Ancient(ish) Companies That Refused to Disappear

I saw an Instagram post about how Nintendo started as a playing-card company back in 1889. This made me think, “Wow, Nintendo must be one of the oldest companies in the world.”

I googled it and realized I could not have been more wrong.

Nintendo began making Japanese hanafuda playing cards in Kyoto in 1889.¹ That is old by almost any normal standard, but not by the standard of this rabbit hole. There are hotels that welcomed guests before the first millennium ended, breweries that predate the countries they now operate in, and companies that had already been trading for centuries by the time Nintendo sold its first card.

Of course, I started with Wikipedia’s List of oldest companies.²

Before going further, there is one boring but important caveat. Wikipedia’s list is not a census of every old company that has ever existed. It reflects what has been documented, sourced and added by editors. “Survival” is also complicated. Sometimes the same family remains in control. Sometimes the business survives under a new legal entity. Sometimes only the brand or craft continues after an acquisition. A 1,000-year-old inn and a 1,000-year-old public institution are not the same kind of thing.

Still, once the list is cleaned and organized, the patterns are hard to ignore.

01Japan dominates the list

The first thing I noticed was the absolute dominance of Japan. Of the 1,023 companies in the cleaned list, covering founding years from 578 to 1699, 531, or 51.9%, are Japanese.

That includes the oldest operating business represented in the dataset: Kongō Gumi, a construction company whose traditional foundation date is 578 AD. That was nearly 1,450 years ago.³

There is an important continuity caveat. The historic family firm was liquidated, and its work continued under Takamatsu Construction; Kongō Gumi's current corporate entity dates from 2006. So “founded in 578” describes the continuity of the name, craft and operating tradition, not one unchanged legal corporation.

Let that number settle for a moment. Kongō Gumi was already centuries old before the Norman conquest of England. It specialized in building and restoring Buddhist temples, a narrow craft tied to institutions that also think in centuries.

Another thing that jumped out was how heavily the list leans toward food and drink. Alcohol and other beverages account for 25.1%, while food and restaurants account for another 21.9%. Together, that is almost half of the dataset.

This makes Kongō Gumi’s longevity even more impressive. It is a construction business, and construction and property companies make up only 23 rows, or 2.2% of the entire list.

In hindsight, the dominance of food and drink makes sense. Technologies change. Governments change. Borders change. But people continue to eat, drink, gather, celebrate and travel. The product may evolve, but the underlying need does not disappear.

Key takeaway

The dataset is not dominated by glamorous or futuristic industries. It is dominated by businesses attached to recurring human needs and by products whose value often increases when customers believe they carry history.

02The picture changes depending on the century

If we divide the data into five historical periods, with a pre-1300 group followed by four century blocks, we see something more interesting than simple Japanese dominance.

Japan leads before 1300, with 23 of 66 companies (34.8%). It then falls behind Germany in the 1300s and behind Germany and Austria in the 1400s, dropping to just 11.0% of represented companies in the fifteenth century. Germany moves the other way, reaching 36.6% in the 1400s.

Then the picture changes completely. Japan rises to 51.8% in the 1500s and 63.8% in the 1600s.

So the story is not “Japan was always dominant.” Japan's share of the represented companies is high very early, falls behind Germany in the 1300s and Germany and Austria in the 1400s, and then becomes overwhelmingly dominant among the rows from 1500 onward.

There are a few other things worth noticing:

• Switzerland briefly punches far above its size. It contributes 15.9% of the 1300s and 11.0% of the 1400s, before falling to about 2% in the 1600s.
• The number of rows explodes after 1500. There are 82 companies from the 1400s, 226 from the 1500s and 580 from the 1600s.
• That explosion should not be mistaken for a historical startup boom. Later businesses had fewer centuries in which to fail, and surviving records become more plentiful closer to the present.

The category mix changes too, although less dramatically than the country mix.

Alcohol remains remarkably persistent, accounting for between roughly one-quarter and three-tenths of the represented companies in most periods. Hospitality is much more prominent among the earliest survivors: it accounts for 25.8% before 1300, but only 10.2% in the 1600s. Crafts and manufacturing, meanwhile, become much more visible after 1500.

03The Big Guys: the top 30

Being old does not necessarily mean being big. Many of these companies are local hotels, restaurants, breweries, pharmacies or craft businesses. But a few have grown into global corporations.

