Japan has approximately 33,000 companies that have been operating for more than a century. Of the roughly 5,500 businesses worldwide that have survived more than 200 years, more than half are Japanese. The oldest company whose records survive is Kongō Gumi, a construction firm founded in Osaka in 578 CE by a Korean craftsman brought to Japan to build Buddhist temples. It operated for 1,428 years — surviving civil war, famine, invasion, industrialisation, the Meiji restoration, two world wars, and the Hiroshima bombing — before being absorbed into a construction conglomerate in 2006. It did not fail. It was bought.
The question these firms raise is not sentimental. In a period of accelerating disruption — AI restructuring labour markets, tariff shocks reshuffling supply chains, geopolitical fragmentation reordering capital flows — what does it actually take for an institution to endure? The evidence from Japan’s shinise, as these ancient firms are known, is specific and counter-intuitive.
The first structural feature is conservative balance sheet management. Shinise characteristically carry minimal debt. During Japan’s asset bubble of the late 1980s, many ancient firms were pressured by banks to expand and borrow; those that refused, or borrowed modestly, survived the subsequent collapse. Those that borrowed heavily did not. Kongō Gumi itself is a cautionary illustration: it had survived 14 centuries with negligible external debt before taking on significant loans during the bubble years, and it was that debt — not the failure of its core craft — that ultimately required it to seek a larger parent.
The firms that lasted longest are, by the metrics of contemporary finance, boring. They did not disrupt. They did not scale. They did not pivot.
The second feature is domain discipline. Shinise do not diversify opportunistically. A ryokan that has operated for 400 years does not open a technology subsidiary when venture capital becomes cheap. The companies that last tend to be the ones that ask, with each generation, not “what can we enter?” but “what are we actually for?” This is not conservatism for its own sake; it is a structural recognition that genuine expertise requires decades to build, and that it cannot be arbitraged.
The third feature is succession architecture. Many shinise practice what amounts to an apprenticeship model of leadership succession, in which the incoming leader has typically spent 20 to 30 years inside the firm before assuming authority. Some Japanese family firms extend this further: they practise mukoyoshi, the adoption of a talented non-family member as a son-in-law to carry on the business, when the biological heirs are not suited to leadership. The institution is considered more important than the bloodline.
The fourth feature — the one most at odds with current management orthodoxy — is that shinise tend to prioritise long-term relationships over short-term margin. They maintain suppliers, customers, and partners across generations. They will, and do, accept lower short-term returns in exchange for relationship stability. This is not altruism; it is a calculation about what kind of value is durable.
Research by Makoto Kanda at Meiji Gakuin University, and subsequent work by Randel Carlock at INSEAD, suggests that these features are not specific to Japanese culture — they appear in long-lived family firms across Germany, Italy, France, and India. The German Mittelstand, the clusters of multigenerational manufacturing firms that form the backbone of German export capacity, show similar balance sheet conservatism, domain discipline, and succession models.
What this body of evidence contradicts is the dominant contemporary model of business success, which prizes aggressive expansion, external capital deployment, and quarterly margin optimisation. The firms that produce the best returns in the financial press over five-year periods are rarely the firms that survive two centuries.
AI is accelerating the obsolescence of specific task-competencies while leaving intact the demand for organisations that can be trusted over time. The evidence from shinise suggests that trust — built slowly, maintained at cost, never arbitraged — is not a soft feature of business. It is a structural one.
The lesson is uncomfortable for an era that has equated growth with health, speed with competence, and valuation with value. The firms that lasted longest are, by the metrics of contemporary finance, boring. They did not disrupt. They did not scale. They did not pivot. They did the same thing, carefully, for centuries — and they are still here.
