-efficiency business, with combined annual revenue of roughly ¥4.1 billion — and made a single acquisition, an internet service provider with about 40,000 users contributing roughly ¥120 million of stock profit.¹³ Management's stated policy is to sell at fair valuations and buy cheap.⁷ The stated sustainability framework goes further, committing the group to scale capital-efficient businesses and shrink, exit or sell capital-inefficient ones.⁵ Five disposals against one purchase in a record year is at least consistent with that language, and it is the opposite of empire-building.

Which brings us to the other half of the company, where the profits do not arrive in the income statement at all.

---

↑ Back to Top

VI. Capital Allocation & The Listed Equity Investment Engine (2:07 – 2:32)

In July 2019, the company added a line to its corporate chronology that would eventually rival everything above it: 純投資 net investment, begun in earnest.⁵

Seven years later, the scale is difficult to overstate. Against an acquisition cost of ¥852.3 billion, the portfolio carried a market value of ¥1,476.4 billion at March 2026 — an unrealised pre-tax gain of ¥624.1 billion.³ The equity investments sit on the balance sheet as ¥1,366.2 billion of non-current financial assets plus ¥319.7 billion of equity-method investments, against ¥539.9 billion of cash.⁴

What is actually in it. The securities report discloses the holdings of the group's three largest investing entities. A partnership called 光通信KK投資事業有限責任組合 held 525 listed names worth ¥378.9 billion; an overseas partnership held 115 listed names worth ¥222.6 billion; and the parent company itself held 37 names worth ¥21.8 billion.⁵ Separately, 42 listed companies were equity-method affiliates at March 2026, with voting stakes from 20.2% to 43.0%.³ The independent analyst Michael Fritzell, who reconstructed the book from Japanese disclosure filings, characterised the strategy as targeting companies with recurring revenues, rising profitability and strong balance sheets at earnings yields of 15% or higher, often in illiquid small caps.¹⁴

The affiliate list reads like a directory of unfashionable Japanese cash generators: シナネンホールディングス Sinanen Holdings at 42.8%, ベルパーク Bell-Park at 36.1%, 第一実業 Daiichi Jitsugyo at 29.4%, 東建コーポレーション Token Corporation at 26.4%, フォーバル Forval at 25.4%, イーレックス Erex at 22.6%.³ Several are themselves distributors of telecom and energy services — meaning Hikari owns large stakes in companies that are simultaneously peers, channels and suppliers.

The metric management actually watches. Rather than mark-to-market, the company reports look-through operating profit: its ownership share of each investee's operating profit, summed. That figure reached ¥139.9 billion in fiscal 2026, giving an "earnings yield" of 16.3% on acquisition cost, up from 15.8%.³ Cash dividends received on cost reached 4.3%, versus 3.0% five years earlier.³ Management put the trailing nine-year pre-tax internal rate of return at 18%.³

That construction deserves scrutiny in both directions. It is genuinely useful, because it measures the underlying earning power bought rather than the price the market currently assigns — the discipline of an owner, not a trader. It also cannot decline when share prices fall, which makes it a poor early-warning indicator, and the improvement from 15.8% to 16.3% partly reflects investees' own profit growth in an unusually strong period for Japanese corporate earnings.

The exit engine is where the returns crystallise, and it is not stock-picking alone. Fiscal 2026 saw the company tender into 16 takeover bids and management buyouts, deploying ¥10.5 billion of cost for ¥19.9 billion of proceeds — a cumulative pre-tax gain of ¥9.4 billion at a cost-weighted average internal rate of return of 54%.¹³ The standout was レジル Rezil, taken private by Bain Capital: a 16.1% stake bought for ¥4.3 billion sold for ¥8.3 billion, a 73% IRR.¹³ Others included ナカヨ Nakayo acquired by あいホールディングス Ai Holdings, Fast Fitness Japan taken private by its founding family and a fund, MUTOH Holdings acquired by ブラザー工業 Brother Industries, and カイノス Kainos acquired by デンカ Denka.¹³

This is the crucial insight about the portfolio: Hikari Tsushin is not primarily betting on multiple expansion. It is positioned as a large minority holder in exactly the kind of Japanese small caps that get bought — cheap, cash-rich, controlled by ageing founders, and under mounting pressure from the Tokyo exchange's campaign against sub-book valuations. Every private equity buyout and management buyout in that universe is a liquidity event for Hikari at a premium. Whether that continues depends far less on Hikari's skill than on the durability of Japan's corporate reform and buyout wave.

A case study in using an affiliate as an operating asset. The clearest example arrived in the fiscal 2026 materials. Hikari Tsushin had spent roughly two years building its own propane gas business, working with nine regional LP gas operators because it owned none of the filling, delivery or safety infrastructure the product requires — which meant its margin was thin and split nine ways.⁷ Its solution was not to buy the infrastructure. It was to consolidate almost all of that business into Sinanen Holdings, in which it already held about 43%, and cross-sell its own insurance and water products into Sinanen's customer base.⁷ Sinanen is Japan's third-largest propane supplier, serving roughly 560,000 households, with forecast revenue of ¥367.3 billion and operating profit of ¥4.4 billion for the year to March 2026.⁷

The logic is worth spelling out, because it is unusual. Rather than earning a small distribution margin itself, Hikari routes the volume to a company whose profits it books at 43% through the equity method, improves that company's logistics density, and simultaneously raises revenue per customer on both sides. Wada was explicit that the stake in Sinanen was not acquired in order to consolidate it — the test was simply whether the shares were cheap relative to intrinsic value — but that once held, it created options a purely financial investor would not have.⁷ This is where the "no wall between operations and investment" claim earns some credibility. It is also a reminder that the portfolio's reported earnings yield captures none of these operating benefits, and none of the conflicts they create for minority holders of the affiliate.

