FORENSIC SYSTEM ARCHITECTURE — SERIES: THE TREATY THAT WON'T LET GO — POST 3 OF 7 The Conduit Layer: The Private Court System

FSA Series ► The Treaty That Won't Let Go ① The Rockhopper Moment ② The Source Layer ③ The Conduit Layer ④ The Conversion Layer ⑤ The Shadow Trader Layer ⑥ The Escape ⑦ FSA Synthesis

Forensic System Architecture — Series: The Treaty That Won't Let Go — Post 3 of 7

The Conduit Layer:
The Private Court System

When the EU Court of Justice ruled in 2018 that investor-state arbitration clauses between EU member states violated EU law, every ECT arbitration tribunal that was asked to apply that ruling declined to do so. Not one or two — all of them. With one single exception across dozens of cases, the ECT's private arbitration network looked at the EU's highest court and said no. The European Commission documented this in its own official text: the tribunals had "disregarded the applicable rules of public international law." Post 3 maps the conduit layer that made this possible — three private tribunal networks, a revolving door between counsel and arbitrator, proceedings held in secret, no meaningful appeal, and costs that make losing a lawsuit cheaper than winning one.

By Randy Gipe & Claude  ·  Forensic System Architecture (FSA)  ·  Series: The Treaty That Won't Let Go  ·  2026

Human / AI Collaboration — Research Note

Post 3's primary sources are: European Commission official document COM(2022)523 (eur-lex.europa.eu), which formally states the Commission's position that ECT tribunals have "disregarded" applicable international law; Gibson Dunn client alerts on Achmea and Komstroy (2021, 2022); the CJEU judgments in Achmea (Case C-284/16, 2018) and Komstroy (Case C-741/19, 2021); the Journal of International Economic Law's quantitative study of the arbitration network (Oxford Academic, 2017) documenting the double-hatting phenomenon and the 1% of lawyers dominating half of all investment arbitration cases; the Kluwer Arbitration Blog's case-by-case tracking of post-Achmea tribunal responses; UNCTAD's Investment Dispute Settlement Navigator; and the European Arbitration Review 2026 (Global Arbitration Review). FSA methodology: Randy Gipe. Research synthesis: Randy Gipe & Claude (Anthropic).

I. The Moment the Conduit Disclosed Itself

FSA's conduit layer is where the architecture moves. The source layer builds the instrument. The conduit layer is the mechanism through which that instrument operates — the institutions, personnel networks, and procedural structures that translate the treaty's investment protections into binding legal awards against sovereign states. In the ECT series, the conduit layer is three private arbitration networks, a small elite of lawyer-arbitrators who move between representing claimants and sitting in judgment, proceedings conducted largely in secret, and a jurisdictional independence so complete that even the EU Court of Justice's direct rulings could not penetrate it.

That last property — the conduit's immunity from the EU's highest court — is the post's foundational finding. It is documented not in activist literature, not in academic criticism, but in an official document of the European Commission itself.

Primary Source — Official European Commission Document

EUROPEAN COMMISSION  ·  COM(2022)523  ·  Brussels, September 22, 2022  ·  eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=CELEX:52022DC0523

The Commission's official text, issued in the context of EU withdrawal from the ECT, formally states the following regarding the CJEU's Achmea (2018) and Komstroy (2021) rulings:

"CONSIDERING that arbitral tribunals established on the basis of Article 26 ECT have held in the past and continue to hold, overwhelmingly, that they are not bound by the judgments of the CJEU, and have held, including after the Komstroy judgment, that Article 26 ECT applies to disputes between a Member State and an investor of another Member State..."

"REGRETTING that those arbitral tribunals have thus disregarded the applicable rules of public international law and the clearly expressed intention of the relevant Contracting Parties to the ECT..."

These are the words of the European Commission — the EU's executive body — in an official document formally published in the EU Official Journal. Not an advocacy organization's assessment. Not an academic paper's conclusion. The EU's own executive branch, in binding official text, documented that a private arbitration network had refused to be bound by the EU's highest court, and that this refusal constituted a "disregard" of applicable international law.

