South Korea inverts the usual shape of the beneficial-ownership problem. Most countries keep a register they won't show you; Korea keeps no UBO register at all — and yet, once a year, the state publicly names the individual who controls each of its largest business groups, and publishes the ownership maps to prove it. Above a certain size, Korean ownership is the most exhaustively charted on earth: the chaebol disclosures, DART's filing archive and the 5% rule leave the listed tier nowhere to hide. Below that line, the lights go out — the court registry doesn't even show shareholders, and the beneficial-ownership question lives inside bank due-diligence files built on a statutory cascade. Verifying a Korean UBO means knowing which side of that line your counterparty sits on, and working each side with completely different tools.
01The Korean inversion
There is no beneficial-ownership register in South Korea — not public, not closed, not pending. What exists instead is a division of labour among three systems. The Korea Fair Trade Commission maps and publishes the ownership of the conglomerates. The financial disclosure system — DART, run by the Financial Supervisory Service — forces the listed tier into continuous public disclosure. And the AML system under the Financial Transaction Reports Act pushes the beneficial-ownership question onto banks, one customer file at a time, with a statutory cascade telling them who to record when no one crosses the threshold.
The result is a jurisdiction where the hardest question of global compliance — who really controls this company — has an official, published answer for the top of the economy, and no searchable answer at all for the rest.
02The chaebol maps: the state names the controller
Korea's signature institution has no parallel in this series. Under the Monopoly Regulation and Fair Trade Act, the KFTC designates the country's large business groups each year — those with assets of KRW 5 trillion or more — and for each group formally identifies the dongil-in, the "same person": the individual (or, in some cases, entity) deemed to control the group. The designation is published, argued over in the press, and legally consequential — it fixes the perimeter of the group, whose affiliates then owe intra-group shareholding disclosure, cross-shareholding restrictions and related-party rules.
Think about what that means for a verifier. For the conglomerates that dominate Korean commerce, the controlling individual is not something you infer — it is something the state has ruled, annually, with published group ownership charts behind it. The dongil-in system exists because of Korea's defining ownership pathology: founding families controlling vast groups through small direct stakes, circular shareholdings and affiliate loops. The KFTC's answer was not a register but a designation — name the controller, map the loops, and regulate the group as a unit.
03The FATF picture
Korea has never been grey-listed. Its FATF/APG mutual evaluation, published in 2020, credited strong inter-agency cooperation, effective use of financial intelligence and a strong compliance culture in financial institutions — and flagged two gaps that matter here: Korea could make more use of international cooperation to seek and provide beneficial ownership information, and its non-financial gatekeepers — lawyers, accountants, real-estate agents, dealers in precious metals and stones — sat outside the AML regime entirely. The FATF has since recorded progress on the technical deficiencies through follow-up reports, but the structural point stands: in Korea, the beneficial-ownership burden is carried almost entirely by the financial sector.
04The 25% cascade: who banks must record
The operative beneficial-ownership rule lives in the Financial Transaction Reports Act (the FTRA — the Act on Reporting and Using Specified Financial Transaction Information) and its Enforcement Decree, administered by the KoFIU. Financial institutions performing customer due diligence on a corporate customer must identify the beneficial owner through a statutory cascade.
Around the cascade sits the rest of the CDD machinery: due diligence at account opening and for occasional transactions of KRW 10 million or more, enhanced measures for high-risk customers, and — since the FTRA's 2021 extension — the same obligations applied to virtual-asset service providers.
05The real-name system and the borrowed name
Beneath the AML rules lies an older, more distinctively Korean layer: the Real Name Financial Transactions Act, the 1993 reform that outlawed anonymous and pseudonymous bank accounts overnight and required every financial transaction to run under the holder's verified legal name. It is the reason Korean financial rails are identity-rich by default — and the reason Korea's classic concealment device is not the offshore shell but the chamyeong, the borrowed name: assets held in the name of a relative, employee or associate. The pattern has surfaced repeatedly in high-profile chaebol investigations, and it is what a Korean-file verifier should actually be probing for: not a hidden register entry, but a registered name that belongs to someone else.
