A total of 801 KOSPI-listed companies were put on the analysis table. On July 23, 2026, the National Assembly Budget Office released "ESG Disclosure, Who Does It and Why: An Analysis of the Determinants of Voluntary Disclosure," comparing how corporate characteristics, capital market pressure and global carbon regulation divide whether a company makes voluntary ESG disclosures. Corporate characteristics included assets, years in business and debt, while the capital market pressure category was measured by National Pension Service investment and the share held by foreign investors. Carbon regulation was divided by whether a company was subject to the emissions trading scheme and the carbon border adjustment mechanism.
What stood out in the results was the capital market side. When the asset size factor rose by 1, the probability of voluntary disclosure was 1,910 times higher, and when the National Pension Service investment factor rose by 1, it was 1,442 times higher, according to the analysis. Looking at the multiples alone, the force of size comes first, but the difference in magnitude between the two factors is not large. Kim Yun-hee, an analyst at the National Assembly Budget Office, assessed that the finding that the probability of disclosure rises as National Pension Service investment grows shows stewardship activity functioning as an effective policy instrument.
These multiples are hard to read at face value. The published material does not state in what unit "a factor of 1" is 1, or from what estimation method those multiples were derived. The direction of movement can be read, but the magnitude is a figure that is hard to carry over as it stands.
Years in business showed a flow in the opposite direction. The voluntary disclosure rate for companies less than 23 years old was 41.5%, while for companies of 64 years or more it stopped at 22.6%. The gap between the two brackets is 18.9 percentage points. The result runs against the common assumption that older companies are larger and hold more assets, and because the material does not present disclosure rates for the middle brackets between the two, the shape of the curve is hard to follow.
The analysis came out just after the regulatory timetable was finalized. On July 8, 2026, the government announced the final plan for institutionalizing sustainability disclosure. It calls for making sustainability disclosure a statutory disclosure requirement for KOSPI-listed companies with consolidated assets of 10 trillion won or more, starting with fiscal year 2027, that is, from disclosures made in 2028.
Compared with the February 2026 consultation draft, the consolidated asset threshold came down from 30 trillion won to 10 trillion won, and the two-stage approach of moving to statutory disclosure by way of exchange disclosure was changed to statutory disclosure at once. Disclosure of Scope 3, or supply chain emissions, was pushed back to after 2031.
The name of the framework changed as well. "ESG disclosure" became "sustainability disclosure."
For companies outside the scope of the mandate, it remains a voluntary area until 2028. This analysis shows what is drawing out disclosure in that area. Most companies with assets under 10 trillion won bear no statutory obligation, but the situation is different if the National Pension Service holds a substantial stake.
The possibility that asset size and National Pension Service investment are entangled with each other does remain. Pension fund money flowing into large companies is a natural thing.
Figures from Europe, where the mandate is already in force, show a different picture. EFRAG's "2026 State of Play Report" shows that 69% of companies subject to the Corporate Sustainability Reporting Directive (CSRD) disclosed a climate transition plan (on a fiscal year 2025 basis). That is up 14 percentage points from 55% in fiscal year 2024.
On the same fiscal year basis, 57% set decarbonization targets aligned with a 1.5-degree pathway, and 63% tied executive pay to those targets. The survey went through 905 fiscal year 2025 sustainability reports that had undergone third-party assurance, using 18 items from the European Sustainability Reporting Standards (ESRS).
The variation by country is also wide. Spain was highest at 89%, followed by France at 85% and Denmark at 81%. By sector, real estate was highest at 95%, followed by administrative and support services at 82% and banking at 78%.
Sectors that hold assets for long periods and sectors that supply capital are clustered at the top. Companies reporting under the CSRD selected an average of 6.4 of the 10 ESRS topics as material, and "E1 Climate change" and "S1 Own workforce" were the most frequently identified at 99% each, with "G1 Business conduct" at 95%.
The two sets of figures are hard to place side by side directly. The European 69% is the share disclosing climate transition plans within a mandatory disclosure regime, while the domestic 41.5% and 22.6% are shares disclosing voluntarily with no obligation in place. The starting lines differ, so the levels are not something to compare.
What can be compared is what moves companies. In Europe it was CSRD coverage, and in Korea it was the scale of National Pension Service investment that drew out disclosure.
Activity on the practitioner side continues as well. A revised and expanded edition of "Sustainability Management Reporting and Assurance: A Compass for the Era of Mandatory Disclosure" was published on July 13, 2026. The first edition came out in January 2025, so it has been a year and a half. The author is Lee Jong-jae, a sustainability management consultant who serves as vice chairman of the Korea Sustainability Assurance Forum and vice chairman of Etoday.
The first statutory disclosure will be made in 2028. The results it will contain are for fiscal year 2027, so the data collection systems of the companies covered have to be running from the point that fiscal year begins. On what actually divides whether companies below the 10 trillion won asset threshold disclose, this analysis by the National Assembly Budget Office has put forward one answer. The size of the multiples needs further verification, but the direction the arrow points is clear.