FSA Structural Finding: The Commission document is the conduit layer's self-disclosure — the moment the architecture's institutional independence becomes visible in the record of the institutions it refused to recognize. The tribunal network did not argue that it was following EU law. It argued that EU law did not apply to it. For FSA purposes, the conduit layer's most precise architectural property is not that it defies the EU Court of Justice — it is that it has documented grounds for doing so, and has used those grounds consistently, in dozens of cases, over eight years, without exception.

---

II. The Three Networks: ICSID, SCC, UNCITRAL

ECT claims flow through three primary arbitration institutions. FSA maps them not as neutral procedural venues but as architectural components — each with specific properties that shape how capital moves through the conduit.

FSA: Conduit Layer — The Three Private Tribunal Networks

ICSID

International Centre for Settlement of Investment Disputes

Location: World Bank headquarters, Washington D.C.

Governance: Part of the World Bank Group. Administered under the ICSID Convention (1966), ratified by 158 states.

ECT role: Primary venue for large ECT fossil fuel claims. Rockhopper v. Italy, RWE v. Netherlands, Uniper v. Netherlands all filed at ICSID.

Award enforcement: Under the ICSID Convention, member states must enforce awards "as if they were a final judgment of a court in that State." No national court can review the merits. Annulment only available through ICSID's own internal mechanism.

FSA note: ICSID's World Bank affiliation provides institutional legitimacy. Its enforcement mechanism — binding on member states without national court review — is the strongest in international arbitration. The conduit's most powerful enforcement channel runs through a multilateral institution headquartered inside a US government building on Pennsylvania Avenue.

SCC

Stockholm Chamber of Commerce Arbitration Institute

Location: Stockholm, Sweden.

Governance: Swedish private institution. The ECT explicitly names SCC as an alternative arbitration venue under Article 26.

ECT role: Significant ECT caseload, particularly for intra-EU disputes where claimants prefer a non-ICSID seat. Sweden is outside the EU's enforcement pressure on intra-EU awards.

Post-Achmea response: Swedish courts initially enforced intra-EU ECT awards from SCC proceedings. However the Swedish Arbitration Act's "ordre public" provision was eventually used to void one award, creating split outcomes depending on specific case facts.

FSA note: SCC's Stockholm seat outside direct EU enforcement pressure made it the preferred conduit for intra-EU ECT claims where claimants anticipated enforcement resistance from EU courts. The venue selection is itself an architectural optimization — choosing the channel most likely to produce enforceable outputs.

UNCITRAL

United Nations Commission on International Trade Law

Structure: Not an institution but a set of procedural rules. Under UNCITRAL rules, arbitrations are ad hoc — no permanent institution administers them. The parties and arbitrators operate under the rules without institutional oversight.

Transparency: No institutional case registry. No default publication requirement. Awards under UNCITRAL rules may remain entirely confidential unless both parties agree to disclosure.

ECT role: Named as the third arbitration option under ECT Article 26. Used in some of the most opaque ECT proceedings — the full case record may never become public.

FSA note: UNCITRAL proceedings are the conduit's most opaque channel. No registry, no default publication, no institutional case tracking. FSA Axiom V applies: the cases that do not appear in publicly available case records are evidence gaps that are data. The full volume of ECT arbitration is unknown because UNCITRAL proceedings may leave no public trace.

---

III. The Numbers: Scale, Concentration, and Cost

150+

Known ECT investor-state arbitration cases as of 2022, making the ECT the most used multilateral investment treaty in history

Source: UNCTAD Investment Dispute Settlement Navigator

~40%

Share of known ECT cases involving fossil fuel investments — the single largest sector category in ECT arbitration history

Source: IISD Energy Charter Treaty case analysis

$10M+

Average cost to a government of defending a single ECT arbitration case, regardless of outcome — before any award is paid

Source: IISD; Columbia Journal of Transnational Law analysis

FSA Structural Finding — The Cost Architecture

The $10 million average defense cost figure is the conduit layer's most efficient insulation property. A government that receives an ECT threat faces a choice with no neutral option: pay tens of millions to defend the arbitration — regardless of whether it wins — or modify the climate legislation that triggered the threat. For smaller ECT member states, the arithmetic is direct: it is cheaper to weaken the policy than to win the lawsuit.

