Two years ago we wrote that we could describe the machinery built to close Korea’s valuation gap but could not tell you what the market paid for it. That gap in our own coverage has now closed. Companies filing Value-up plans have overtaken those that did not, 92 trillion won left corporate accounts in one year, and the national pension’s offshore share fell for the first time in two decades.
Last verified: 5 August 2026. Built from the Korea Exchange 2026 Value-up programme white paper, a Korea Capital Market Institute comparison of disclosing and non-disclosing companies, and our own reading of the National Pension Service portfolio page. Pension figures are provisional at end-May 2026 and the Service labels them so. We do not quote the Korea Value-up Index return: the Exchange index page renders through script and would not open for us, and the published return figures do not reconcile arithmetically with a KOSPI comparison.
Editor’s opinion
Measurable at last, and narrower than it looks
Our view is that the reform argument has crossed from assertion into evidence, and that the evidence proves something smaller than its promoters claim. Both halves of that matter.
The strongest single number is a reversal. At the end of 2021 the companies that would later file Value-up plans traded at 1.4 times book, and the companies that never filed traded at 1.7. By the first quarter of 2026 the filers stood at 1.9 and the non-filers at 1.5. The cohort that started cheaper finished dearer.
That kills the obvious objection. If good companies had simply selected themselves into the programme, they would have looked better at the start. They looked worse. Something happened between those two dates that attached to the act of disclosing.
Here is the limit. What this measures is one group of Korean companies against another, not Korea against the world. A relative re-rating inside a market is real and it is not the same claim as the discount closing. We think the honest position is that the first has happened and the second remains open.
What argues the other way
Three things. Deciding to file may itself signal a management already changing, in which case the multiple followed the managers and not the policy — a defensible reading, though it still credits the programme with sorting them into view. Five years is a short window, and one of those years carried a market-wide rally that lifted both cohorts. And the funds tracking this theme measure the performance of selected constituents, which is a narrower fact than a market re-rating and is sometimes reported as though it were the same thing.
Andrew Lee. This section argues a view rather than establishing a fact. It is not a forecast and not a recommendation, and the material it draws on is listed in the sources below.
What we could not measure last time
When we set out the case for Korean equities we stopped at a wall and said so. The note read that we could describe the machinery and measure the earnings, “but we cannot tell you what multiple the market currently pays for them”. The argument table listed the same shortfall against us: no valuation series we could stand behind.
What has arrived since is not a national valuation series. It is something more useful for this purpose — a comparison of Korean companies against other Korean companies, split by whether they filed a Value-up plan.
That design matters, and what the Korea discount actually measures explains why. Comparing Korea to the United States mixes the discount together with industrial composition, and nobody can separate the two from published data. Comparing Korean manufacturers to Korean manufacturers holds composition roughly still. Whatever moves after that has a chance of meaning something.
The disclosure cohort overtook the rest
Participation is no longer marginal. Cumulative Value-up disclosures now come from 733 listed companies, together more than 80 per cent of total market capitalisation. Whatever the programme is, it is not a pilot.
Price-to-book, disclosing versus non-disclosing companies
Multiples of book value: 1.4 and 1.9 for filers, 1.7 and 1.5 for non-filers. Bar length tracks the multiple. The two groups changed places. Source: Korea Capital Market Institute, comparison of listed-company performance by Value-up disclosure participation.
Read the crossing rather than the levels. A cohort that traded at a discount to its peers in 2021 traded at a premium to them by 2026, while the peers went the other way. Self-selection by quality cannot produce that shape, because quality does not arrive backwards.
What the shape cannot tell us is how much of the move belongs to disclosure and how much to the reforms running alongside it. The 2021 starting point also sits near a market peak, which flatters any subsequent comparison for whichever group fell less.
Treasury shares and the cash that actually left
Disclosure is talk. The test that matters for governance is whether money moves, and in 2025 it moved.
| Returns to shareholders, 2025 | Amount | Change on 2024 |
|---|---|---|
| Treasury shares bought and cancelled | 41.5tn won | +27% |
| Cash dividends | 50.9tn won | +11% |
| Total | 92.4tn won | — |
The growth rates carry the signal. Buybacks rose at more than twice the pace of dividends, and that is the change a governance reform should produce. Dividends have long been the Korean habit; retiring shares is the newer behaviour and the one that permanently lifts what remains.
One caution we cannot resolve. The source reports purchases and cancellations as a single figure, so we cannot separate them. The distinction is not cosmetic: a company that buys shares and keeps them in treasury can sell them back into the market later, and only cancellation is irreversible. Treat 41.5 trillion won as an upper bound on permanent return.
Behind the behaviour sits a legal change. From 22 July 2025 an amendment to the Commercial Act extended directors’ fiduciary duty to shareholders, not only to the company as an entity. The payout figures add evidence that it was not merely symbolic.