06DART and the 5% rule: the glass tier
For listed companies, Korea is among the most transparent markets in the world. DART — the Financial Supervisory Service's electronic disclosure archive — carries every listed company's periodic reports, free and in full, including disclosure of the largest shareholder and their affiliates and officers' holdings. On top sits the 5% rule under the Financial Investment Services and Capital Markets Act: any person whose holding in a listed company reaches 5% must publicly report it, and material changes thereafter — the standard tool for watching stakes build in KOSPI and KOSDAQ names.
| Tier | Visibility | Where |
|---|---|---|
| Designated groups (≥ KRW 5tn) | Named controller | KFTC designation, published group ownership structures |
| Listed companies | High | DART reports, largest-shareholder disclosure, the 5% rule |
| Private companies | Dark | Court registry shows officers and capital — not shareholders; ownership sits in corporate books, tax filings and bank CDD |
07The corporate registry: officers, not owners
Korea's corporate register is run by the courts — extracts are available online for a small fee and are the standard proof of a company's existence, showing its name, registration number, address, capital, purpose and registered officers. What the extract does not show is shareholders: the shareholder register is a corporate book kept by the company itself. The state is not blind — companies file statements of share movements with their corporate tax returns, so the National Tax Service holds shareholder detail — but that is a tax record, not a public one. The public paper trail on a private Korean company therefore establishes existence, management and capital, and stops precisely where the ownership question begins.
Infrastructure reinforces the same boundary. Since the Electronic Securities Act took effect in 2019, listed shares exist only as electronic entries in the book-entry system operated through the Korea Securities Depository — identity-tied accounts, transfers that leave records, no certificates to hand over. Unlisted companies, by contrast, may still run on paper share certificates and physical shareholder books. The dematerialised tier is precisely the visible one; paper survives exactly where the public record is weakest — which is one more reason the borrowed-name probe belongs in every private-company file.
08Where to look: sources, access and cost
Korean verification runs on a small set of official sources, most of them free — the constraint is language and coverage, not price.
| Source | What it gives you | Access & cost |
|---|---|---|
| DART (dart.fss.or.kr) | Listed companies: business reports, largest shareholder & affiliates, officers' holdings, 5% filings | Free — full archive, no account needed; mostly Korean, with partial English |
| KRX market disclosure | Exchange filings and event disclosures for KOSPI/KOSDAQ names | Free |
| KFTC large-group disclosure | Designated groups: the named controller and published intra-group ownership structures | Free — published annually with the designations |
| Court registry (IROS, iros.go.kr) | Any company: existence, registration number, address, capital, officers, representative | Nominal fee — roughly KRW 1,000 per extract; Korean-language; no shareholders |
| Company's own books | Private companies: the shareholder register and share-movement history | Via counterparty — request and reconcile against financial statements |
| NTS tax filings | Share-movement statements filed with corporate tax returns | Not accessible — a state record only |
| Bank CDD files | The cascade-derived beneficial-owner record | Not accessible — the institution's file, not yours |
Two Koreas, one workflow
Korean verification splits in two: a glass tier where the challenge is navigating vast Korean-language disclosure, and a dark tier where the record must be assembled from registry extracts, filings and the chain above. Zavia.ai connects directly to official registries in 100+ countries, retrieves and structures the Korean corporate layer, follows ownership through the holding and offshore layers where private chains resolve, and returns an auditable map to the natural person — with sanctions and PEP overlays, in English.
