Author: IY Joseph Cho*
| Jurisdictions: | Topics: |
North Korea is a paradox for the law of enforcement. On paper, it is among the world’s most judgment-laden states: U.S. federal courts have entered judgments exceeding US$2.8 billion under the FSIA terrorism exception, including the US$501 million Warmbier judgment. In practice, it is among the most enforcement-proof: the forfeited cargo vessel Wise Honest remains among the few notable execution successes.
The standard explanation is treaty non-membership – the DPRK is party to neither the New York Convention nor ICSID -plus the absence of attachable assets. A recent article in Pacific Focus accordingly argues that Pyongyang should accede to both conventions. This post argues that the accession framing, while not wrong, misses the binding constraint. The Convention already covers more of the field than the framing implies: a Singapore-, Seoul-, or London-seated award involving a DPRK party is a Convention award, because Convention status turns on the place of the award, not the nationality of a sanctioned party. The real wall stands elsewhere. Even a procedurally flawless, Convention-covered, recognizable award may be impossible to monetize, because satisfying it can itself constitute a prohibited transfer of value under sanctions law.
Four stages, not one
Sanctions discussions habitually compress four legally distinct stages into the single word “enforcement.” Disaggregating them is where the analysis starts.
Recognition – the forum’s acceptance of the award’s existence and preclusive force – transfers no value: a court may recognize an award in favor of a designated entity, thereby fixing the debt, interrupting limitation periods, supporting an insolvency proof, without moving a cent. Exequatur converts the award into a domestic judgment and likewise moves nothing. Execution – attachment, garnishment, forced sale – is where asset-freeze rules engage. Payment is where the problem is sharpest: transferring funds to a designated or embargo-covered party is the paradigm prohibited dealing.
The distinction is not hypothetical. In the Russia-sanctions context, German appellate courts have divided between recognition-without-enforcement and outright refusal, prompting recent commentary that declaratory recognition preserves the award’s legal force while blocked or trust accounts supply the compliant terminus. U.S. practice draws the same line: under OFAC FAQ 808, no license is needed merely to litigate against a blocked person, but one is required for settlement or for execution altering blocked property.
Why the DPRK is the clean case
For Russia and Iran, the hard question is whether an enforcement forum should give effect to another state’s unilateral or secondary sanctions – the terrain mapped by the post-2022 literature and by recent scholarship on Article V(2)(b) and sanctions. The DPRK is different. Its core sanctions layer is multilateral: Security Council resolutions 1718 (2006)through 2397 (2017), adopted under Chapter VII and binding on every UN member state under Articles 25 and 103 of the Charter. Resolution 2375 (2017) prohibited all joint ventures with DPRK entities and required existing ones closed within 120 days – meaning the contracts most likely to contain arbitration clauses were precisely the contracts states were obliged to terminate.
Because every enforcement forum sits under the identical Chapter VII obligation, the DPRK measures function as genuinely transnational public policy under Article V(2)(b): the rare case in which the content of the exception does not vary with the forum.. This narrows forum shopping rather than eliminating it: no compliant forum could permit unrestricted payment to a UN-designated or otherwise sanctions-covered DPRK creditor, but domestic implementing regimes still differ on stays, frozen-account crediting, licensing routes, and the treatment of particular asset classes – variation that sits at the remedial layer rather than at the level of the prohibition. North Korea is unusually clean at the level of source – one of the rare sanctions subjects for which the public-policy analysis can proceed on the multilateral layer alone.
Clean at the source does not mean binary at the remedy. Combining the Chapter VII analysis with the four-stage taxonomy yields a graduated menu rather than an on/off switch: (i) refuse enforcement outright where the relief sought would directly effect a prohibited transfer; (ii) recognize but stay execution; (iii) recognize and permit execution only into a blocked account; (iv) condition execution on licensing and exemption efforts; (v) deny release while preserving the claim, with interest continuing to accrue. Sanctions, in other words, do not necessarily destroy the award; more often they suspend its monetization. Note the fine print: Resolution 1718’s ¶9(c) carve-out reaches only liens and judgments entered before the resolution and not benefiting designated persons – near-useless for any new award in a DPRK creditor’s favor – and the DPRK resolutions, unlike a number of other UN asset-freeze regimes, contain no express provision for crediting interest or pre-existing obligations to frozen accounts, leaving that mechanics to national implementing law. Any plausible lawful exit would have to be built around crediting into frozen accounts and the 1718 Committee’s exemption machinery. That is a design brief, not a dead end.
The Korean jurisprudence: paying without releasing
What makes this more than theory is that Korean courts have now litigated the underlying question – can sanctions exposure excuse an ordinary contractual payment obligation? – to the Supreme Court and back.
In Bank Mellat v. Woori Bank, the Iranian bank (a U.S. SDN) sued for the return of roughly KRW 20.2 billion frozen at Woori since the 2018 re-imposition of Iran sanctions. It lost at every level, with the Supreme Court dismissing the final appeal in April 2025. On the reported reasoning, the good-faith principle of Civil Code Article 2(1) limits even a bare contractual repayment claim; Mellat’s SDN designation was a grave supervening change within its own sphere; and compliance would have exposed Woori to an existential secondary-sanctions risk, rendering performance – in the courts’ striking phrase – normatively impossible. The High Court added that contractual interest had been credited to the frozen account throughout, defeating the damages claim.
