KOSÉ Holdings Corporation

Company history

Financial history 2006–2025 — revenue, cost structure, balance sheet, cash flow and key ratios, year by year →

Founded
1946
Head office
Nihonbashi, Tokyo, Japan
Listed
2000
Founder
Kobayashi Kozaburo
Revenue · FYE Mar 2025
$2.2B (¥330bn)
Net profit · FYE Mar 2025
$100.9M (¥15bn)
KOSÉ Holdings Corporation: long-term performance & turning pointsSales (revenue) and profit-margin ratio
Sales (¥ bn)Net margin (%)

1946Quality over volume, and a partner from Paris

Revenue (¥ bn, bars) · net margin (%, line)
Source: securities reports
  1. 1946Kobayashi Kozaburo founds Kobayashi Gomei in Oji, Tokyo
  2. 1948Reorganized as Kobayashi Kosé
  3. 1956Albion established for prestige cosmetics
  4. 1957La Bonne — the first major hit
  5. 1963Technical alliance with L’Oréal (to 2001)
  6. 1965Head office moves to Nihonbashi

KOSÉ began in March 1946 as Kobayashi Gomei, set up by Kobayashi Kozaburo in Oji, north Tokyo, on the belief that cosmetics could lift the spirits of a country living in the ruins of defeat. He had no capital behind him, so he invented one: a contract-sales system under which retailers paid in before goods were distributed, funding the working capital from the selling side. He then went round the country signing shops up, packing even tiny consignments and paying the freight himself to reach provincial retailers who had nothing to sell. The habit of putting the retailer’s trust first became the company’s standing idea of trade. The firm was reorganized as Kobayashi Kosé in June 1948, and its sales arm split off as a separate company in 1951.

That year’s Pearl-lite Skin, a lotion formulated not to crystallize, sold for years. It was a market in which anything would sell and shoddy goods circulated freely; Kozaburo, disturbed by products priced above their worth, insisted on the best raw materials and fragrances he could buy. When voices inside the company argued for mass production and scale, he said plainly that KOSÉ would seek quality, not volume — a small company able to compete on the excellence of what it made. La Bonne (1957) and Aurique (1962) followed, and a reputation for product development dates from these years.

Two structural choices from the same period still shape the group. In March 1956 KOSÉ established Albion as a separate maker of prestige cosmetics — a parallel product world under separate management, seven years before any foreign tie-up. And in May 1963 it signed a technical alliance with L’Oréal of France, a relationship it would hold for thirty-eight years, until the joint venture was dissolved in 2001. A Sayama factory and a Tokyo research laboratory both opened in 1964, the head office moved to Nihonbashi in 1965 — where it remains — and a Hong Kong company in 1968 began selling abroad.

Read the full history in Japanese →


1970Two price tiers, two sales companies

Revenue (¥ bn, bars) · net margin (%, line)
Source: securities reports
  1. 1970Cosme Decorté launched against the foreign brands
  2. 1979Gunma plant opens
  3. 1980First Deming Prize won by a Japanese cosmetics plant
  4. 1981Kobayashi Reijiro becomes president
  5. 1985Sekkisei launched
  6. 1988Selling split in two; KOSÉ Cosmeport established

Capital liberalization had opened the industry to foreign houses, and department-store counters were the first thing they took. KOSÉ answered in 1970 with Cosme Decorté, the prestige line Kozaburo had long wanted to make, and defended its position by refusing to fight only where the foreigners were strongest: it secured space in independent cosmetics specialty shops as well as department stores and began posting trained beauty consultants into them. Kobayashi Yasukiyo later called that the survival move. Powder foundation Fit-on followed in 1976, and in 1985 came Sekkisei, a lotion of Japanese and Chinese botanical extracts that is still one of the group’s pillar brands worldwide.

Manufacturing earned its own reputation. In 1980 KOSÉ became the first Japanese cosmetics maker to win a Deming Prize for a plant — recognition that later brought OEM work from foreign brands and ISO 9002 certification across every factory. A Gunma plant opened in 1979, and in 1981 Kobayashi Reijiro, the founder’s son, took over as president. Local companies went up across Asia in sequence: Singapore (1971), Malaysia (1972), Thailand and Taiwan (1984, with manufacturing), China (1988).

