Which offences qualify as predicates varies by jurisdiction, and that variation is operationally significant. FATF Recommendation 3 sets a baseline, designating categories of offences that member states must treat as predicates — but each state determines how those categories are defined in domestic legislation, which offences fall within them, and what evidentiary threshold establishes that the predicate was committed. A state may implement the Recommendation narrowly, excluding conduct that another jurisdiction would capture, without being formally non-compliant.

Where predicates are defined narrowly, or where the originating conduct was legal where the crime occurred, the money laundering charge becomes difficult or impossible to establish. In cross-border cases, this interacts directly with dual criminality: if the predicate conduct is not recognised as criminal in the requested state, mutual legal assistance may be refused, the asset trail goes cold, and the legal architecture quietly converts delay into impunity.

Definition drawn from FATF Glossary and AML/CFT Library (aml-cft.net).

In practice, the distinction between process and consequence is narrower than the formal framing implies. Grey-listed jurisdictions face heightened scrutiny from correspondent banks, international lenders, and sovereign credit rating agencies. The IMF and World Bank factor grey list status into country assessments. Foreign direct investment typically contracts. The pressure to exit the list is therefore considerable — which is precisely the mechanism FATF intends.

The grey list currently contains between twenty and thirty jurisdictions at any given time, spanning every region. A jurisdiction exits when FATF determines that deficiencies have been sufficiently addressed, a determination made by the same peer review process that identified them. Whether the exit reflects genuine systemic improvement — or successful performance of compliance — is a question the mechanism is not designed to resolve.

See also: FATF. FATF Blacklist. Peer Review. Sovereignty.

Source: Financial Action Task Force, Jurisdictions under Increased Monitoring (updated February 2025)

The consequences of blacklisting are substantial. FATF calls on member jurisdictions to apply enhanced due diligence to transactions involving listed states, and in the most serious cases, to apply countermeasures — actively restricting financial flows rather than merely scrutinising them. Correspondent banking relationships become difficult to maintain, access to international capital markets contracts, and the reputational signal to private financial institutions is significant, independent of any formal requirement.

As of 2023, the blacklist has rarely included more than three jurisdictions at any one time. Its influence stems partly from its rarity — and partly from the credible threat of placement, which encourages most states to remain engaged with the grey list process.

See also: FATF. FATF Grey List. Mutual legal assistance.

Source: Financial Action Task Force, High-Risk Jurisdictions Subject to a Call for Action (updated February 2025).

The method is not arbitrary. It reflects the foundational principle of international law, sovereign equality: states do not submit to external supervision in the way domestic institutions might. Peer review offers a means of collective accountability that preserves that equality while creating at least formal pressure toward compliance.

The structural limitation is inherent. States reviewing their peers have their own compliance records, their own political relationships, and their own interest in a process that does not set uncomfortable precedents. The UNCAC Implementation Review Mechanism operates on this model — states review each other under confidentiality provisions, with no independent verification of findings and no enforcement mechanism for non-compliance. The review exists. The accountability it produces is a separate question.

See also: Sovereignty. Reciprocity. FATF.

Source: United Nations Convention against Corruption, opened for signature 9 December 2003, 2349 UNTS 41 (entered into force 14 December 2005) arts 63–64; Conference of the States Parties to UNCAC, Terms of Reference of the Implementation Review Mechanism, Resolution 3/1 (2009).

The central authority is the mandatory routing point for all formal MLA. A request sent directly to a foreign court, prosecutor, or police agency — bypassing the central authority — has no formal legal standing and cannot compel any response. Every request must enter the system at this point.

Central authorities exercise significant discretion. They determine whether a request meets the formal requirements of the relevant treaty, assign it a priority within their workload, decide whether to seek clarification before transmitting it, and — in some jurisdictions — make an initial assessment of dual criminality and proportionality before the request proceeds further. No treaty prescribes how long any of these steps should take.

In grand corruption cases, where the target may have connections to the government of the requested state, the central authority is the point at which legitimate administrative process and deliberate delay become structurally indistinguishable.

Definition drawn from UNODC MLA Writer's Guide and Basel Institute on Governance MLA resources.

The distinction matters because it shapes how international accountability mechanisms are designed and how states respond to them. A mechanism that formally respects sovereignty may nonetheless limit autonomy — requiring reporting, conditioning aid, or imposing reputational consequences for non-compliance. Conversely, a state may invoke its autonomy to resist measures it has formally agreed to, arguing that implementation is a matter of domestic discretion.

In the cases examined in this publication, the invocation of state autonomy typically appears at the point where a formal commitment meets a politically inconvenient obligation. The commitment remains. The obligation does not.