Then the counterpoint. In a parallel action over the Bank of Korea’s 2019 suspension of Mellat’s access to its deposit facility, the same court ruled for Mellat in May 2026, awarding a KRW 10 billion partial claim. Two grounds distinguished rather than contradicted Woori: the sanctions risk to the central bank had not become concrete or realized, and the BOK had contemporaneously justified the suspension on a different ground, raising sanctions only in litigation. The BOK has appealed.
Read together, the cases mark an emerging, fact-sensitive doctrine: sanctions exposure can render performance of payment obligations normatively impossible, but only where the compliance risk is grave, real, and concretized rather than speculative; where the sanctions rationale is genuine and contemporaneous rather than post hoc; and with attention to whose sphere produced the supervening change. Notice also the interest symmetry – Woori escaped damages because it credited interest throughout the freeze; the BOK incurred damages because its suspension cut the facility off entirely. Crediting without releasing is what shields the compliant obligor. Translated into award mechanics: paying into a blocked account may be both the ceiling of what sanctions law permits and the floor of what good faith requires.
The extrapolation to a DPRK-linked award is short. Every element of the doctrine appears in stronger form: the prohibition is not a foreign state’s policy whose enforcement risk needs factual proof, but a Chapter VII obligation of Korea itself, implemented in domestic law with criminal consequences; and the post hoc problem is materially reduced where the prohibition is contemporaneous, public, and statutory.
One intermediate step deserves note: Korean procedure separates the enforcement decision from execution proper, and where execution then proceeds by attachment and collection against a bank as third-party obligor, the bank stands in a position functionally analogous to Woori’s – the doctrine enters the execution phase principally as the garnishee’s defence, while at the payment and release stage the implementing statutes govern directly. Mellat’s contribution to award mechanics is thus the preservation logic: debt not extinguished, interest credited where lawful, release deferred.
Two cautionary artifacts
Korean peninsula practice supplies two demonstrations. The South–North Commercial Arbitration Committee — fully drafted in the inter-Korean agreements and revived for Kaesong with arbitrator lists exchanged in 2014 – never heard a case; an institutional framework existed on paper but never matured into an operable adjudicatory mechanism; and even had it functioned, any award in either direction after 2016–17 would likely have required a licensed payment channel that did not exist. A forum existed; a payment channel did not. KEDO tells the same story earlier: its light-water reactor Supply Agreement appears to have provided for arbitration only with the other side’s consent – an agreement to agree, not a mechanism against a non-cooperating DPRK counterparty. When the project collapsed, mutual claims dissolved into diplomacy and write-off, not adjudication. Neither mechanism ever produced a constituted tribunal and award, so the payment wall was never actually tested – which is itself the design lesson: unilateral commencement and an external appointing authority are load-bearing for the existence of an award; blocked-payment and licensing architecture, for its realization.
Designing for the wall
If commerce with the DPRK ever again becomes lawful, the architecture should be built around sanctions contingency from day one: a unilateral right to commence arbitration plus an external appointing authority (the two elements the historical instruments lacked); a neutral Convention seat with institutional administration of licensed fund flows, as institutions already manage in Russia-related cases; and – the core – blocked-account performance as the contractual terminus, where the applicable implementing regime permits the credit with interest credited Mellat-style, release obligations suspended rather than extinguished, and the clause specifying whether a lawful credit operates as discharge, provisional satisfaction, or mere preservation of the payment obligation, paired with a separate cooperation covenant: calibrated efforts to pursue exemptions and licenses, warranting neither their grant nor eventual release. The clause allocates sanctions risk rather than eliminating it – and that allocation, made expressly by the parties rather than judicially implied, is the point. Contract drafters need not wait: calibrated sanctions triggers (the gulf between Lamesa and Mamancochet), payment cascades that MUR Shipping v RTI confirms tribunals will not judicially patch, and blocked-account fallbacks belong in any DPRK-adjacent contract now.
One closing inversion. The familiar complaint is that creditors cannot collect from North Korea. The quieter fact is that few parties can safely pay, or receive payment under, an award involving a designated or otherwise sanctions-covered DPRK creditor absent a licensed channel meaning sanctions have largely disabled formal dispute resolution for precisely the residual commerce they tolerate. A narrow, monitored, blocked-account award-payment channel would surrender nothing of coercive substance while restoring a rule-of-law path that currently does not exist. That is a design choice, not a legal impossibility.
* IY Joseph Cho is based in Seoul, where he handles commercial litigation, international arbitration, and regulatory matters in Korea and internationally. He is admitted to practice in New York and as a solicitor of England and Wales, and holds a J.S.D. from Seoul National University School of Law. His scholarship on international arbitration, sanctions, and Korean commercial law has appeared in the Cornell International Law Journal, the University of Pennsylvania Asian Law Review, and the Australian Journal of Asian Law.