The most consequential decision of the period was organizational. In 1988 KOSÉ split its domestic selling in two — a sales company for its own counselling-sales products in April, and KOSÉ Cosmeport for the general-goods route in July, charged with reaching a wider public at accessible prices. High-priced lines sold face to face in department stores and specialty shops; low-priced lines were picked off a shelf in mass retailers; different companies, different products. A decade later the two still accounted for 76% and 20% of sales, and separating the selling organizations first is what made it possible to separate the brands in 1991.

Read the full history in Japanese →


1991A brand for every channel, and a public listing

Revenue (¥ bn, bars) · net margin (%, line)
Source: securities reports
FY2006 · consolidated
Revenue$1.5B
Net income$86M
Net margin5.6%
FY2006 · consolidated
Revenue$1.5B
Net income$86M
Net margin5.6%
  1. 1991Renamed KOSÉ; brand-per-channel marketing adopted
  2. 1997Resale price maintenance abolished; Kobayashi Yasukiyo president
  3. 1999Shares registered OTC; revenue $1.2B (¥138bn), No.3 in Japan
  4. 2000Listed on the Tokyo Stock Exchange, first section
  5. 2001L’Oréal joint venture dissolved

In August 1991 Kobayashi Kosé became simply KOSÉ, part of a corporate-identity programme adopted as domestic brands came under pressure from foreign rivals. The research done for it pointed to brand marketing, and KOSÉ acted on the conclusion: it developed a different brand for each distribution channel — department stores, specialty shops, mass retailers, drugstores — where Japanese practice had been to put the same goods everywhere. Internally it was called the selective distribution system; it was also what L’Oréal did, and the alliance supplied the example. Asked what distinguished KOSÉ from Shiseido, president Kobayashi Yasukiyo named this system and the group structure behind it.

The method was strong against price erosion and weak against churn in the channels themselves. Resale price maintenance for cosmetics was abolished in 1997, the year Yasukiyo became president, and discounting broke out; the winners of the shake-out were drugstores, whose head offices bought centrally and could not be served by dispatching beauty consultants store by store. Yasukiyo later admitted the company was slow to retrain its sales force for that model. By 1999 specialty shops were still 28.8% of sales and drugstores 20.5%, and in the year to March 2006 the drugstores’ shift to profitability caught KOSÉ out again, in falling sales and profit.

KOSÉ registered its shares over the counter in December 1999 — consolidated revenue of $1.2B (¥138bn), an 8.4% share of the industry, third in Japan behind Shiseido and Kanebo, with only 5.9% of sales from abroad — and moved to the first section of the Tokyo Stock Exchange in December 2000. In August 2001 the L’Oréal joint venture was dissolved after thirty-eight years. The years that followed added brands with distinct worlds of their own, Stephen Knoll in 2003 and JILL STUART beauty in 2005, and pushed further into China with a sales company in 2005.

Read the full history in Japanese →


2007Attack, tarte, and the holding company

Revenue (¥ bn, bars) · net margin (%, line)
Source: securities reports
FY2007 · consolidated
Revenue$1.5B
Net income$59M
Net margin3.9%
FY2025 · consolidated
Revenue$2.2B
Net income$101M
Net margin4.6%
  1. 2007Kobayashi Kazutoshi becomes president
  2. 2011From defensive to offensive reform
  3. 2014tarte acquired; entry into North America
  4. 2017Recurring profit ¥39.6bn — past Shiseido for the first time
  5. 2019Record revenue $3.1B (¥333bn) on inbound demand
  6. 2024China restructuring; ¥4.4bn extraordinary loss
  7. 2026Holding-company structure; first president from outside the family

Kobayashi Kazutoshi, the founder’s grandson, became president in June 2007 and began with a defensive reform aimed at efficiency. It did not work: consolidated revenue fell from ¥177.8 billion in the year to March 2006 to ¥166.5 billion six years later, and recurring profit from ¥20.6 billion to ¥11.8 billion. In 2011 he switched to an offensive reform — widening both the fields and the geography of the business, opening a direct-sales company that year and placing subsidiaries in India (2013) and Indonesia (2014).