What the index and the funds do not settle
Money has followed the theme into funds. The thirteen Value-up exchange-traded funds held 4.3 trillion won in net assets at end-May 2026, against their launch in November 2024 — a rise of 787.8 per cent.
We leave the index return out, for two reasons. We could not open the Exchange index page to verify it. The figures in circulation also fail arithmetic: a reported index return sits beside a KOSPI comparison implying the market more than tripled since September 2024. One of those numbers is being described wrongly, and we will not pass on either until we know which.
The deeper problem would remain even with clean figures. An index measures its constituents, and constituents are chosen. Strong index performance tells you the selection worked; it does not tell you the market re-rated. Fund inflows measure demand for the theme, which is a fact about investors rather than about Korean companies.
A selloff is not a re-rating reversed
All of this reads oddly against the screen. Korean equities have fallen hard on fears that Chinese memory supply is closing the gap on the domestic champions, which we examined in how far CXMT has actually come and again in the case for buying into the selloff.
Two different clocks are running. The cohort reversal covers five years and a change in what companies are obliged to do with their capital. The selloff covers weeks and a change in the outlook for one industry’s pricing. A price fall does not undo a governance change, and a governance change does not put a floor under a semiconductor cycle.
The honest version of the bearish objection is sharper than the price action. If Korean governance has genuinely improved, why does the whole market still swing on one sector’s margins? Because that is cause number two, industrial composition, and no disclosure rule touches it. Korea makes memory chips. Reform can change how the proceeds are shared and cannot change what generates them.
Testing the point properly would need the two cohorts compared through this drawdown specifically, and we do not have that data.
What the pension shift adds
The largest domestic buyer has turned around. At end-May 2026 the National Pension Service held 543.6 trillion won in Korean equities, 29.4 per cent of the fund. The same series stood at 139.7 trillion at the end of 2024 and 263.7 trillion at the end of 2025.
One line in that table matters more than the holdings. Overseas investment had climbed almost every year since 2001, reaching 59.06 per cent of the fund in 2025. In 2026 it fell to 53.54 per cent. After two decades of sending money abroad, the fund reversed direction.
Two limits on reading it. Rising prices and fresh buying both inflate a holding, and the table cannot separate them. We also skip the target weights from press coverage: the Service’s page does not publish them, and the actual weight already sits well above the reported target.
Where that leaves the question. Two of the government’s three stated causes now show movement you can point to in primary sources, and the third never was going to move. What nobody can yet show is the market-wide re-rating, because the comparison that would prove it is the one the data does not support. Measurable, then, and unfinished.
Sources
We take facts from these and write our own sentences. Where a figure originates in a company filing we say so, and where it originates in a research house estimate we name the house. We link to a document only where we hold the exact address for it; a link to a publisher’s front page would not get you to the source, so where we do not have the direct address we give you enough to find it.
Exchange and research material
- Korea Exchange, 2026 corporate value enhancement programme white paper, published as a PDF on the KIND Value-up reference page. Korean-language document. Source for the 733 cumulative disclosing companies and their share above 80 per cent of market capitalisation, for 2025 returns to shareholders of 92.4 trillion won split into 41.5 trillion of treasury-share purchase and cancellation up 27 per cent and 50.9 trillion of cash dividends up 11 per cent, and for the thirteen Value-up exchange-traded funds at 4.3 trillion won against a November 2024 launch.
- Korea Capital Market Institute, comparison of listed-company performance by Value-up disclosure participation and remaining tasks. Korean-language page. Source for price-to-book of 1.4 rising to 1.9 for disclosing companies and 1.7 falling to 1.5 for non-disclosing companies, between end-2021 and the first quarter of 2026.
- Korea Exchange, Korea Value-up Index page. We could not read the index level or return from this page, which renders through script. We therefore quote neither, and we do not repeat the widely reported cumulative-return comparison against the KOSPI because the two figures do not reconcile.
Pension fund data we read directly
- National Pension Service Investment Management, portfolio by asset class, read 5 August 2026. Source for domestic equities of 543.6 trillion won at 29.4 per cent of the fund at end-May 2026, for the annual series showing 139.7 trillion in 2024 and 263.7 trillion in 2025, and for the overseas investment share falling from 59.06 per cent in 2025 to 53.54 per cent in 2026. The Service marks the monthly figures provisional. The page publishes no target weights.
Corrections
None yet. If you can point to a primary document that contradicts anything above, we will correct it and say what changed.
Disclosure. This article concerns corporate governance policy and the evidence for its effects. It is not investment advice, and it is not a recommendation to buy or sell any security. Naming a company or a fund here is a citation, not a view on it. The section headed Editor’s opinion is comment: it argues a view, and we fence it off from the sourced material for that reason. This site has no commercial relationship with any company, exchange, fund manager or regulator named above. See the Editorial & Sourcing Policy and Disclaimer.