Decoding RaymondsIndex: Deterioration Risk (WP), Explained

1. The event

On 16 September 2026, at its 16th regular meeting, Korea's Securities and Futures Commission — an arm of the Financial Services Commission — imposed sanctions on Haitai Confectionery and Foods. The findings: revenue and cost of sales were overstated from 2016 through 2019, by 13.678bn won, 15.317bn won, 11.256bn won and 12.777bn won respectively, through false tax invoices and the omission of sales discounts and sales incentives. The commission also found violations in securities registration statements and obstruction of the external audit.

The obstruction is the detail that matters here. Haitai presented false transaction records to its auditor, and asked business partners to send false replies to accounts-receivable confirmation requests. Sanctions include a three-year auditor designation, a dismissal recommendation and six-month suspension for the head of the finance division, an action equivalent to a dismissal recommendation against the former auditor, and fines — with the finance head and former auditor referred to prosecutors. Fine amounts are to be finalised by the Financial Services Commission.

2. What WP actually measures

Deterioration risk is routinely read as "how bad is it." That is not what the metric is asking. WP is built around a gap: the distance between the moment a fact becomes true and the moment the record says so.

Haitai shows what governs the length of that gap. A misstatement begins as a journal entry inside the company. Keeping it in place for four years did not. An accounts-receivable confirmation request is the single procedure designed to verify a company's books by asking someone other than the company. To neutralise it, the party at the other end has to answer a certain way. At that point the matter stops being an accounting event and becomes a relational one — and the number of people required to sustain it is what sets how long the record stays silent.

The inverse case appeared two days later. A KOSDAQ-listed entertainment company disclosed that an unregistered executive is suspected of breach of trust, stating it would determine the specifics and the scale of the loss through an audit and outside legal counsel before pursuing recovery. The event is in the record; the amount is not in any financial statement line yet. Screened on ratios, that company looks untouched today.

3. Korea parallel

RaymondsIndex is built on 3,109 KOSPI and KOSDAQ listed companies. Within that population, 276 companies were subsequently suspended from trading. Of those 276, 85.9% exhibited a statistically significant relational signal — a change in board composition, an executive departure, or a concentration of privately placed convertible bonds — before deterioration became visible in financial indicators. Effect sizes exceeded Cohen's d > 0.8 across all three signal dimensions. The point of those figures is not that relational data predicts trouble in the abstract. It is that the benchmark being beaten is a financial-statement benchmark, and the margin is a timing margin.

Hong Kong has written a version of the same logic into rule. HKEX now requires auditors to be appointed or removed only at general meetings, requires specific audit fees or ranges in disclosures, and treats any company action causing an auditor's resignation as a removal requiring a shareholder vote. Behind it: the Securities and Futures Commission has warned that late-stage resignations are warning signs of governance and internal control problems, noting in a review that auditors at 89 listed issuers resigned within four months of annual reporting deadlines, with 66 of those departures linked to "fee disagreements." Regulators there maintain that outright financial fraud is not widespread in Hong Kong.

4. Academic frame

Khanna, Kim and Lu (Journal of Finance, 2015, 70(3), 1203–1252) find that appointment-based CEO connectedness in executive suites and boardrooms both raises the likelihood of fraud and lowers the likelihood of detection, partly by reducing the coordination costs of carrying it out. Detection probability, in other words, is a function of the network — which is precisely the quantity a ratio cannot hold. Dechow, Ge, Larson and Sloan (Contemporary Accounting Research, 2011, 28(1), 17–82), working from 2,190 SEC Accounting and Auditing Enforcement Releases issued between 1982 and 2005, examine misstating firms along five dimensions, one of which is explicitly non-financial measures — an early acknowledgement that the accrual data alone was not carrying the signal.

5. What this means for an individual investor

Two records arrived this week. One attached numbers to years that closed in 2019. The other named an event and said the amount is still unknown. In both cases the financial statements were the last document to speak, not the first. The practical question is not which ratio to watch more closely — it is which documents you are reading at all, and in what order.

Facts above are drawn from public regulatory rulings, disclosures and reporting; general observation, not investment advice.

#RaymondsRisk #RelationalRisk #CorporateGovernance #CounterpartyRisk #DetectionLag #AuditQuality