This cost structure is not a side effect of the conduit's design. It is the conduit's design operating as intended. The regulatory chill documented in Post 2 — France modifying legislation after a single company's threat, New Zealand and Denmark designing weaker phaseout plans to avoid exposure — is the cost architecture producing its designed output before a single case is filed.

The legal costs are further asymmetric in a second dimension: fossil fuel companies can structure ECT claims on a no-win, no-fee basis with specialist litigation funders — exactly as Rockhopper did. The government defending the claim pays its defense costs regardless of outcome. The company bringing the claim may pay nothing if it loses. The conduit's cost architecture, like its procedural architecture, is systematically oriented toward the claimant.

---

IV. The Revolving Door: Double-Hatting and the Elite Network

The conduit layer's personnel architecture has been quantitatively mapped by academic researchers using the full case record of international investment arbitration. A 2017 study published in the Journal of International Economic Law at Oxford Academic identified 2,699 distinct lawyers who had represented parties in international investment arbitrations. The distribution is extreme.

The Network's Concentration — Oxford Academic Quantitative Study, 2017

Only 14% of identified lawyers — 382 individuals — have litigated more than two cases. The top 1% — 25 lawyers — have each litigated more than 13 cases. A member of this top 25 has appeared, on average, in every second international investment arbitration case in the dataset. Half of all investment arbitration was handled by 25 lawyers.

The same study identified what it called "double-hatting" — the practice of lawyers serving simultaneously as counsel in one ECT arbitration and as arbitrators in another. The same lawyer can spend the morning drafting arguments for a fossil fuel company as counsel in one case, and the afternoon writing an award as arbitrator in another case involving the same contested legal principles. The study documented this practice across the network's core group and noted that the "competitive advantage" of ECT specialists includes — in one practitioner's documented statement — what is "not written down."

Philip Sands QC, a leading international lawyer who has litigated major ECT cases, wrote and lectured repeatedly about the legitimacy concerns: "Can that lawyer, while acting as arbitrator, cut herself off entirely from her simultaneous role as counsel? The issue is not whether she thinks it can be done, but whether a reasonable observer would so conclude."

There is no binding rule prohibiting double-hatting in ECT arbitration. Disclosure requirements vary by institution. The practice continues across all three tribunal networks documented above.

---

V. The Defiance Record: Every Tribunal, Except One

The most architecturally significant property of the ECT conduit layer is documented in the case-by-case record of how its tribunals responded to the CJEU's rulings. This record, compiled from the Kluwer Arbitration Blog's systematic case tracking and confirmed in the Gibson Dunn client alerts, shows a pattern of remarkable consistency.

Documented Tribunal Response to CJEU Rulings — Case Record

2018

CJEU rules in Achmea (Case C-284/16): Investor-state arbitration clauses in bilateral investment treaties between EU member states violate EU law. The European Commission and most EU member states declare this applies to intra-EU ECT claims. The Commission pressures member states to terminate intra-EU BITs.

ECT tribunal response: Every ECT tribunal that considers the Achmea ruling as a jurisdictional objection rejects it. The reasoning: Achmea concerned a bilateral treaty, and the ECT — as a multilateral treaty to which the EU itself is a party — is legally distinct.

FSA note: The "multilateral treaty" distinction is the tribunals' legal rationale for continuing jurisdiction. It is a defensible legal argument under the Vienna Convention on the Law of Treaties. FSA's interest is not in the legal merits of the argument but in the structural fact: the conduit network found a legal argument that allowed it to continue operating after the EU's highest court ruled against it, and used that argument unanimously.

2021

CJEU rules in Komstroy (Case C-741/19): The CJEU explicitly extends its Achmea reasoning to the ECT itself, ruling that Article 26 ECT does not apply to intra-EU disputes. The Commission states this closes the "multilateral treaty" distinction the tribunals had relied on since 2018.