See also: Sovereignty. Proportionality.

Sources: Declaration on Principles of International Law concerning Friendly Relations among States, UN GA Res 2625 (XXV) (24 October 1970); Charter of the United Nations, Article 2(7).

In practice, sovereignty means that no state is legally subordinate to another. A state cannot be compelled to extradite a suspect, execute a foreign court order, or disclose financial records held within its jurisdiction without its consent — typically formalised through a treaty. The principle protects genuine independence. It also, in precisely the same operation, protects deliberate non-cooperation.

This publication will encounter sovereignty frequently — not as an abstract principle, but as a procedural argument. When a requested state declines a mutual legal assistance request, cites domestic legal constraints, or limits the scope of an international review, it is exercising sovereign authority. The argument is legally legitimate. The outcome is that accountability stops at the border.

See also: Autonomy. Mutual legal assistance.

Sources: Charter of the United Nations, Article 2(1); Declaration on Principles of International Law concerning Friendly Relations among States, UN GA Res 2625 (XXV) (24 October 1970).

In the context of mutual legal assistance (MLA), reciprocity operates as both an incentive and a constraint. States are more likely to cooperate with MLA requests from jurisdictions that have cooperated with their own. Conversely, a state that routinely delays, refuses, or conditions its responses may find that its own requests receive equivalent treatment. The principle creates informal accountability where formal enforcement mechanisms are absent.

Reciprocity has limits. It cannot justify a state violating a binding obligation on the grounds that another state has done the same. But as a driver of cooperation in a system with no enforcement authority, it does significant work — and its absence is conspicuous.

See also: Mutual legal assistance. Sovereignty.

Sources: ICJ, Case Concerning Military and Paramilitary Activities in and against Nicaragua (Nicaragua v United States) [1986] ICJ Rep 14; Declaration on Principles of International Law concerning Friendly Relations among States, UN GA Res 2625 (XXV) (24 October 1970).

The mechanism relies on strict administrative criteria, including dual criminality, proportionality, confidentiality, and compliance with local evidentiary rules. Typical requests seek witness evidence, banking and communications records, production orders, search and seizure, asset restraint, or service of process.

No treaty mandates a response timeline. Central authorities retain significant discretion over sequencing and priority. Each condition attached to a request represents a legitimate basis for delay or refusal — and in cases involving politically connected targets or complex cross-jurisdictional asset structures, these conditions stack.

The Azerbaijani Laundromat moved USD 2.9 billion through four UK shell companies while investigations proceeded separately in four jurisdictions: Denmark, Estonia, the United Kingdom, and the United States. Total recovery: approximately 0.2 per cent.

Definition drawn from the Basel Institute on Governance (Introduction to MLA) and AML/CFT Library (aml-cft.net).

In mutual legal assistance, proportionality is an assessment criterion applied by the requested state before executing a request. The assistance sought must be commensurate with the seriousness of the offence and the centrality of the evidence to the proceedings. A sweeping production order in a minor regulatory matter may not satisfy it. A targeted request for specific banking records in a billion-dollar fraud will generally pass. The assessment is legitimate, necessary, and — in the hands of a requested state with its own reasons for delay — extensible.

The principle operates similarly in international sanctions and the use of force, where it requires that coercive measures not exceed what the triggering conduct warrants. In every context, the logic is the same: the means must fit the end. In grand corruption cases, determining what fits takes time. That time is rarely neutral.

Sources: ICJ, Case Concerning Military and Paramilitary Activities in and against Nicaragua (Nicaragua v United States) [1986] ICJ Rep 14; Charter of the United Nations, Article 2(4); FATF Glossary and AML/CFT Library (aml-cft.net).

Here is a simple question. A company enters into a contract with your local council. Public money changes hands. So, who owns the company?

The answer, for most of recorded corporate history in the United Kingdom, was: you do not get to know. The law did not require anyone to tell you. It did not require anyone to tell the government. In many cases, it did not require anyone to record the information at all.

This is not a historical curiosity. It is a design feature — and the one that made the Scottish Limited Partnership one of the most reliable vehicles in the architecture of grand corruption.

The Instrument

A Scottish Limited Partnership (SLP) is a legal entity registered in Scotland under the Limited Partnerships Act 1907 (UK). Unlike its English equivalent, the SLP has a separate legal personality — meaning it exists as a legal entity in its own right, capable of holding assets, opening bank accounts, and entering contracts as though it were a person rather than a collection of people. There are two classes of partner. The general partners, who manage the business and carry unlimited liability, and the limited partners, who contribute capital and are shielded from liability beyond their contributions.