The accounting point that matters more than the returns. Hikari elects to carry these equities at fair value through other comprehensive income. In plain English: gains and losses — including realised gains on sale — never pass through the income statement. In fiscal 2026, ¥134.5 billion was transferred directly from other comprehensive income into retained earnings, bypassing reported profit entirely.⁴ The tender-offer gains catalogued above were recognised as comprehensive income, not net profit.¹³

The consequences cut both ways, and investors should hold both in mind:

• Reported net profit understates the economics in good years. Comprehensive income attributable to owners was ¥306.0 billion in fiscal 2026 — double reported net profit — and equity rose 29.6% to ¥1,185.7 billion.⁴³
• Reported net profit will also overstate the economics in bad years, because a portfolio drawdown will hit equity and book value without ever appearing in the profit line investors quote.
• The ¥624.1 billion of unrealised gain carries a deferred tax liability, and the balance sheet duly shows ¥210.7 billion of deferred tax liabilities, up from ¥152.0 billion.⁴ Any "net asset value" that counts the portfolio at gross market value is overstating what shareholders own by roughly that amount.

The funding side, and the turn. This portfolio is not funded purely by operating cash. Interest-bearing debt stood at ¥1,088.5 billion at March 2026.⁴ Fiscal 2026 bond issuance totalled ¥199.5 billion, including a US dollar bond equivalent to ¥73.5 billion at 6.130%, and yen bonds at 1.654% to 2.656%.⁴ In April 2026 the board approved two further issues, at 2.823% for five years and 3.510% for seven.⁴

Read the coupon history and the strategic pivot explains itself. Bonds issued in 2021 carried coupons as low as 0.20%; the 2026 issues carry 3.510%.¹⁵ Interest paid rose 19% to ¥15.9 billion in fiscal 2026.³ Buying equities yielding 16% on cost with money borrowed at 0.2% is close to a free lunch. At 3.5%, it is a decision.

Management has drawn the obvious conclusion. Wada said the phase of aggressive net buying is over, disclosed that the fourth quarter turned to net selling of ¥61.9 billion, described the company as entering a period of gradual debt reduction, and committed to raising required investment yields in line with funding costs across equities, microfinance and the core business.⁷ Long-term debt is projected to decline gradually while net debt to EBITDA stays below 4.0 times.¹⁶ The financial discipline framework is specific: hold liquidity of at least roughly 140% of bonds and loans maturing within three years. At March 2026 that ratio was 174%, with ¥764.5 billion of cash and liquid bonds against ¥439.0 billion of three-year maturities.¹⁶

Japan Credit Rating Agency affirmed A+ with a stable outlook in January 2026, and its language is a fair independent summary: counting the listed portfolio, the company is net debt free; but JCR noted it will continue paying attention to the large holding of listed securities carrying price fluctuation risk, and to the effect of stock price and interest rate trends on the group's financial position.¹⁵

Shareholder returns. The fiscal 2026 dividend of ¥751 per share marked the fifteenth consecutive annual increase and the twenty-third year without a cut; ¥780 is forecast for fiscal 2027.³ Buybacks totalled ¥6.1 billion in fiscal 2026, with ¥11.8 billion planned for fiscal 2027 including a new authorisation of up to 350,000 shares or ¥10 billion resolved on May 13, 2026.³ Shares outstanding fell from 44.27 million to 43.99 million, and 280,000 treasury shares were cancelled in December 2025.⁴ Over ten years the total payout ratio has averaged 34%.³

The dividend record is genuinely impressive. The payout ratio is the sore point, and it is where the sharpest question of the year was asked.

---

↑ Back to Top

VII. Helmer's 7 Powers & Porter's 5 Forces Analysis (2:32 – 2:52)

Evaluating Hikari Tsushin through standard strategic frameworks requires testing each claimed advantage against disclosed operational performance and clear falsification criteria.

Process power — the strongest candidate, and not what it used to be. The traditional narrative credits the group's former 1990s in-house sales force. That sales force no longer exists in its original form; the group maintains 4,012 employees across seven business units, and the high-revenue electricity operation employs just 261 people.⁵ What persists is a repeatable framework for pricing, deploying, and measuring third-party sales efforts across varied product categories under a unified financial formula. In fiscal year 2026, the group deployed ¥69.6 billion in acquisition costs to generate ¥19.2 billion in new stock profit, tracking returns precisely by product line and voltage class.³

The test of this process power is whether returns hold when management refuses to compete purely on price. Execution in fiscal 2026 aligns with that test: high-voltage contract additions fell by design as management targeted ¥130 billion in fiscal 2027 operating profit rather than committing an additional ¥2 billion to ¥3 billion in acquisition spend to lower-yielding contracts.⁷ The counter-evidence appears in regulatory filings: an outsourced distribution network structured around aggressive contract conversion has twice drawn regulatory warnings over customer disclosures.¹⁰¹¹ Process power reliant on external sales networks introduces ongoing compliance exposure.

Cornered resource — the underrated one. The group's most durable asset is not its sales model, but its residential rental distribution network: long-term partnerships and joint ventures with most of Japan's top ten property management companies, which serve as the primary origination channel for low-voltage electricity contracts.⁷ A property manager that enrolls incoming tenants into Hikari's electricity service creates a captive channel across a steady stream of new households. President Hideaki Wada contrasted this channel with online customer acquisition, where price competition is intense and switching rates are high.⁷ This arrangement functions as a cornered resource that limits direct competitive access.

Counter-positioning — plausible, unverified. The bundling rationale is straightforward: a single customer relationship encompassing electricity, water, insurance, and telecom services spreads customer acquisition costs across multiple margin streams, creating an economics structure single-product competitors struggle to match. However, Hikari Tsushin does not disclose cross-sell ratios, average revenue per user across bundled lines, or the proportion of multi-product households. Without empirical disclosure, bundling remains a structural hypothesis rather than a verified strategic moat.