ECT tribunal response: Overwhelmingly unchanged. Post-Komstroy ECT tribunals continue rejecting intra-EU jurisdictional objections. The reasoning now: the CJEU's rulings operate within the EU legal order; international arbitration operates under public international law; the two systems are separate, and the CJEU cannot bind an international arbitration tribunal seated outside the EU.

FSA note: The two-legal-order argument is the conduit layer's most architecturally precise self-insulation mechanism. It does not deny the CJEU's authority within the EU. It claims a separate jurisdiction — international law — where the CJEU has no reach. The tribunal network is not defying EU law within the EU legal order. It is operating in a parallel legal order that the EU cannot directly control.

2022

European Commission issues COM(2022)523: The Commission formally documents, in official EU text, that ECT tribunals "have thus disregarded the applicable rules of public international law." The Commission calls for EU collective withdrawal from the ECT. It states it "regrets" the tribunal network's consistent refusal to apply the CJEU's rulings.

ECT tribunal response: Continued rejection of intra-EU jurisdictional objections. The UK High Court, in a 2023 enforcement case, states directly regarding the CJEU's reasoning: "with the greatest of respect to the CJEU, it is not the ultimate arbiter under the ICSID Convention."

FSA note: The UK High Court's 2023 statement is the conduit layer's clearest public articulation. An English court — applying the ICSID Convention, which is international law — says the EU's highest court is not the final authority on a question of international treaty law. From inside the conduit, this is legally defensible. From outside the conduit, it is a private arbitration network overruling democratic institutions. Both of those things are simultaneously true.

2024–2025

The one exception: Of all ECT tribunals that have considered intra-EU jurisdictional objections across this entire period, one — Green Power v. Spain — applied the Achmea/Komstroy reasoning and declined jurisdiction. One. Every other tribunal rejected the objection and proceeded to hear the case.

EU internal enforcement battles: EU member states begin bringing anti-suit injunctions in domestic courts to stop investors from pursuing ECT claims. Spain and Poland take investors to home-country courts. The Amsterdam court declines to issue the injunction in one case. The legal battlefield expands. The ECT arbitration network continues operating.

FSA structural summary: One exception in dozens of cases over eight years. The conduit layer's jurisdictional self-insulation has an empirical record approaching unanimity. The mechanism did not need to win legal arguments in every jurisdiction. It needed only to find one enforceable seat — and seats outside the EU, including Washington D.C. (ICSID), Stockholm (SCC pre-2025 changes), London, and New York — remain operative enforcement channels.

---

VI. What the Conduit Layer Builds

FSA's conduit layer is not the source of the ECT's power. The source layer built the instrument — the 1994 treaty text with its asymmetric investment protections and its survival clause. The conduit layer is what makes the source layer's power operational and self-sustaining. Without the three tribunal networks, the revolving door of elite lawyers moving between counsel and arbitrator roles, the secret proceedings and limited appeal rights, and the jurisdictional independence that even the EU Court of Justice could not overcome, the ECT would be a document with words. The conduit layer is what makes those words into €190 million awards, $10 million defense costs, and climate legislation modified before any case is filed.

FSA Structural Finding — The Conduit Layer's Defining Properties

Jurisdictional autonomy: The tribunal networks operate under international law, not EU law, not national law. Their jurisdictional independence is not a defect — it is the conduit's designed operating condition. An investment protection system that could be overridden by the domestic courts of the states being sued would not function as an investment protection system.

Personnel concentration: The top 25 lawyers in the network have appeared in half of all international investment arbitrations. The same individuals rotate between representing claimants and sitting as arbitrators. The network's legitimacy depends on the reputation of those individuals. The network's outputs — including the legal arguments that consistently defeat Achmea-based jurisdictional objections — are produced by a group whose professional and financial interests are served by the continued operation of the system they adjudicate.

Cost asymmetry as structural property: The $10 million average defense cost operates as a permanent deterrent against regulatory action regardless of the legal merits of any specific claim. The conduit does not need to win every case. It needs only to make the cost of fighting every case higher than the cost of accommodation. The regulatory chill documented in Post 2 is the conduit's most efficient output — the cases that never need to be filed because the architecture's existence was sufficient.