09Trusts and funds: the sintak layer
Korea has a substantial trust industry — the sintak — and it sits squarely in the verifier's path. Trust business is a licensed activity conducted by banks and securities firms under Korea's trust legislation, and the structure a compliance team most often meets is the real-estate trust: title to property is transferred to a licensed trustee company, while the client holds beneficiary rights. On the land register, the trustee appears as owner — so a Korean property-holding chain can dead-end at a trust company unless you ask the beneficial question: who holds the beneficiary certificates, and who was the settlor. The trustee is a regulated institution and its records are the route, through the counterparty.
The fund layer runs in parallel under the Capital Markets Act: collective investment vehicles managed by licensed asset managers, with investor identities sitting in distributor and depository records rather than any public register. The Korean pattern matches the corpus-wide one — regulated intermediaries substitute for ownership registers — with the usual consequence: for a fund or trust in the chain, the answer exists in a licensee's books, and the request routes through the customer to the institution. Where the vehicle or its manager touches the listed market, DART disclosure picks the thread back up.
10The entity landscape
| Vehicle | Form | Where ownership sits |
|---|---|---|
| Stock company | Jusik hoesa | The dominant form; shareholder register kept by the company, share movements filed with tax returns |
| Limited company | Yuhan hoesa | Members' equity in corporate records — long favoured by foreign subsidiaries for lighter disclosure |
| Listed company | KOSPI / KOSDAQ | DART reports, largest-shareholder and affiliate disclosure, the 5% rule |
| Designated-group affiliate | Chaebol member | Inside the KFTC's published group structure, under a named controller |
| Foreign-invested company | FIPA-registered | Foreign-investment notification on record with the authorities — a state trace of the inbound owner |
Foreign-owned companies leave one extra trace: inbound investment is notified under the Foreign Investment Promotion Act, creating a state record of the foreign investor behind a Korean subsidiary. Like Korea's other ownership records it is not a public search — but it means a foreign-invested company's declared parent is on file with the authorities, and the registration certificate is a document the counterparty can be asked to produce.
11Chains and loops: Korea's ownership geometry
That geometry is now under sustained official pressure. The Corporate Value-Up programme launched in 2024 pushes listed companies to address the governance discount, and the July 2025 amendment to the Commercial Act extended directors' fiduciary duty to run to shareholders — a change aimed directly at the controller-versus-minority conflicts that Korea's loop structures produce. Neither creates a UBO register, but both raise the disclosure temperature around exactly the structures this guide describes, and the direction of travel is toward more visibility at the top, not less. One mechanism to watch inside the loops: treasury shares, which carry no votes while held by the company but regain them when sold — historically a lever for re-arming control by placing blocks with friendly parties.
12A worked example
Take a Korean counterparty of each kind. First, a KOSPI-listed manufacturer: DART gives you the business report with the largest shareholder and affiliates, the 5% filings show every major stake, and if the company belongs to a designated group, the KFTC has already named the controlling individual — your job is corroboration and screening, not discovery.
Second, a private parts supplier — a jusik hoesa with three corporate shareholders. The registry extract proves existence, capital and officers, and nothing more. The shareholder register is the company's book; you request it, and the share-movement history behind it, through the counterparty. The bank that holds its accounts has a cascade file — but that is the bank's record, not yours. If a shareholder is itself a company, you repeat the exercise upward; if the chain exits Korea, you resolve the foreign layer in its own jurisdiction. And throughout, you apply the Korean-specific test: are the named individuals plausibly the real parties, or does the pattern — a young relative, an employee-shareholder, a cluster of small stakes summing to control — suggest borrowed names? In Korea the register is honest about what it is; the question is whether the names in the corporate books are.