Switching costs — genuine in water, weak in electricity. Approximately 90% of Premium Water's new subscribers sign five-year contracts, establishing clear contractual friction.¹² Conversely, switching retail electricity providers in Japan is frictionless by regulatory design. The primary barrier to churn in residential power is consumer inertia rather than financial penalties. While inertia generates predictable cash flows, it does not constitute a structural moat and degrades as price-comparison tools gain adoption.

Scale economies — modest, and mostly overhead. Operating a three-person holding company overseeing ¥735 billion in revenue illustrates lean central management, generating ¥24 million in operating profit per employee.⁵³ However, the core energy business achieves little purchasing scale: procuring 100% of its power from the wholesale market means the group pays market clearing prices identical to its peers.⁷ Scale in this model reduces administrative overhead rather than direct cost of goods sold.

Absent powers. Hikari Tsushin lacks brand power; Japanese households do not deliberately select Hikari Tsushin based on consumer brand equity. It also lacks network effects, as adding subscribers does not enhance service value for existing customers. Analyzing the company as a network platform misinterprets its underlying economics.

The peer question, which is genuinely strange here. The competitive dynamic in Japanese SME telecom and energy distribution is unusual. Hikari Tsushin's direct peers include ベルパーク Bell-Park, フォーバル Forval, ラストワンマイル Last One Mile, and INEST — all of which appear on Hikari's balance sheet as equity-method affiliates, with voting stakes of 36.1%, 25.4%, 36.5%, and 40.8% respectively.³ The company maintains substantial equity positions in a major segment of its direct operating peer group.

This ownership structure carries two distinct analytical implications. From an operational standpoint, board-level visibility into neighboring distributors provides empirical benchmarking data on industry-wide customer acquisition yields. From a market structure standpoint, high equity stakes make competitive intensity partly endogenous: holding 40% of a competitor reduces incentives to engage in aggressive price competition. However, this cross-ownership exposes both the investment book and operating revenue to identical sector headwinds, creating concentration risk during downturns in Japanese small-business distribution.

Beyond affiliated distributors, primary competitive pressure originates from digital price-comparison and switching platforms that convert inertial accounts into active shoppers — a substitution risk management explicitly identified.⁷

Porter, applied.

• Supplier power (moderate, and structurally neutralised). Telecom carriers and wholesale power generators remain concentrated, but market-linked tariff structures pass wholesale cost fluctuations directly to end consumers. The trade-off is that Hikari Tsushin achieves no structural input cost advantage, competing on distribution efficiency rather than commodity pricing.
• Buyer power (low individually, high in aggregate). Individual households hold negligible negotiating leverage. However, collective buyer behavior — specifically churn responses to bill adjustments — represents a major operational variable, prompting management to cap bill spikes by amortizing excess charges over three months.⁷
• Threat of entry (low for the channel, low-to-moderate for the product). Entering retail electricity reselling requires limited capital. Conversely, replicating three decades of formal partnerships with Japan's major property management firms presents high time and relationship barriers. Wada noted that rapidly duplicating this physical distribution network is improbable.⁷
• Substitutes (moderate and rising). Web-based aggregator platforms drive intense price sensitivity. Management acknowledged that digital switching services pose the most direct challenge to customer retention.⁷
• Rivalry (intense in high-voltage, restrained in low). The group responded to aggressive pricing in high-voltage commercial power by scaling back originations rather than cutting prices.⁷ This response reflects capital discipline alongside an absence of pricing power among sophisticated corporate buyers.

Myth versus reality. Three common market perceptions are contradicted by corporate filings. Hikari Tsushin is not primarily an SME direct-sales organization; its primary growth channel centers on residential rental properties, with customer acquisition spending directed predominantly toward individual households.⁷ It is not a corporate software vendor; the Solutions segment accounts for under 4% of consolidated revenue.⁴ Finally, it does not operate as a traditional keiretsu cross-holder: public filings show zero listed strategic holdings, carrying only 231 unlisted positions valued at ¥692 million and classifying every listed equity holding as a pure financial investment.⁵ Governance analysis must focus on disclosed capital allocation rather than legacy cross-shareholding assumptions.

↑ Back to Top

VIII. Management Credibility, Governance, & Activist Stress Test (2:52 – 3:12)

Shigeta, born February 25, 1965, has served as Representative Director and Chairman since June 2003 and as chief executive since November 2000 — the month the crisis was at its worst.⁵ He attended all nine board meetings in fiscal 2026.⁵ He gives almost no interviews. And he is paid almost nothing.

That last fact is worth dwelling on. Total director remuneration at the parent, excluding audit-committee and outside directors, was ¥569 million across four people in fiscal 2026 — of which Wada took ¥383 million and Managing Director 髙橋正人 Masato Takahashi ¥178 million.⁵ Arithmetic leaves under ¥10 million for the remaining two. There is no performance-linked component anywhere in the structure; every yen is fixed base salary.⁵

For most companies that would be a governance failing. Here it reflects where the incentive actually sits. Shigeta's family vehicles and personal holding together controlled roughly 48% of the voting stock at his most recent large-shareholding report, and the March 2026 register shows 有限会社光パワー Hikari Power at 29.06%, 合同会社光パワー本家 at 5.37% and Shigeta personally at 2.73%, alongside 株式会社鹿児島東インド会社 Kagoshima East India Company at 7.53%, whose representative director is his son 重田光時 Mitsutoki Shigeta.⁵ The top ten holders account for 69.87%.⁵ A founder owning nearly half the company does not need a bonus plan; he needs the share price to compound. That is the cleanest possible alignment on long-term value per share — and the reason minority holders have essentially no ability to force change.

Wada, born December 1973, is the operator. He joined in April 1997, ran the network and information-communications divisions, became a director in 2004, took over sales in 2017, became President in June 2019 and added the COO title in April 2026.⁵ He also sits on the board of エムティーアイ MTI, one of the group's affiliates.⁵

The company devotes an entire appendix to a single fact about him: since taking office, Wada has applied 100% of his after-tax compensation to buying Hikari Tsushin shares — ¥780 million of cumulative net pay over six years, against ¥1.78 billion of share purchases, 2.2 times his take-home.¹⁷ The appendix notes, pointedly, that he began doing this before Berkshire Hathaway's Greg Abel made similar remarks.¹⁷ It also carries a footnote: the ¥1.78 billion excludes share purchases funded by loans from Hikari Tsushin.¹⁷ A chief executive buying stock with his own money is a strong signal. A chief executive borrowing from the company to buy its stock is a related-party arrangement, and the terms of those loans are not disclosed.