What made the SLP structurally useful for purposes beyond the legitimate was a specific combination of features. It was not required to file accounts, nor did it have to disclose its ultimate beneficial owner to any public registry. Furthermore, its general partner—the entity legally responsible for management—could itself be a company registered in an offshore jurisdiction. To ask who ran the SLP was to be pointed to an address in the British Virgin Islands; to ask who ran that company was to find the trail run cold.

The SLP was not designed as a vehicle for financial crime. It was designed in 1907, when the relevant concerns were different, and the relevant sums were smaller. The structure simply did not care what it was used for.

The Scene of the Crime

Between 2012 and 2014, approximately USD 2.9 billion passed through Danske Bank’s Estonian branch. A significant portion moved through UK-registered shell companies — entities with no active business operations, designed to hold assets or move money while obscuring their ownership. Four of them — Polux Management, Hilux Services, Metastar Invest, and LCM Alliance — were at the centre of what became known as the Azerbaijani Laundromat: a scheme that used these structures to move funds from the International Bank of Azerbaijan, a state institution, through the European financial system, financing luxury purchases for Azerbaijan’s ruling elite and payments to European politicians.

The beneficial owners of the companies at the centre of the scheme were not publicly known because they were not required to be. The law did not ask. The registry did not record. The money moved through structures that were entirely legal at the point of registration, in a jurisdiction that had not yet decided that transparency was an obligation rather than a courtesy.

The Organised Crime and Corruption Reporting Project (OCCRP) and the Danish newspaper Berlingske published the investigation in September 2017. By then, the money had been moving for five years.

The Verdict Arrived Too Late

In 2017, the United Kingdom introduced a requirement for SLPs to register persons with significant control — their beneficial owners — through the Scottish Limited Partnerships (Register of People with Significant Control) Regulations 2017 (UK SI 2017/694). The Azerbaijani Laundromat operated between 2012 and 2014. The regulations came into force three years after the scheme had concluded and the money had moved.

This is the precise gap the piece examines. Not that the law failed — it did not fail, it functioned as written. Not that regulators were negligent — they operated within the framework available to them. The gap existed for a simpler reason: there was no legal obligation to disclose beneficial ownership at the time the money moved.

The structure was legal. The opacity was legal. The gap between what the law permitted and what accountability required was not a malfunction. It was the law.

The Reason Nothing Changed

Beneficial ownership opacity persists not because governments cannot solve it, but because solving it requires sustained political will in the face of sustained lobbying by the financial and legal industries that profit from the existing architecture. The 1907 Act was not amended for 110 years. The register of persons with significant control — the PSC register — introduced under the Companies Act 2006 (UK), Part 21A, applied to private companies but not, initially, to SLPs: an omission that required entirely separate regulations to address, a decade later.

The Financial Action Task Force, or FATF, the international standard-setter for anti-money laundering frameworks, has recommended beneficial ownership transparency as a core standard since 2003. The gap between international recommendations and domestic legislative implementation is where the money moves. It is a gap measured not in weeks but in decades, sustained by the same lobbying interest that made the amendment politically inconvenient for the 110 years before it.

The Azerbaijani Laundromat is not an anomaly. The 1MDB scheme, examined in the first piece in this series, used shell companies across Singapore, Switzerland, Luxembourg, and the United States — each jurisdiction offering its own version of the same structural permission: you may hold assets here without disclosing who you are.

The Answer

So, who owns the company that just contracted with your local council?

If it is a UK private company incorporated after 2016, there is a PSC register entry — though its accuracy depends on self-reporting, and enforcement against false entries is limited. If it is an SLP incorporated before 2017, the answer may simply be: the record does not exist. If it is a company in a jurisdiction that has not yet implemented FATF recommendations on beneficial ownership — and many have not — the answer is the same one that covered the Azerbaijani Laundromat for five years.

We do not know because the law didn’t ask.

That is not a gap in the system. That is the system.

Next in this series: The Architecture of Delay: Mutual Legal Assistance, USD 2.9 billion and the ‘Azerbaijani Laundromat’

Mutual legal assistance was designed to enable cross-border cooperation. In grand corruption cases, its structural features enable something else entirely.

A movie about hiding illicit profits in a perceived foreign safe haven, produced with illicit profits hidden in a perceived foreign safe haven.