13Common failure modes
| The mistake | Why it fails |
|---|---|
| Searching for a Korean UBO register | None exists — the answer lives in KFTC designations, DART filings, corporate books and bank CDD |
| Reading the registry extract as ownership | The court register shows officers and capital, never shareholders |
| Taking the cascade name at face value | A representative recorded at rung three is an executive, not necessarily an owner |
| Ignoring the group context | A "standalone" company may be a designated-group affiliate under a named controller with published loops |
| Equating small stakes with no control | Korean control runs on webs — single-digit family holdings can command entire groups |
| Forgetting the borrowed-name pattern | The chamyeong — assets under a relative's or employee's name — is the classic Korean concealment device |
| Stopping at a trustee on the land register | Real-estate trusts put the licensed trustee on title — the beneficiary certificates hold the real answer |
| Assuming gatekeeper files exist | Lawyers, accountants and real-estate agents sit outside the AML regime — the FATF's standing criticism |
14How to verify a Korean UBO: workflow
- Classify the tier. Designated-group affiliate, listed company, or private entity — the tier determines the entire workflow.
- Pull the registry extract. Establish existence, capital, officers and the representative from the court register.
- Work the public tier hard. For listed names: the DART business report, largest-shareholder and affiliate disclosure, and the 5% filings. For group affiliates: the KFTC designation and published group structure.
- Request the books for private companies. The shareholder register and share-movement history, obtained through the counterparty — and reconciled against its financial statements.
- Apply the cascade consciously. Benchmark any Korean KYC file by asking which rung produced the name: a 25% owner, a largest shareholder, or a mere representative.
- Probe for borrowed names. Test the named individuals against the pattern — age, role, plausibility, clusters of small stakes — rather than accepting the register at face value.
- Resolve foreign layers and screen. Trace non-Korean shareholders in their own jurisdictions, then run sanctions, PEP and adverse-media checks with an audit trail.
15Practical takeaways
| Scenario | What you can rely on | What you must supplement |
|---|---|---|
| Designated-group affiliate | A state-named controller and published group map | Screening, and the family layer's private vehicles |
| Listed company | DART reports and 5% filings | Stakes below 5% and concert-party patterns |
| Private jusik hoesa | Registry extract for existence and officers | The shareholder register itself, via the counterparty |
| Foreign-owned subsidiary | Foreign-investment notification trail | The parent chain in its own jurisdiction |
Korea rewards the verifier who respects the cliff. Above it, use the abundance: the state has already done the hardest work, naming controllers and publishing the loops, and DART hands you a disclosure archive most registries can't match. Below it, change tools entirely: the registry proves existence and nothing more, the ownership record is a corporate book you must request, the KYC name may be a cascade artefact, and the risk worth probing is the borrowed name rather than the hidden shell. One jurisdiction, two regimes of visibility — and the professional error is using the tools of one on the other.
Korea & cross-border ownership data — however you build
Skip the layer-by-layer unwinding. Get Korean corporate data and cross-border ownership resolution sourced directly from official registries — structured, translated and mapped to the natural person at the top.
16Frequently asked questions
Does South Korea have a beneficial ownership register?
No. Korea maintains no UBO register of any kind — public or closed. Beneficial-ownership information exists instead in three places: the KFTC's annual designation of large business groups and their controllers, with published group ownership structures; the disclosure system for listed companies through DART and the 5% rule; and the customer-due-diligence files financial institutions must build under the Financial Transaction Reports Act, using a statutory cascade. For private companies outside those systems, ownership lives in the company's own shareholder register and its tax filings — neither of which is publicly searchable.
What is the beneficial ownership threshold in South Korea?
25%, with a cascade beneath it. Under the Enforcement Decree of the Financial Transaction Reports Act, a financial institution performing due diligence on a corporate customer identifies the natural person owning 25% or more; where no one crosses that line, it records the largest shareholder or the person exercising dominant influence; and where no controller can be identified, the company's representative is recorded. The cascade guarantees a name in every file — but it also means a recorded "beneficial owner" may be an executive rather than an owner, so the rung that produced the name matters.
What is the dongil-in in Korean regulation?