Behaviour over time — the test that matters. Management credibility is not measured by tone; it is measured by what happens to prior promises.

In February 2026, Hikari Tsushin raised its full-year guidance to revenue of ¥760 billion, operating profit of ¥115 billion and net profit of ¥120 billion.¹⁸ The outcome was revenue of ¥734.8 billion — a shortfall of ¥25 billion, or 3.3% — with operating profit above guidance and net profit ¥31 billion above it, largely on foreign exchange.⁴ The pattern is consistent: the operating line is delivered or beaten, the revenue line is loose, and the net line is sandbagged on non-operating items. Fiscal 2027 guidance repeats the shape, projecting net profit down 20% to ¥120 billion on an assumed rate of ¥157 to the dollar and no one-off gains, while operating profit rises 11% to a record ¥130 billion.⁴³

Two behaviours weigh in management's favour. Wada volunteered that fiscal 2026 operating profit included ¥3.5 billion of one-off disposal gains, and that the underlying fiscal 2027 growth rate was therefore 14% rather than the headline 11% — correcting his own optimistic number downward.⁷ He also disclosed a decision investors would never otherwise have seen: the original fiscal 2027 plan assumed heavier acquisition spending and about ¥127 billion of operating profit, and he chose instead to ease off in light of Middle East energy conditions and book the profit.⁷ He added that the company has no notion of needing to spend its acquisition budget.⁷

Two behaviours weigh against. Asked on the same call about a ¥100 billion fund initiative with レオス・キャピタルワークス Rheos Capital Works and other asset-management diversification, and about new businesses, the answers were that things are progressing well but not yet disclosable — with a claim that existing businesses alone can reach ¥200 billion of operating profit within four to five years, and that new businesses under incubation are meant to carry the group from ¥200 billion to ¥300 billion.⁷ That is a bold long-range target with no supporting detail, delivered by a company that publishes no formal medium-term plan; its stated management indicators are simply stock profit and consolidated operating profit.⁵ Investors should log it as an aspiration, not a commitment.

The activist case, and management's answer to it. The sharpest exchange of the year came from an overseas investor who did the arithmetic out loud: fiscal 2026 net profit, realised gains and unrealised gains summed to ¥285.5 billion, while shareholder returns were ¥40.5 billion — 14% of the economic gain — even as book value per share rose 30% and the price-to-book ratio fell to about 1.4 times.⁷ The question was whether the payout on comprehensive income could rise.

Wada's response cited the foreign exchange component of net profit, the desire for stable and continuous dividends, and an intention to raise ROE over the medium to long term.⁷ It did not answer the question. That is a fair criticism to log: the dividend policy in the securities report commits only to "stable dividends" with no payout target or capital allocation framework, and the board — not shareholders — sets the dividend.⁵

A skeptical investor's full charge sheet would read:

• Complexity as a discount machine. Wada himself identified this as the risk investors underestimate — that a conglomerate blending stock-type operating businesses with a ¥1.5 trillion equity book is not well understood, and that few Japanese companies are run primarily as capital allocators.⁷ Management's arithmetic on the March 2026 call: net cash assets of about ¥1.1 trillion, a core business earning roughly ¥100 billion a year, and therefore an implied "real" price-to-earnings ratio near 6 times.⁷ That calculation is management's, it counts the portfolio at gross market value before the deferred tax liability, and readers should treat it accordingly.
• A third-party valuation on the record. In the FT Group share exchange documentation, the financial adviser 青山トラスト会計社 Aoyama Trust & Accounting valued Hikari Tsushin shares at ¥40,030–¥43,096 on market prices but ¥60,130–¥75,514 on discounted cash flow.¹⁹ The DCF used company-supplied projections, so it is not independent evidence of value — but it is a disclosed, auditable statement of how management's own plan translates into per-share value.
• Buyback capacity versus float. Asked whether repurchases are constrained by the low free float that 48% family ownership implies, Wada said over ¥100 billion of capacity remains and pointed to the use of treasury shares in the FT Group transaction as a balancing mechanism.⁷ The constraint is real but not yet binding.
• Related-party optics. The FT Group deal, announced March 31, 2026, converted a 72.6%-owned listed subsidiary into a wholly owned one at 0.03 Hikari shares per FT Group share, delisting it from the Standard Market on July 30, 2026 and taking effect August 1.¹⁹⁵ Eliminating a parent-subsidiary listing is a genuine governance improvement. But it was structured as a simplified share exchange requiring no Hikari shareholder vote, and FT Group's second-largest holder was 株式会社HCMAアルファ HCMA Alpha, whose representative director is Wada.¹⁹⁵ Nothing disclosed suggests impropriety; the point is that a transaction transferring Hikari shares to a vehicle headed by the chief executive, without a shareholder vote, is exactly what an activist would ask about.

One more disclosure asymmetry belongs on the list. Hikari Tsushin publishes remarkable granularity about things that flatter it — deal-by-deal tender-offer returns, look-through earnings yields, the chief executive's personal share purchases — and comparatively little about the things that would test it. There is no disclosed churn rate, no revenue-per-customer figure, no cross-sell statistic, no cohort payback curve, and no entry yield on the current year's equity purchases, which management declined to quantify when asked directly.⁷ None of that is a violation of any rule. It is, however, a pattern: the metrics chosen are the metrics that are working.

