This Week's Risk Radar: The Number That Flips Sign Depending on Where You Stand

On 19 August, a small KOSDAQ manufacturer disclosed that its largest-shareholder block had changed hands. The filing was unusually bare. Purpose of acquisition: not applicable. Source of funds: not applicable. Plans to appoint or remove officers: not applicable. Reason for change: commencement of inheritance following the death of the largest shareholder.

DUOBACK (KOSDAQ 073190), an ergonomic-chair maker founded in 1987, had 44.10% of its shares held by its chief executive and five related parties — 5,278,486 shares, of which 4,351,102 (36.36%) stood in the chief executive's own name. He died on 17 August; two days later the company filed the change; the next morning trading was suspended pending a delisting eligibility review.

Read that sequence carefully. The causal chain is not the obvious one.

What actually halted the stock

The company had lost money at the operating line every year since 2021 — roughly ₩3.2bn, ₩4.2bn, ₩3.9bn, ₩3.4bn and ₩3.0bn across FY2021–FY2025. Its market capitalisation had sat below the ₩15bn continued-listing floor for thirty consecutive sessions, which put it into administrative-issue status in June. Its shares had fallen from an intraday ₩4,105 in February 2025 to ₩656 by mid-July 2026.

None of that halted trading. Five years of it did not.

What halted trading was a rule with two conditions: a change of control, and administrative-issue status at the moment it happens. The rule is indifferent to cause — sale, gift and inheritance count equally. Condition one is common: Korea Exchange recorded seven KOSPI and eight KOSDAQ changes of largest shareholder in the first nineteen days of August, and 39 and 83 respectively from January through July. Condition two is the scarce one.

Why this is a measurement problem, not a forecasting one

Risk work persistently confuses predicting events with measuring exposure to them. A chief executive's death is not forecastable, and no scoring system should claim otherwise. The question a deterioration score answers is different: if any shock arrives, by what factor does this company convert it?

That factor was fully observable before 17 August, and the most instructive input is the one that looks benign in isolation.

A 36.36% stake in a single individual's name is, in a healthy issuer, ordinarily read as a stability signal — a controlling owner aligned with the equity, insulated from short-horizon pressure. Inside administrative-issue status, the identical figure is a single point of failure, because it is precisely the block whose transfer trips the rule. The number does not change. Its sign does. Governance concentration is therefore not a risk variable on its own; it becomes one only when multiplied by the zone the company occupies. This is why screening governance metrics in isolation produces such unstable results.

The academic frame

That unexpected executive deaths move share prices is an old and well-established finding — Johnson, Magee, Nagarajan and Newman documented the reaction in the Journal of Accounting and Economics in 1985, reading it as information about the manager's marginal value. Bennedsen, Nielsen, Pérez-González and Wolfenzon later showed, in the Quarterly Journal of Economics (2007), that succession outcomes in family firms turn on the structure surrounding the transition rather than the transition itself.

Both point the same way. The event carries information; the structure decides the magnitude.

Korea parallel, and one contrast

Korean market structure makes this multiplier unusually live. Founder-controlled issuers dominate the small-cap end of KOSDAQ, control blocks are concentrated, and the continued-listing thresholds tightened this year have widened the population sitting inside the zone where the multiplier applies.

Japan answered the same problem differently. The Tokyo Stock Exchange grants an issuer that breaches continued-listing criteria an improvement period of one year in principle — and then publishes the expiry. As of 1 April 2026, 21 issuers were under supervision, one was designated for delisting, and 64 improvement periods were still running with their end months mapped through March 2027. TSE went further, advertising in the Nikkei and asking brokers to notify retail clients.

Korea publishes the status. Japan publishes the clock. Neither publishes the multiplier — and the multiplier is what turned an unforecastable Monday into a suspended stock by Thursday.

So the useful question is not which of your holdings might receive bad news. It is which of them would convert ordinary news into something structural — and whether you know that before the news arrives.

Observation drawn from public filings. Not investment advice, and not a view on any issuer's merits.

#RaymondsRisk #RelationalRisk #CorporateGovernance #SuccessionRisk #KeyManRisk #KOSDAQ