The dongil-in — literally the "same person" — is the individual (or in some cases entity) the Korea Fair Trade Commission formally designates as the controller of a large business group. Each year the KFTC designates groups with assets of KRW 5 trillion or more, names each group's dongil-in, and publishes the group's ownership structure. The designation fixes the group's legal perimeter: its affiliates then owe intra-group disclosure and fall under cross-shareholding and related-party rules. For verifiers it is unique globally — a state-ruled answer to the control question for the top of the economy.
What is DART and what does it disclose?
DART is the Financial Supervisory Service's electronic disclosure archive — Korea's equivalent of EDGAR — carrying listed companies' periodic and event disclosures, free and in full. For ownership purposes the key contents are the business report's disclosure of the largest shareholder and their affiliates, officers' shareholdings, and the flow of large-holding reports under the 5% rule, which requires any person whose stake in a listed company reaches 5% to report it publicly, with updates for material changes. For any listed Korean counterparty, DART is the first stop and often effectively answers the ownership question.
What does the Korean corporate registry show?
Officers, not owners. The corporate register is maintained by the courts, and extracts — available online for a small fee — show a company's name, registration number, address, purpose, capital and registered officers including its representative. Shareholders do not appear: the shareholder register is a book the company itself keeps, and share-movement detail reaches the state through corporate tax filings to the National Tax Service, which are not public. The extract therefore proves existence and management, and the ownership trail must be built from disclosure, corporate books or the counterparty itself.
Is South Korea on the FATF grey list?
No — Korea has never been grey-listed. Its FATF/APG mutual evaluation, published in 2020, found strong inter-agency cooperation, effective use of financial intelligence and a robust compliance culture in the financial sector, while criticising two things relevant to ownership: limited use of international cooperation to seek and share beneficial-ownership information, and the exclusion of non-financial gatekeepers — lawyers, accountants, real-estate agents and dealers in precious metals and stones — from AML obligations. Korea has since recorded progress on technical deficiencies through follow-up reports.
What is the real-name financial system?
The 1993 reform that outlawed anonymous finance in Korea. The Real Name Financial Transactions Act requires every financial transaction to be conducted under the holder's verified legal name, ending the era of anonymous and pseudonymous accounts overnight. Its legacy cuts two ways for verifiers: Korean financial rails are identity-rich by default, but the country's characteristic concealment device became the chamyeong — the borrowed name, with assets registered to relatives, employees or associates — a pattern that has surfaced repeatedly in major chaebol investigations and that Korean-file due diligence should actively probe for.
What is the 5% rule in Korea?
The large-holding reporting rule under the Financial Investment Services and Capital Markets Act: any person whose holding in a listed company reaches 5% of its voting shares must publicly report the position, its purpose and funding, and then report material changes. The filings flow through DART, making stake-building in KOSPI and KOSDAQ companies visible in near-real time. For ownership verification it is the sharpest public tool below the largest-shareholder disclosure — though stakes below 5%, and coordinated holdings structured to stay beneath it, remain the residual blind spot.
Why are chaebol ownership structures hard to analyse?
Because control runs on geometry rather than percentage. Korea's conglomerates were historically built so that founding families command entire groups through small direct stakes multiplied by webs of affiliate cross-holdings and circular shareholdings — loops in which group companies hold each other. A single-digit family stake can translate into absolute control of dozens of companies. This is precisely why the KFTC designation system exists: rather than pretending a threshold test could capture the pattern, the regulator maps the loops, names the controller and regulates the group as a unit — publishing the results.
How do you verify the ownership of a Korean company?
Classify the tier first. For a designated-group affiliate, start from the KFTC's designation and published group structure; for a listed company, pull the DART business report, largest-shareholder disclosure and 5% filings; for a private company, obtain the court-registry extract for existence and officers, then request the shareholder register and share-movement history through the counterparty and reconcile them with its financial statements. Benchmark any bank KYC file against the cascade, probe the named individuals for borrowed-name patterns, resolve foreign shareholders in their own jurisdictions, and screen everyone identified, keeping an auditable trail.