Countervailing, and worth stating: an Investment Audit Committee has existed since June 2022, chaired by independent outside director 新村健 Ken Shinmura with a majority of independent members, meeting quarterly with full attendance and reviewing both the investment portfolio and the financial discipline framework.⁵ Three of eight directors are outside directors, two of eight are women.⁵ And the disclosure package — segment-level stock profit and acquisition cost, deal-by-deal tender-offer IRRs, fifteen-year cumulative taxes paid of ¥311.7 billion — is more granular than most Japanese large caps offer.³⁷

Which makes the regulatory file the more jarring counterpoint.

---

↑ Back to Top

IX. Material Risk Radar & What Could Go Wrong (3:12 – 3:27)

On May 19, 2026 — six days after the group reported record results — Haruene disclosed that it had received a business improvement recommendation from Japan's Electricity and Gas Market Surveillance Commission under the Electricity Business Act.¹¹

The commission found that sales agents had given explanations liable to mislead customers and that pre-contract disclosure documents were deficient.¹¹ Haruene stated it had terminated its contracts with the involved agents, halted their activities, and confirmed that existing customer contracts were unaffected.¹¹ It was the second such regulatory action in two years: in April 2024, the commission cited Haruene for taking an extended period to provide switching contract numbers to certain customers, as well as for agents potentially providing misleading information.¹⁰

This is the sharpest operational risk in the file, and it takes top priority because it is already a live issue rather than a theoretical scenario.

Regulatory and sales-channel risk. The risk is structural rather than incidental. Converting sales into a variable cost paid per contract to third-party agents shifts volume risk away from payroll, but it shifts compliance risk directly onto the regulator's desk. Two regulatory recommendations against a subsidiary that generates over 20% of group revenue — occurring two years apart for similar agent behaviors — point to a recurring control challenge.⁵¹⁰¹¹ In Japan, regulatory enforcement escalates from recommendations to formal administrative orders and operational license conditions. Hikari Tsushin's own disclosures acknowledge that legal violations by the group or its sales partners could disrupt operations.⁵ Outside analysts offer a blunter assessment, questioning the durability of the group's sales moat and comparing high-pressure agent tactics to multi-level marketing.¹⁴

Funding cost and refinancing. This represents the clearest quantified risk. A group carrying over ¥1 trillion in debt refinanced borrowing with 0.20% coupons into new debt carrying 3.510% coupons within five years.¹⁵⁴ Refinancing every ¥100 billion of debt at a 300-basis-point increase adds ¥3 billion in annual pre-tax interest expense against annual operating profit of ¥116.7 billion. Certain debt facilities also carry financial covenants.⁵ Management's mitigations are substantial: liquidity coverage stands at 174% of three-year debt maturities, net debt to EBITDA is capped at 4.0 times, and management intends to deleverage after modestly expanding debt to pre-fund major maturities due in five and nine years.¹⁶⁷

Equity market drawdown, with an accounting twist. The equity portfolio's ¥624.1 billion unrealized gain represents 53% of total shareholders' equity. A 25% drop in Japanese small-cap equities would erase roughly ¥370 billion in gross portfolio value and a substantial portion of book value per share. Because the group elects to account for these holdings at fair value through other comprehensive income, such a drop would bypass the reported income statement entirely. Two sub-exposures warrant attention: the group's overseas partnership holds 115 listed equities valued at ¥222.6 billion, carrying ¥102.3 billion in unrealized gains while generating only ¥448 million in dividends — a 0.2% yield that presents a markedly different risk profile from the domestic value portfolio — and recorded a ¥2.9 billion realized loss during the fiscal year.⁵ Furthermore, the exit strategy relies on an ongoing wave of corporate reforms and buyouts in Japan, which management noted has reduced the availability of sub-book-value opportunities compared to prior decades.⁷

Cash conversion. The most overlooked figure in the recent financial results is cash conversion. Business operating cash flow fell 71% to ¥21.8 billion even as EBITDA rose 11% to ¥133.1 billion, because receivables — primarily from the expanding Asian microfinance portfolio and customer installment contracts — absorbed ¥110.5 billion.³ While balance-sheet growth in a lending business naturally consumes cash — reducing reported operating cash flow to ¥57.1 billion from ¥84.8 billion a year earlier — this gap is precisely where asset-quality issues in a receivables-heavy business would first surface.⁴

Accounting judgments to watch. Management highlights three critical accounting estimates: the impairment of non-financial assets, the recoverability of deferred tax assets, and the recoverability of capitalized contract costs.⁴ The third judgment is the most active, given the ¥60.6 billion capitalized contract balance. Consolidated goodwill remains modest at ¥18.4 billion, indicating that recent acquisitions have not overpaid for assets.⁴ Meanwhile, equity-method earnings of ¥26.9 billion included ¥5.6 billion in one-time bargain purchase gains from newly added affiliates, which should be excluded when evaluating sustainable operating performance.⁴ Auditors raised no going-concern reservations.⁴

Microfinance credit. The group maintains a ¥160 billion credit book offering loans at statutory ceiling rates in Cambodia and Malaysia. Management cites small-ticket diversification, strict underwriting, and physical address verification as key safeguards, responding to cyclicality concerns by asking investors to trust its underwriting discipline.⁷ However, emerging-market consumer lending at maximum statutory rates historically performs well until a local credit cycle deteriorates. Disclosure on this portfolio remains thinner than on any other major group segment.

Utility working capital under geopolitical stress. Wholesale electricity prices spiked in March and April 2026 amid Middle East tensions before moderating.⁷ While market-linked tariff structures protect operating margins over time, working capital remains exposed to short-term billing deferrals over three-month smoothing windows, as well as potential customer churn if elevated prices persist without government intervention.⁷

Demographics and market share. In response to Japan's aging population and rural depopulation, management argues that its current market shares in electricity and bottled water remain small enough to support organic growth for another four to five years.⁷ While this points to near-term market expansion, it does not alter the long-term reality that a shrinking pool of Japanese small businesses ultimately constrains the addressable market for all of the group's core commercial offerings.

