What Individual Investors Don't See Until It's Too Late: The Record That Was Never Written

On September 2, 2026, at its fifteenth regular meeting, Korea's Securities and Futures Commission referred an investor-relations officer at a KOSDAQ-listed company to prosecutors on suspicion of trading on material non-public information. The facts, as the regulator described them, are compact. The officer, working in IR, obtained the topline result of a Phase 1 clinical trial and news of a signed drug-licensing agreement before either became public. Between March and June 2024, shares in the company were bought through an account held in another person's name. The gain was approximately 20 million won. The officer also failed to file the ownership report that Korean law requires of insiders.

Under the Financial Investment Services and Capital Markets Act, an insider who uses material non-public information in trading faces imprisonment of at least one year, or a fine of up to six times the unlawful gain, with an administrative penalty of up to twice the gain also available. Separately, officers and major shareholders must report their holdings to the Financial Supervisory Service within five days of becoming an insider, and any subsequent change within five days of the change. Failing to report carries up to one year's imprisonment or a fine of up to 30 million won. The authorities did not disclose the individual or the company, citing the effect on the pending investigation.

The concept: alignment is a comparison, not a quantity

RaymondsIndex treats Momentum Alignment as one of its leading indicators, and it is routinely described in shorthand as the fit between revenue growth and capital-expenditure growth. That shorthand describes an output. The method underneath is different and more general: alignment testing places two independently produced records of the same underlying event next to each other and asks whether they agree. Revenue is recorded by one process, capital expenditure by another. Neither can validate itself. The comparison is what makes it a test.

This has an uncomfortable corollary. An alignment test is only as strong as the weaker of its two inputs, and it fails silently — not loudly — when one input is missing. A missing series does not register as a bad score. It registers as nothing at all.

That is precisely the shape of this case. During March to June 2024, the market held exactly one series about this company's insider activity: price and volume. The second series, the insider's ownership filing, was never produced, because the purchases were routed through a nominee account. The trades were therefore not an outlier during the months they occurred. An outlier requires a benchmark, and the benchmark had not been written.

Korea parallel: who is obliged to create the second record

The Korean disclosure architecture assumes the second record will exist. The five-day filing rule is the mechanism that is supposed to produce it, and it is fast — five days is a tight window by any international standard. But speed only helps a record that gets written. The cost of not writing it, in this case, was the cost of using someone else's account name.

What follows is a question of who can restore a record that was never made. Not the market: an ordinary investor cannot trace account ownership. Only a regulator with the power to follow accounts can reconstruct the second series after the fact, and that reconstruction is expensive and slow. From March 2024, the first month of the trading window, to the September 2026 referral is thirty months — a figure obtained simply by subtracting the two dates the regulator stated, not one the regulator published.

And the reconstruction is still not shared. Because the criminal referral is pending, neither the officer nor the issuer has been named. Investors in that company today know that a violation was found somewhere in the KOSDAQ market. They do not know whether it was theirs. The information asymmetry that opened in 2024 has not been closed; it has been transferred from an insider to an investigator.

This is not a uniquely Korean design problem. In a March 10, 2026 note for Nomura Research Institute's lakyara series, Sadakazu Osaki set out the amendments to Japan's Financial Instruments and Exchange Act expected in 2026 on the basis of Financial System Council working-group reports — among them a broadening of the scope of insider-trading regulation and a streamlining of information-disclosure requirements. The direction of travel in both markets is toward widening the set of events that must be recorded. That is the right instinct, and it addresses the coverage of the rule rather than the ease of stepping outside the name it applies to.

Academic frame

Two papers frame why this matters more than the sum involved. Akerlof (1970), in "The Market for 'Lemons': Quality Uncertainty and the Market Mechanism," showed that when buyers cannot distinguish quality, the price they are willing to pay converges toward the average — and the good sellers withdraw. Undetected insider activity works on the same channel: it does not merely take money from a counterparty, it degrades what every price in that market is understood to mean.

Grossman and Stiglitz (1980), in "On the Impossibility of Informationally Efficient Markets," made the complementary point that information is costly, so prices cannot fully reflect it — someone must be paid for gathering it. This case shows what happens when the gathering cost is paid entirely by the regulator, after the fact, and the result is then withheld while an investigation proceeds. The information gets produced. It simply does not reach the people whose money was exposed to its absence.

For the individual investor

The practical takeaway is not "watch for insider filings." It is narrower and more useful: when a signal you rely on is a comparison, ask what happens if one side of the comparison is simply not filed. Concealment rarely produces a strange number. It produces a blank where the second number should have been, and blanks do not trigger alerts.

So the question worth carrying into next week: in the signals you actually use, how many of them would still fire if the counterparty record were quietly never created?

General observation based on a public regulatory announcement. No company is identified, and nothing here is investment advice.

#RaymondsRisk #RelationalRisk #CorporateGovernance #BeneficialOwnership #EnforcementLag #MarketIntegrity