Opacity by design: UNCITRAL proceedings may leave no public record. ICSID publishes case registration and final awards but not all procedural decisions. The full scope of ECT litigation — and, more importantly, the full scope of ECT settlement and pre-filing accommodation — is structurally unknowable. The opacity is not incidental. It is the conduit's third insulation mechanism, after jurisdictional independence and cost asymmetry.

"The existence of a core group of counsel, which includes a significant number that also act as arbitrators, may enhance the maintenance of information asymmetries... one arbitration lawyer recently boasted that their competitive advantage lies in what is 'not written down.'" — Journal of International Economic Law, Oxford Academic
Quantitative Empirical Study of the Investment Arbitration Network, 2017

Post 4 maps the conversion layer: how the conduit's outputs — the tribunal awards, the settlement threats, the jurisdictional findings — convert democratic climate legislation into a compensable harm. The RWE and Uniper cases against the Netherlands are the conversion layer's most architecturally precise examples: two companies, one democratically enacted coal phase-out law, €2.8 billion in combined claims, and a legal theory under which the act of passing a climate law is the injury requiring compensation.

Source Notes

[1] European Commission COM(2022)523: "Communication from the Commission — Energy Charter Treaty" (September 22, 2022). Full text at eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=CELEX:52022DC0523. The exact language — "CONSIDERING that arbitral tribunals... have thus disregarded the applicable rules of public international law" and "REGRETTING that those arbitral tribunals" have continued to accept jurisdiction — is from this official document. This is European Commission language, not characterization by FSA or any advocacy source.

[2] CJEU Achmea ruling: Slovak Republic v. Achmea BV, Case C-284/16, March 6, 2018. CJEU Komstroy ruling: Republic of Moldova v. Komstroy, Case C-741/19, September 2, 2021. Both fully documented and publicly available through the CJEU.

[3] Tribunal network analysis and double-hatting: "Revolving Door in International Investment Arbitration," Journal of International Economic Law, Oxford Academic, Vol. 20, Issue 2, 2017 (available at academic.oup.com). The 2,699 lawyers identified, the 14%/1% distribution figures, and the "not written down" quote are from this peer-reviewed quantitative study. The Philip Sands quotation on double-hatting is documented in the same study and confirmed across multiple arbitration law publications.

[4] Post-Achmea case record: Kluwer Arbitration Blog, systematic coverage 2018–2025; Gibson Dunn client alerts (gibsondunn.com) on Achmea (2018) and Komstroy (2022); Queritius legal analysis on post-Achmea intra-EU disputes (queritius.com, July 2023); European Arbitration Review 2026 (Global Arbitration Review). The "one exception" — Green Power v. Spain — is documented across all these sources as the sole ECT tribunal to have applied the Achmea/Komstroy reasoning to decline jurisdiction.

[5] UK High Court quote — "with the greatest of respect to the CJEU, it is not the ultimate arbiter under the ICSID Convention": Infrastructure Services Luxembourg S.A.R.L., Energia Termosolar/Spain, [2023] EWHC 1226 (Comm), paragraph 80 (May 24, 2023). Cited in the CELIS Institute analysis of investment arbitration and EU law (celis.institute).

[6] Cost of defending ECT arbitration: IISD Investment Treaty News; Columbia Journal of Transnational Law, "The Energy Charter Treaty: Reform or Retreat?" (jtl.columbia.edu, March 2025). The $10 million average defense cost figure is widely cited across ECT reform literature and confirmed by multiple independent sources.

FSA: The Treaty That Won't Let Go — Series Structure

POST 1 — PUBLISHED

The Rockhopper Moment: The Anomaly

POST 2 — PUBLISHED

The Source Layer: 1994 and the Architecture of Capture

POST 3 — YOU ARE HERE

The Conduit Layer: The Private Court System

POST 4

The Conversion Layer: Democracy as Compensable Harm

POST 5

The Shadow Trader Layer: Geneva, Zug, and the Invisible Architecture

POST 6

The Escape: Nations That Tried to Leave

POST 7

FSA Synthesis: The Treaty as Template for Permanent Insulation