Cybersecurity and data privacy. Data privacy presents an operational risk tied directly to asset valuation. Hikari Tsushin's risk disclosures explicitly connect cyberattacks, system disruptions, and personal data breaches to the recoverability of its receivables and capitalized contract costs: a loss of customer trust would drive cancellations, impairing the assets that represent future billing streams.⁵ Managing utility, insurance, healthcare-adjacent, and consumer-lending data for millions of customers across 170 subsidiaries and independent sales channels creates a wide and complex attack surface, though no material incidents have been reported to date.

Key person risk and technology strategy. Shigeta, at age 61, continues to direct overall capital allocation, and no formal succession plan has been disclosed. When Wada assumed the COO title in April 2026 and joined the board of the group's investment management subsidiary that same month, the moves signaled a potential transition of investment oversight, though management has not made a public statement on executive succession.⁵ On technology, management chose not to act as a reseller for large language model platforms, with Wada explaining that the opportunity lacked sufficient margin and alignment with the company's core strengths — a pragmatic stance that nevertheless skips a growth path pursued by peers like SoftBank.⁷ The more significant long-term artificial intelligence risk is structural: AI tools that allow households and small businesses to automate price comparison could erode the customer inertia that underpins the group's recurring subscription cash flows.

Financial reporting changes ahead. International Financial Reporting Standard 18 becomes mandatory for the group for the fiscal year ending March 2028. Management has not early-adopted the standard and is currently evaluating its impact.⁴ While the rule change will not alter underlying cash flows or business economics, it will revise the presentation of the income statement — a relevant shift for a company whose operating profit differs substantially from its pre-tax income.

In aggregate, these risks do not present an immediate threat to solvency. However, they reflect an operational design that leaves key exposures unhedged by choice — including variable power procurement, floating debt costs, and uncommitted dividend frameworks. That strategy has generated strong historical returns, but its ongoing success relies entirely on disciplined execution.

↑ Back to Top

X. Acquired Playbook: Core Business & Investing Lessons (3:27 – 3:47)

1. The unit of account determines the company you become. Hikari Tsushin's 1990s model recognized the reward for a sale before customer cash arrived; its modern architecture instead presents the annuity income it owns alongside the price paid to acquire it in a single table.³ That shift in measurement altered corporate behavior more effectively than any strategy document could. Companies do not become disciplined simply by resolving to be disciplined; they become disciplined by changing what they measure.

2. The most durable asset is usually the channel, not the product. Everything in the group's product lineup—electricity, water, insurance, broadband—is a third-party commodity. The asset that cannot easily be copied is a thirty-year network of relationships with the property management firms that sign incoming tenants.⁷ When evaluating a distributor, the key question is not what it sells, but who guarantees it access. Products can be swapped in an afternoon, while physical distribution partnerships take decades to construct and cannot be acquired quickly at any reasonable price.

3. Converting fixed selling costs to variable expenses creates a real option with a real cost. Cutting headcount by 60% while expanding operating profit per employee twelvefold gave the group the flexibility to curtail customer acquisitions whenever returns disappointed. Management exercised that flexibility in fiscal 2027 by targeting ¥130 billion in operating profit rather than spending down to ¥127 billion to acquire lower-yielding volume.³⁷ The trade-off for that flexibility is reduced control over independent sales agents, a cost reflected in the regulatory record.¹⁰¹¹

4. Where a profit lands in the accounts determines who can see it. Electing fair-value treatment through other comprehensive income keeps realized investment gains out of reported earnings.⁴ While the company's full economics appear in comprehensive income and book value per share, its price-to-earnings ratio offers limited analytical insight. Any business carrying a large equity portfolio requires similar scrutiny: identify the accounting classification first to determine which earnings figure reflects economic reality.

5. Acting as a natural seller in an external buyout wave functions as a business model. Generating a 54% average return across sixteen tender offers in a single year did not stem from forecasting equity prices.¹³ It resulted from accumulating large minority stakes in cash-generative small caps before private equity firms, strategic acquirers, or founding families launched buyouts. The limitation of this model is its dependence on external market cycles, and management has acknowledged that sub-book-value opportunities have become scarcer.⁷

6. Surviving a 99% drawdown required abandoning the business that drove early growth. Hikari Tsushin did not simply attempt to repair its legacy mobile distribution unit; it exited the line, altered its fiscal year-end, closed roughly a thousand retail outlets, and retired the vast majority of its debt.⁸⁵ While typical turnaround narratives emphasize expansion, this recovery relied on systematic scaling back followed by a decade of quiet execution.

7. Alignment and accountability remain distinct properties. An executive structure without performance-linked pay, nominal salary for the chairman, and nearly half the voting equity held by the founding family produced fifteen consecutive years of dividend growth alongside a shrinking share count.⁵³ However, that same concentration of voting power leaves minority shareholders with limited leverage to compel changes in capital allocation, as illustrated by management's unspecific answers on shareholder payout policy. Alignment does not guarantee external accountability.

8. Refusing to deploy capital is an active strategy. Management's decision to walk away from high-voltage commercial electricity contracts in fiscal 2026—refusing to match competitor discounts or reallocate unspent acquisition budgets simply to sustain top-line growth—highlights a rare form of capital discipline.⁷ Most sales-driven organizations struggle to restrict volume when returns compress. The willingness to defer originations when yields fall below target represents a distinctive operational capability.

9. Metrics designed to move in one direction require counter-weighting. Look-through earnings yield on acquisition cost provides a useful measure of underlying corporate earning power.³ Because its denominator reflects historical cost and its numerator tracks investee operating profit, the metric remains insulated from market price declines. Management-selected performance indicators naturally emphasize favorable trends, making it essential for analysts to evaluate them alongside mark-to-market valuations.

10. Pass-through pricing protects margins without creating a competitive advantage. Relying exclusively on market-linked electricity tariffs protected gross margins during wholesale power price spikes that bankrupted competing new entrants, supporting continued segment growth.³⁷ However, this structure provides no procurement cost advantage, as the company purchases power at standard market clearing prices.⁷ Pass-through pricing serves as an effective risk mitigation tool, but it does not replace a distribution moat—which remains tied to the property management partnerships that secure the end customer.