To compare them, I calculated what I call the Central Value Proxy.

The calculation follows a simple evidence hierarchy:

1. For listed companies, use observed equity market capitalization.
2. For recently acquired or privatized companies, use the disclosed transaction value.
3. For private companies with usable accounts, use disclosed book capital, revenue or EBITDA as the anchor.
4. Where no market price exists, apply a transparent valuation multiple and show a range rather than pretending the estimate is exact.

For a modeled private company, the central value uses the disclosed book value or a stated central multiple; it is not automatically the midpoint of the range. The low and high cases remain attached to the row, along with the reason for choosing the anchor and multiple. That is where subjective judgment enters the ranking.

Foreign-currency values were converted into US dollars using European Central Bank reference rates from 14 August 2026.⁴ The numbered rank uses the central estimate, but overlapping ranges should be read as tiers, not precise positions.

At the top is Barclays, the British financial group founded in 1690, with a market capitalization of approximately $94.73 billion in the snapshot.⁵ It is followed by Merck KGaA, Saint-Gobain and Banca Monte dei Paschi di Siena.

* The source dataset uses 1473 for Gebrüder Weiss. The company's current history documents the Lindau courier service from 1487 and the Gebrüder Weiss name from 1823, so “fifteenth-century roots” is the safest wording outside the table.⁶

† Santa Maria Novella's 1612 date marks the opening of the public pharmacy; its monastic medicinal tradition dates to 1221.⁷

A surprisingly even spread of sizes

If we divide the valued top 30 into four size tiers, they are surprisingly evenly distributed. Five have central value proxies above $10 billion, while the other 25 are split almost evenly across large, mid-market and smaller corporate tiers.

The top 30 are not representative of the whole list. They are the 30 highest-ranked companies in a best-effort screen of entities for which a defensible value proxy could be built. A broader Teikoku Databank study of 46,708 Japanese firms aged at least 100 years provides a useful benchmark for the size distribution of old Japanese firms: among firms with known revenue, 79.0% generated less than JPY 1 billion a year. Only 1.0% generated JPY 100 billion or more.⁸

Yet there is an interesting counterpoint: long-established firms represented 19.4% of all Japanese companies in that JPY 100 billion-plus revenue class. The result looks less like a single winning size and more like a longevity barbell: a huge base of small firms and a thin but meaningful tail of very large ones.

04Large-value firms have a different sector mix

We see a clear shift among the bigger players. Health and Pharmacy has the largest share of the top 30, with Finance and Insurance in second place. Food, alcohol and beverages together fall to 20.0% of the top 30, even though they account for 47.0% of the full dataset.

This is an important distinction. Most represented survivors are concentrated in focused businesses attached to food, drink, craft or place. Within the top-30 value-proxy subset, health and finance are much more prominent.

In other words, the composition suggests two possible models of extreme longevity. Focused and place-attached businesses may benefit from local loyalty, patient ownership and specialized knowledge. Larger firms may benefit from professional management, access to capital, diversification and institutional continuity. This is a hypothesis generated by the survivor data, not a measured difference in survival rates.

05Most survivors look like specialists, not giant corporations

The sector data tells us what these companies do. It does not tell us how they are managed. To generate hypotheses about that question, I divided the 1,023 companies into six operating-format and continuity proxies.

The word proxy matters. For most companies, the list does not provide current revenue, employee count, ownership or management structure. These are transparent clues based on the type of operation and the verified large-company screen, not measured management patterns or audited legal classifications.

The biggest group is the specialist producer or merchant archetype, at 69.2%. Another 20.5% are local-premises operators. But this does not mean that 89.7% of the dataset is made up of small businesses. The specialist group is explicitly size-unresolved, and an old business can remain focused while becoming quite large.

What exactly does “local premises” mean?

It means the place is part of the product. An inn sells a night in a particular building. A pub sells a drink, but also a room, a neighbourhood and a social ritual. A bathhouse, restaurant or hotel depends on customers entering a specific physical setting.

It does not necessarily mean one location, few employees, low revenue or family ownership. A local-premises business can expand, franchise or become part of a larger group. The label simply says that its historic advantage is strongly tied to an address and the experience around it.