Its clearest expression came in April 2014, when KOSÉ bought the American make-up brand tarte. Founded in 1999 by Maureen Kelly, tarte had no factory and no laboratory — it competed on concept and idea, which is precisely the value Kazutoshi said his own company lacked. Inside Sephora and Ulta, it reached 2,300 doors by 2017, and in three years its revenue rose 3.6-fold to ¥28.2 billion and operating profit 5.6-fold to ¥8.4 billion. At home, prestige lines and Albion drove recurring profit to ¥39.6 billion in the year to March 2017, passing Shiseido’s figure for the first time; inbound tourist demand then carried revenue from ¥190.0 billion in the year to March 2014 to $3.1B (¥333bn) five years later, with record recurring profit of ¥54.0 billion. As Japanese-made gained prestige, Chinese-made lost it, and KOSÉ sold its Chinese plant in 2017.

Then the reversal. COVID-19 cut demand for lipstick and its neighbours; revenue fell to ¥279.4 billion in the year to March 2021 and recurring profit to ¥18.7 billion, roughly a third of the peak. The fiscal year-end moved from March to December in 2021, the listing moved to the Prime Market in 2022, and China turned into a burden — inventory clearance and store and headcount cuts produced a ¥4.4 billion extraordinary loss in 2024. Revenue recovered to $2.2B (¥330bn) in 2025, back at the 2019 level, but recurring profit reached only ¥21.5 billion. On 1 January 2026 the company reorganized into a holding structure as KOSÉ Holdings, and at the March 2026 general meeting the holding company’s presidency passed to Shibusawa Koichi — the first president from outside the founding family — with Kazutoshi remaining as chairman and group CEO.

Read the full history in Japanese →


Key decisions — the author’s view

Revenue (¥ bn) · net margin % · around FY1963

The technical alliance with L’Oréal (1963)

Choosing to let the foreigner in and learn

Rather than shut the foreign company out, take it inside and learn from it. Kobayashi Kosé’s choice in May 1963 rests on that single point. Shiseido, the industry’s largest firm, rejected acquiring a local company even when it entered the United States, choosing instead to set up its own subsidiary and sell under the Shiseido name. KOSÉ, seventeen years old, did the opposite: it signed a technical alliance with what would become the world’s largest cosmetics company and held the relationship for thirty-eight years. Cosme Decorté in 1970 and the 1980 Deming Prize — the first won by a cosmetics maker’s plant — are both results reached partway through those thirty-eight years.

The relationship, though, did not run one way. For L’Oréal, 1963 was also the year it landed in Japan, and by placing its partner’s brands alongside its own in its distribution channels KOSÉ was welcoming into the domestic market a competitor that would later grow on its own. When the alliance ended thirty-eight years on, L’Oréal Japan described it as “buying back the shares from KOSÉ.” What the price of partnering with a foreign firm in order to learn actually was appears in that choice of words.

Revenue (¥ bn) · net margin % · around FY1991

One brand per channel: the 1991 identity change (1991)

What it means to read the sales floor in detail

What KOSÉ stepped away from was the commercial custom of the day: handing every sales floor the same goods. Put Cosme Decorté in department stores, Predia in specialty shops and mass retailers, Marie Claire in drugstores, and you need as many development programmes, advertising campaigns and inventories as you have names. That the company could commit to it in 1991 appears to be because in 1988 it had already separated its counselling-sales and general-goods companies — dividing the selling organization before dividing the products. The two answers Kobayashi Yasukiyo gave when asked what distinguished KOSÉ from Shiseido were exactly these: brand marketing, and group management.

But a method of raising a brand per channel is weak against the speed at which the channels themselves change places. Yasukiyo himself named inventory burden and the consolidation of inefficient brands as problems in 1999, and in the year to March 2006 the company sank to falling sales and profit because it was late adapting to the drugstores’ turn toward profitability. The more finely you read the sales floor, the heavier the rework when the party you read changes. A design that prevents price erosion and the speed to switch to a new channel were hard to hold together inside the same method.

Revenue (¥ bn) · net margin % · around FY2014

Buying tarte: a shelf in North America (2014)

What it means to buy a sales floor

What remained as assets from the $135 million consideration was ¥9,498 million of goodwill, ¥3,519 million of trademarks and ¥2,717 million of customer-related assets. No factory came with it, and no laboratory. A company that had built plants at Sayama and Gunma and formulated its products in its own research institute had bought nothing but a brand and a sales floor. When Kobayashi Kazutoshi called Kelly a “female Steve Jobs,” the praise was at the same time an admission that there were territories his own way of making things could not reach.