↑ Back to Top

XI. The Investment Thesis: Bull vs. Bear Case & Key KPIs (3:47 – 4:02)

Why this wins from here. The bull case does not depend on optimism about any individual product. It rests on four factors already evident in the financial disclosures.

The recurring profit base compounds mechanically: stock profit has grown at a 12.4% annual rate over fifteen years and is guided to ¥201.5 billion in fiscal 2027, expanding whenever the group adds contracts at yields above its cost of capital.⁷³ Economics are spread across seven reporting segments with no single product dependency, reflecting a deliberate design principle rather than an accident.⁵ The equity portfolio provides both an earnings stream — delivering ¥55.2 billion in cash dividends and interest compared to roughly ¥100 billion from core operations — and an asset base equal to well over half the company's market capitalization.⁷³ Meanwhile, shareholder returns rise annually, funded from a payout ratio that has averaged just one-third of profit, leaving substantial room for expansion.³

This profile benefits from a structural tailwind: ongoing Japanese corporate governance reform continues to pressure sub-book-value companies to buy back stock, raise dividends, or go private, and Hikari Tsushin ranks among the largest minority shareholders across that target group.

A quieter strength in the bull thesis lies in how those returns were delivered. The fifteen-year track record was achieved while distributing capital, not by hoarding it. Cumulative dividends of ¥248.2 billion and buybacks of ¥103.0 billion — totaling ¥351.2 billion in shareholder payouts — were funded alongside an equity base that expanded nearly twelvefold.⁷ Compounding that survives a sustained distribution policy demonstrates greater operational durability than growth dependent on total profit retention.

What breaks the case. The bear case centers not on a single catastrophe, but on a convergence of structural pressures.

Funding costs are rising faster than investment yields. A portfolio generating a 16% yield on historical cost against debt issued at 0.2% functions as a powerful compounding engine; measured against new debt carrying coupons of 3.5% and higher, it becomes a spread business with a narrowing margin — a reality highlighted by management turning into a net seller of equities in the fourth quarter.⁷¹⁵ Expansion in electricity relies on an external distribution channel and government subsidies: the property-management alliances belong to third parties that rivals could court, while the ¥12.8 billion in government relief recognized within segment revenue represents more than a third of segment operating profit.⁷⁴ The regulatory file remains open, and a third compliance finding against Haruene would shift scrutiny from control lapses to institutional conduct.¹⁰¹¹ Book value is heavily levered to Japanese small-cap equities, where a market drawdown would compress equity without appearing in reported earnings. Furthermore, the conglomerate discount may persist indefinitely, given that founding-family control prevents external activists from forcing structural change — a dynamic underscored when management deflected questions regarding higher payout ratios.⁷ Finally, long-term targets of ¥200 billion and ¥300 billion in operating profit depend on unproven new incubation ventures; if those fail to scale, the company remains a mid-teens compounder paired with an equity book — a solid outcome, but a fundamentally different thesis.

Underlying these factors is a systemic macro risk that market models frequently overlook. Cash generation, book value growth, and the buyout exit engine all rely on the same favorable environment: Japanese small-cap equities re-rating, corporate governance reforms persisting, domestic interest rates remaining manageable, and households absorbing market-linked utility tariffs. These four conditions have coincided for roughly seven years — matching the duration of the net investment program.⁵ The group has never stress-tested this operational model through a sustained Japanese bear market, leaving a critical gap in historical evidence.

The framework verdict, briefly. Evaluated through Helmer's 7 Powers, the group possesses one probable power in its cornered distribution channel, one credible power in its disciplined acquisition economics, and two unverified claims in counter-positioning via product bundling and non-water switching costs. Under Porter's Five Forces, it operates in structurally unattractive commodity markets — retail power and basic telecom connections — yet extracts superior returns by controlling customer access points rather than owning scarce physical assets. That distribution advantage provides a tangible edge, but it relies on ongoing contract renewals and faces continuous competitive pressure.

The three numbers that matter.

1. Stock profit growth alongside customer acquisition costs. Neither figure provides clarity in isolation. Management explicitly noted the operational trade-off: when stock profit compounds above 10%, acquisition spending accelerates, temporarily restraining operating profit growth; when stock profit expansion slows, operating profit gains accelerate.⁷ Consequently, rising operating profit accompanied by falling acquisition spend signals retrenchment rather than strength. Evaluated together, the two metrics reveal whether the subscription book is expanding at acceptable returns.

2. Portfolio earnings yield on acquisition cost alongside net transaction direction. Look-through yield on cost — which reached 16.3% in fiscal 2026, up from 15.8% — measures underlying investee earning power rather than fluctuating market sentiment, with management reporting the figure annually.³ Tracking this yield against net security purchases or sales provides essential context; the fourth-quarter shift to ¥61.9 billion in net equity sales served as the single clearest operational signal in recent filings.⁷

3. Business operating cash flow relative to EBITDA. The spread between these metrics widened significantly in fiscal 2026 as receivables absorbed ¥110.5 billion.³ For a business operating simultaneously as an annuity distributor and a growing Asian microfinance lender, this ratio serves as the earliest indicator of deteriorating collections, elevated customer churn, or potential impairments of capitalized contract costs.

A brief note on metrics to disregard: reported net profit offers minimal analytical value for this structure. Fiscal 2027 guidance projects a 20% decline in net profit alongside record operating profit, driven entirely by conservative currency assumptions and the exclusion of one-off asset sales.⁴³ Trailing price-to-earnings multiples inherit this distortion. Comprehensive income and book value per share provide far more accurate measures of economic performance, provided deferred tax liabilities on unrealized gains are appropriately deducted.