This operating format is especially visible in the earliest periods. Local-premises operators account for 34.8% of rows before 1300, 39.1% in the 1300s and 36.6% in the 1400s. Their share falls to 17.7% in the 1500s and 15.5% in the 1600s.

That gives us another possible explanation for extreme longevity. Among the earliest periods represented here, a defensible place, such as a hot spring, pilgrimage route, market square or trusted inn, could function like an asset, a distribution channel and a brand at the same time.

Key takeaway

Specialist producers and merchants are the most common operating-format proxy among represented survivors. Listed and institutionally complex firms are more visible in the top-30 value subset, but this dataset does not show which model has a higher survival rate or what scale requires.

06Three instructive failure or closure cases

Going through all these successful stories made me wonder about the opposite: are there companies that survived for centuries and still failed recently?

I found a few, and they may be more useful than the success stories.

Kongō Gumi: the craft survived, but independence did not

Kongō Gumi traces its operating tradition to 578 and spent roughly 1,400 years specializing in temple construction. That sounds almost invincible. It was not.

Demand for traditional temple construction weakened. At the same time, debt and exposure connected to Japan’s real-estate bubble damaged the company. In 2006, Takamatsu Construction acquired the operating business. The Kongō Gumi name, employees and specialist work continued under a subsidiary, but the ancient independent family enterprise ended.³

The lesson: 1,400 years of expertise could not offset leverage and a shrinking core market. An acquisition can preserve the work while ending one form of continuity.

Whitechapel Bell Foundry: the market became too small

The Whitechapel Bell Foundry traced its history to around 1570. It cast famous bells, including Big Ben and the Liberty Bell. Then it closed in 2017.

A UK government planning inquiry reached a fairly simple conclusion: the market for large tower bells had been shrinking for years, demand was very small, and casting them at a central London site was no longer viable.⁹

The lesson: craftsmanship, fame and 450 years of history cannot cover fixed costs when the addressable market contracts below a viable scale.

R. Durtnell & Sons: 428 years did not remove balance-sheet risk

R. Durtnell & Sons traced its history to 1591. The historic operating contractor ceased trading on 4 July 2019 while involved in major restoration work. Contemporary industry coverage pointed to a persistently falling level of net assets.¹⁰

There is also a technical wrinkle: R. Durtnell & Sons Limited is still marked active at Companies House and has filed accounts since the cessation, while its related holding company has continued filing too.¹⁰ So this is best described as the end of the historic contracting operation and a major break in continuity, not the conclusive dissolution of every related legal entity. Even the phrase “the company closed” can become complicated when an operation, legal company, holding company and historic name are not the same thing.

The lesson: long project cycles may amplify cash-flow pressure. The direct warning here is simpler: centuries of experience did not prevent the balance sheet from deteriorating.

07What makes firms last and what makes them fail?

After going through the data and the failure cases, I do not think there is one secret to lasting for centuries. There are several, and they reinforce each other.

1. Continuity is a deliberate objective

Many long-lived family firms are not optimizing only for this quarter or even this generation. They care about continuity, reputation, family identity and their role in the community. The EU–Japan Centre’s review repeatedly finds family orientation, long-term goals, craft tradition, merit-based succession and social engagement.¹¹

That does not make them anti-profit. It means profit is treated as fuel for continuing the institution rather than the only reason it exists.

2. Succession is flexible, not purely hereditary

Japanese family businesses have historically used adult adoption, sons-in-law and other non-blood heirs to keep the business household intact while choosing capable managers.

A study in the Journal of Financial Economics found that non-blood-heir-run Japanese family firms outperformed blood-heir firms and roughly matched founder-run listed firms.¹² This does not explain Japan’s entire pattern, but it shows how a family business can protect continuity without insisting that management talent must arrive through biology.

3. Tradition becomes a resource for innovation

The businesses that last do not preserve every process exactly as it was. They preserve a story, a capability or a promise, then reinterpret it.

Researchers call this “innovation through tradition”: old knowledge becomes raw material for new products, markets and methods.¹³ Heritage works when it is a platform for renewal. It becomes a liability when management treats it as an instruction never to change.

4. Renewal happens before the crisis

Research on Japanese long-established listed companies suggests that business development before a pivotal strategic change matters.¹⁴ Waiting until the historic core is already collapsing may be too late.