Yet what is bought with intangible assets erodes while still intangible. The operating margin fell from its third-year level to 7.9%, raw-material regulation added ¥1.0 billion to cost, and Ulta merely withdrawing from Target is enough to require a returns provision. Earnings that sit on shelves and trends cannot be defended in one’s own hands the way earnings that sit on one’s own equipment can. That KOSÉ nonetheless decided to prioritize marketing investment for the 2026 fiscal year is presumably because it sees this business as not the kind that survives by being defended.

Revenue (¥ bn) · net margin % · around FY2024

Restructuring China: inventory, stores and the local plant (2024)

Sound reasoning at entry, and where the miscalculation lay

From a market where it led all Japanese makers with ¥3.6 billion of sales in 1999, KOSÉ cut its scale back in 2024 while booking a ¥4.4 billion loss. What it withdrew from was not China but the exchange of discounts. The 1988 design — make locally, sell locally — was reasonable measured against the price expectations of a time before Japanese-made cosmetics were prized in China, and the miscalculation lies in the point where the reversal in quality perception emptied a local plant of its meaning. Letting go of manufacturing alone in 2017 appears to have been the response to that single point.

The profit, however, was made by cutting marketing and administrative expenses, and sales have not yet returned. Even into 2026 the mainland is still midway through withdrawing from directly operated stores and shifting to agency distribution, and has not reached the stage of contributing substantial profit. In the same market, while the prestige tier stepped out of the fight, KOSÉ Cosmeport’s lower-priced tier has recovered to nearly 80% of its 2018 overseas sales. This was a decision to re-choose the price tier in which to compete, not a decision to leave the market.

Revenue (¥ bn) · net margin % · around FY2024

The holding company, and a president from outside the family (2024)

An opened presidency, and a founding family that stayed

The appointment of Shibusawa Koichi, who came up through accounting and the president’s office, changes the character of a presidency four generations of the founding family had held. He served as chairman of the Chinese subsidiary, then returned to head general affairs, legal, human resources and quality assurance. Given that the problems of 2024 were inventory disposal and store rationalization in China, and a return on invested capital of 2.6%, the choice has a clear logic to it. Rather than the founding family surrendering the presidency, it is closer to the reality to read this as roles being reassigned to fit the problems that had to be solved.

Because command gathered in the holding company, the weight of Kobayashi Kazutoshi, who remains chairman and group CEO, has if anything increased. What outside directors named as a distinguishing feature of the board was the speed of decision that comes from that chairman’s leadership. Kobayashi Shoichi sits as executive vice-president of the holding company and Kobayashi Takao as vice-president of the operating company. The fact that a president has been drawn from outside the family for the first time and the fact that the family continues to occupy the key posts hold simultaneously. That is why the reading of this as an “interim” arrangement will not go away.

Each heading links to the full Japanese analysis — background, decision and outcome, with sources.


References & sources

This is a condensed English edition. The full, source-by-source history — with the detailed narrative, financial tables, shareholders and executives — is maintained in Japanese: 日本語版(詳細)— KOSÉ Holdings Corporation full history in Japanese →

  1. KOSÉ Corporation — 有価証券報告書 (annual securities reports).
  2. KOSÉ Corporation — 統合報告書 (integrated reports) and 決算短信 (earnings releases).
  3. Kobayashi Yasukiyo — lecture on KOSÉ’s management and brand marketing, 1999.
  4. KOSÉ Corporation — corporate history (沿革) and news releases. kose.co.jp.

Yen amounts are converted at the average rate of each figure’s own year — not today’s rate; revenue charts are shown in yen. Exchange rates & sources — the full ¥/US$ table →


Disclaimer


Data API

KOSÉ Holdings Corporation’s history, financials, executives and shareholders are published as static JSON — no key, plain GET.

Method Endpoint Returns
GET /api/companies.json All companies
GET /api/4922/manifest.json Resource index
GET /api/4922/history.json History overview
GET /api/4922/timeline.json Chronology
GET /api/decisions.json All management decisions (index)
GET /api/4922/decisions.json Management decisions (index)
GET /api/4922/decisions/{slug}.json One decision (full dossier)
GET /api/4922/executives.json Executives
GET /api/4922/shareholders.json Major shareholders
GET /api/4922/financials.json Financial statements
GET /api/4922/financials-longterm.json Long-term results
GET /api/4922/segments.json Business segments
GET /api/4922/regions.json Sales by region
GET /api/4922/workforce.json Workforce