The single metric investors require but cannot inspect is a disclosed churn or customer persistence rate. While management emphasizes account retention and highlights total subscriber growth, corporate filings omit cancellation rates across electricity, telecom, and insurance lines.⁷¹⁵ For an enterprise whose valuation depends on the duration of recurring customer cash flows, this omission represents the most significant blind spot in an otherwise detailed disclosure regime.

↑ Back to Top

XII. Epilogue & Closing Thoughts (4:02 – 4:10)

There is a photograph that does not exist, and it explains this company better than any chart.

It would show Yasumitsu Shigeta in the spring of 2000, 35 years old, sitting in a Tokyo office watching a trading screen on which his company's shares could not clear because no buyers existed. Twenty consecutive mornings of that.

Forty billion dollars in nominal paper wealth evaporated. SoftBank publicly asking him to step down from its board. A magazine exposé alleging that his distribution network had manufactured the contract growth the market had valued at 720 times earnings.

What he did next was practical rather than visionary. He closed storefronts, retired debt, pledged personal capital, changed the fiscal year-end, and ceased public media appearances.

He then spent two decades rebuilding the enterprise around a fundamental question — what does a recurring annuity cost, and what will it return — and another five years applying that same framework to listed equities. The outcome, compounded over decades, is an equity base near ¥1.2 trillion and a share price that has outperformed the broader Japanese market over thirty years, peak drawdown included.⁴³

Twenty-six years after the crash, the group reports record operating profits, delivers its fifteenth consecutive annual dividend increase, and oversees a securities portfolio with a market value larger than the operating subsidiaries that funded it.⁴³

Yet over that same period, the company was cited twice by Japan's energy regulator over misleading disclosures by sales agents, declined to provide clear guidance on higher shareholder payouts, and set long-range operating profit targets without disclosing supporting operational details.¹⁰¹¹⁷ Both aspects define the firm. Evaluating only the operational recovery misses the structural risks that preceded the original collapse.

The primary lesson of Hikari Tsushin's trajectory is not survival alone. It is that institutional crisis forced a fundamental shift in the core unit of account — from gross contracts signed to recurring annuities owned net of customer acquisition costs — and that this accounting discipline drove long-term capital compounding. Organizations ultimately reflect the metrics they prioritize.

The central analytical question is whether the current operational phase will prove as durable as the post-2000 recovery. Management has stated that existing business units can reach ¥200 billion in operating profit within four to five years, with undisclosed incubation ventures intended to drive profits from ¥200 billion to ¥300 billion.⁷ Concurrently, management signaled that the period of low-cost debt expansion to acquire under-valued equities has ended, transitioning the group into a net seller of equities focused on debt reduction.⁷ These two objectives require a strategic shift from capturing financial spreads between 16% look-through earnings yields and sub-1% borrowing costs toward generating organic operational growth in core markets.

For long-term investors, this operational shift represents the core thesis. Hikari Tsushin has demonstrated an ability to acquire recurring cash flows, measure acquisition returns rigorously, and maintain discipline when contract yields compress. It has also successfully accumulated minority stakes in cash-generative small caps and realized gains through external corporate buyouts. What remains unproven is whether the core operating businesses can sustain historical compound returns without the tailwind of cheap debt and equity portfolio expansion. As funding costs rise and the conglomerate discount persists, the company's capital allocation discipline faces its clearest test.

Financial markets have long struggled to value a corporate structure that operates simultaneously as a commercial distributor and a leveraged equity portfolio managed by a lean holding company. That structural ambiguity remains central to evaluating the firm's intrinsic value.

---

References

1. 株価大暴落の記録（光通信241,000円→895円、20営業日連続ストップ安） — ファイナンシャルスター ↩↩↩↩↩

2. Fifteen Years Of Ups And Downs For Yasumitsu Shigeta — Forbes Asia, 2013-04-03 ↩↩↩

3. 2026年3月期 通期 決算説明資料 — 株式会社光通信, 2026-05-13 ↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩

4. 2026年3月期 決算短信〔IFRS〕（連結） — 株式会社光通信, 2026-05-13 ↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩

5. 第39期 有価証券報告書（2026年3月期） — 株式会社光通信 / EDINET, 2026-06-26 ↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩

6. 光通信【9435】株価・企業情報 — 日本経済新聞 ↩

7. 2026年3月期決算説明会（国内・海外）質疑応答 書き起こし — 株式会社光通信, 2026-05-14 ↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩

8. 光通信の歴史 — The社史 ↩↩↩↩↩↩↩↩↩

9. Faces In The News: Yasumitsu Shigeta — Forbes, 2001-02-20 ↩↩↩↩↩↩↩

10. 業務改善勧告等に関するお知らせ — 株式会社ハルエネ, 2024-04-26 ↩↩↩↩↩↩↩

11. 業務改善勧告等の受領に関するお知らせ — 株式会社ハルエネ, 2026-05-19 ↩↩↩↩↩↩↩↩↩

12. プレミアムウォーターHD、2026年3月期 顧客数5％増の182万件 売上高前期比4.5％増の803億円に — 日本流通産業新聞オンライン ↩↩

13. 別紙4 TOB、M&A実績 — 株式会社光通信, 2026-05-13 ↩↩↩↩↩↩

14. Hikari Tsushin's portfolio — Asian Century Stocks ↩↩

15. Credit Rating Affirmation: HIKARI TSUSHIN, INC. (A+/Stable) — Japan Credit Rating Agency, 2026-01-20 ↩↩↩↩↩

16. 別紙3 有利子負債の見通し、財務規律 — 株式会社光通信, 2026-05-13 ↩↩↩

17. 別紙5 和田社長個人による自社株式の取得状況 — 株式会社光通信, 2026-05-13 ↩↩↩

18. 2026年3月期 第3四半期決算短信〔IFRS〕（連結） — 株式会社光通信, 2026-02-12 ↩

19. 簡易株式交換による株式会社エフティグループの完全子会社化に関する株式交換契約締結のお知らせ — 株式会社光通信, 2026-03-31 ↩↩↩