Simple diversification is not enough either. A company can add businesses and still fail if it has not built the skills, capital and management structure needed to make a real transition.

5. The product answers a recurring human need

Food, drink, accommodation, health, tools, finance and ritual goods appear again and again in the dataset. The method of delivery changes, but the underlying need survives.

Beverages, food, manufacturing and hospitality together make up 80.5% of the represented companies. That does not prove these sectors have the highest survival rates, but it is difficult to ignore how often the oldest businesses sell something humans keep needing.

6. Place and community create a moat

Terroir, spring water, pilgrimage routes, temples, local craft networks, protected recipes and civic institutions give many old businesses an identity that cannot simply be copied and moved.

Research on Japanese businesses surviving 300 years or more highlights customers and products, owners and employees, management credo, change and risk management, and the community-oriented idea of sanpo-yoshi, meaning good for the seller, the buyer and society.¹⁵

This is not a moat built from software or patents. It is a moat built from belonging.

7. Conservative finance and reusable assets help

Teikoku Databank finds that old Japanese companies can stabilize themselves through non-operating income and long-held property or financial assets.⁸ A historic business may own the building beneath it, rent unused land, or hold assets accumulated across generations.

That does not mean every old firm is conservative. Kongō Gumi is a warning against saying that. But it does mean some have a financial buffer that a young business living only from current cash flow does not.

Key takeaway

The companies that last are rarely frozen in time. They preserve the part customers trust while changing almost everything else.

08Why are so many represented companies Japanese?

There is no single answer. The safest explanation is layered.

First, the dataset itself is skewed

Wikipedia coverage, historical documentation and editor attention vary by country. Japan and Germany may be cataloged more systematically than large parts of Africa, Asia and Latin America. The list proves what is represented in the list, not a universal national survival rate.

Second, Japan genuinely has an unusually deep population of old firms

Teikoku Databank counted 46,708 Japanese companies at least 100 years old at December 2025. That included 1,836 at least 200 years old, 905 at least 300, 47 at least 500 and 11 at least 1,000.⁸

So the Wikipedia skew is real, but it is not the entire story.

Third, the business-household system supports succession

The Japanese ie tradition treats the household and business lineage as something that can continue beyond any one person. Adult adoption historically allowed families to bring in capable heirs rather than letting the absence of a suitable biological successor end the company.¹²

Fourth, community status reinforces customer loyalty

A recognized shinise is not merely old. It can embody local identity, trusted quality and the continuity of a town or craft. Customers are not buying only the object; they are buying participation in the story.¹⁶

Fifth, Japan’s sector mix is favorable

Sake, confectionery, inns, traditional medicines and craft products combine recurring demand with place-based reputation. In this dataset, all 87 sake companies are Japanese, as are 75 of the 76 confectionery-related businesses.²

Sixth, some regions retained commercial continuity

Teikoku Databank points to Kyoto’s relatively limited Second World War damage as one contributor to its high concentration of long-established firms.⁸ Physical continuity does not explain everything, but it matters when a business depends on a particular street, workshop or customer community.

Finally, Japan is not immune to failure

Ageing owners and absent successors now threaten otherwise viable businesses. A company can have customers, make money and still close because nobody is able or willing to take over. Teikoku Databank recorded 142 bankruptcies of century-old Japanese firms in 2025.⁸

09What commonly ends a centuries-old firm?

The failure cases repeat a familiar set of problems:

• Demand for the historic core product disappears.
• Management protects tradition but fails to adapt the business model.
• Debt or speculative investments overwhelm an otherwise stable operation.
• One project, customer or lender creates a cash-flow shock.
• No capable or willing successor is available.
• Family conflict prevents a timely transfer of control.
• War, policy change, technology or the loss of protected status removes the old advantage.
• The operations survive, but independence disappears through acquisition.

10As close to immortality as a company can get

Humans have not been around for very long on this little blue ball. But within our tiny lives, 1,500 years is an enormous number. Most of us cannot name our ancestors from a few hundred years ago.

The fact that companies have survived, and in some cases thrived, for so long is incredible. They have crossed wars, fires, plagues, political revolutions, new currencies, industrialization, globalization and the internet.

A recurring pattern is that many preserved what customers trusted while changing much of what surrounded it.

This is probably as close to immortality as a company can get.