The Zombie Pattern: When a Ratio Improves Because the Denominator Was Erased

Korea's junior exchange spent July doing something that looks, on a screener, like repair.

As of July 22, KOSDAQ companies had filed 72 capital-reduction decisions this year — more than the 62 filed in all of last year, with five months still to run. The concentration is in no-consideration reductions, where shareholders receive nothing and paid-in capital is written down against accumulated deficit. EV Advanced Materials filed a 90% reduction on July 13, cutting paid-in capital from ₩29.79bn to ₩2.98bn; the stated purpose was "improving financial structure by offsetting accumulated deficit." KS Industry filed a 75% reduction on June 24, taking paid-in capital from ₩20.19bn to ₩5.05bn.

Here is what matters about these filings, and it is the part the ratio hides.

A no-consideration reduction is a zero event for resources. No cash arrives. No cash leaves. The asset side of the balance sheet does not move by one won. What changes is a label on the right-hand side: capital account down, deficit offset, impairment ratio improved. The company is not more productive on July 14 than it was on July 12. It is more compliant.

This is the specific question reinvestment intensity is built to ask. A financial ratio can improve along two entirely different paths — the numerator grows because assets or earnings were created, or the denominator shrinks because capital was erased. Both print the same number. Only the first one leaves behind something that can generate a return next year. So the diagnostic is not "has the ratio improved" but "what share of the improvement came from resource formation" — and, quarter over quarter, whether that share is rising or falling.

The calendar explains the clustering. From July, the Korea Exchange tightened listing-maintenance standards: the market-cap floor moved from ₩15bn to ₩20bn, rising again to ₩30bn in January 2027; a sub-₩1,000 share-price condition was added; disclosure-violation demerits tightened from 15 points to 10. The quietest change is the consequential one — full capital impairment is now tested at the half-year mark, not only at fiscal year-end. The judgment date moved forward by six months. August is when the first half-year reports under that rule arrive.

None of this establishes intent, and it should not be read as such. The Korea Capital Market Institute's Lee Hyo-seop put it carefully in the Seoul Economic Daily: it is hard to say definitively, but the phased tightening of exit requirements probably contributed to firms clearing financial burdens preemptively. What can be said without inference is narrower and still useful: adjustments that create no resources are clustering immediately in front of a test date that just moved closer.

Korea Parallel. Across Korean listed names, the recurring pattern is not the reduction itself — it is the sequence. A ratio improves, the improvement is entirely denominator-side, and reinvestment in the following quarters never steps up to fill the space the write-down opened. That sequence appears in filings before it appears in earnings, which is the whole reason leading indicators exist.

Academic frame. Caballero, Hoshi and Kashyap, in "Zombie Lending and Depressed Restructuring in Japan" (American Economic Review, 2008), showed that credit extended to keep impaired borrowers nominally solvent suppressed investment and job creation across whole sectors — the cost fell on healthy competitors, not only on the assisted firms. Banerjee and Hofmann, in "The Rise of Zombie Firms" (BIS Quarterly Review, September 2018), found that firms unable to cover interest expense from operating profit invest less and lose productivity relative to peers. Both point the same way: the cost of a delayed decision does not vanish. It is redistributed.

Japan is paying that redistribution now. Tokyo Shoko Research counted 5,346 corporate failures in the first half of 2026, up 7.1% year on year and the first half-year above 5,000 since 2014. Firms with fewer than ten employees accounted for 4,844 of them — 90%.

Two markets, opposite ends of one question. Korea pulled its test date forward. Japan is settling a bill it deferred. The practical takeaway is identical: when a distressed company's ratios improve, find out which side of the fraction moved before deciding what you are looking at.

Based on public disclosures and reported figures. General structural observation, not investment advice.

#RaymondsRisk #RelationalRisk #CorporateGovernance #CapitalReduction #BalanceSheetOptics #KOSDAQ