The fraud is extraordinary. But set it aside for a moment. What follows is about something more structural: the international mechanism that was supposed to catch exactly this kind of conduct — and why, by design, it never could. During the same years in which that money was moving, Malaysia was a state party in good standing to the United Nations Convention Against Corruption (UNCAC). Malaysia was actively participating in the treaty’s official monitoring system — a process where diplomat-experts audit each other’s anti-corruption laws. The bureaucratic machinery ran. The boxes were ticked.

It found nothing.

This is not a case of failed oversight. It is a case study in an architecture functioning precisely as intended: designed by the very states subject to it to avoid the conditions under which failure could ever be measured. The question is not why the apparatus missed 1MDB. The question is what, structurally, it was ever capable of finding.

The Convention and the Catch

UNCAC is the only legally binding global anti-corruption instrument. Adopted in 2003 and in force since 2005, it has been ratified by 190 states — near-universal participation. Its four pillars are prevention, criminalisation, international cooperation, and asset recovery. Ratification is not a signalling of good intentions; it is a binding legal obligation.

That obligation comes, as it turns out, almost entirely without teeth. No external authority can compel a state to comply, penalise one that does not, or independently verify whether obligations are being met. What UNCAC has instead is the Implementation Review Mechanism (IRM). Established in 2009 and operational since 2010, the IRM’s stated purpose is to assist states in implementing their obligations through a structured peer-review process.

The mechanism’s own language is worth a moment’s attention: it frames itself not as an auditor but as a partner in capacity building.

It is designed, in its own terms, to help states comply — not to determine whether they have done so.

That framing is not incidental. A mechanism that positions itself as a partner rather than an examiner has already made a structural choice about what it will and will not find.

The Audit, By the Audited

To understand why the IRM missed the scheme taking place, it helps to look at how it actually operates — not as it describes itself, but as it functions. Each element of the review process represents a structural choice. Line those choices up, and the pattern is either a remarkable series of coincidences or entirely by design.

The review is conducted by peers. Each state under review is assessed by two other governments — one from its own regional group and one from outside it. Reviewers are government-designated. This matters because every state in the room will eventually be the one under review. The structural interest is not in rigorous scrutiny. The interest is in a comfortable process that reflects well on everyone. Reciprocity is a powerful norm. Independent scrutiny is, in this context, nobody’s friend.

The primary input is self-assessment. The review opens with a questionnaire completed by the state under review. Reviewers work from this as their foundational document, with no obligation to cross-reference self-reported information against court records, investigative reporting, or the findings of international monitoring bodies. The reviewed state describes itself. Reviewers assess that description.

Site visits are optional and require consent. In-country visits — where reviewers observe implementation in practice rather than on paper — are the most consequential form of scrutiny available. They are also entirely voluntary. A state that has something to conceal may simply decline to host one. Most do, and the mechanism has no response to that beyond noting the absence.

Reports are confidential by default. The actual findings — the full review report — are published only if the reviewed state consents — an option that states with serious corruption concerns exercise freely, which is to say, they decline. What the public sees instead is an executive summary, a document into which the reviewed state has substantial input. The subject of the review edits the account of the review. If this sounds like the ending you already expected, that is because it is.

Civil society has no formal role. Non-governmental organisations, investigative bodies, and human rights monitors — the actors most likely to hold documented evidence of the gap between a state’s legal framework and its operational reality — have no formal standing in the process. They may submit information. Reviewers are not required to consider it. The people who most often know what is actually happening are formally peripheral to a process designed to find out what is actually happening.

Recommendations are not binding. The review concludes with recommendations carrying no compliance deadline, no follow-up enforcement, and no consequences for ignoring them entirely. They are, in the most precise sense of the word, suggestions.

Each of these features might be defended in isolation as pragmatic — a concession to state sovereignty, a recognition of resource constraints, a gesture toward cooperative monitoring. Together, they produce a closed system in which the information that would make accountability possible is systematically excluded, and the information that is included is provided by the party with the greatest interest in a favourable outcome.

The IRM is not a detection system. It is a certification system — and what it certifies is not compliance, but participation in the process of reporting compliance. The distinction between those two things is the black hole at the centre of the system: where accountability enters, and nothing escapes.

The Fund, the Film, and the Review

Between 2009 and 2015, more than USD 3.5 billion was misappropriated from 1MDB—a sovereign wealth fund established by the Malaysian government with a stated purpose of national economic development. Its actual purpose, according to DOJ civil forfeiture complaints, was to become the vehicle for one of the largest state-linked financial frauds in recorded history.

The advisory board was chaired by Prime Minister Najib Razak. Stated purpose met actual execution through a network of shell companies—entities designed to place layers of apparent legitimacy between stolen money and its destination—across Singapore, Switzerland, Luxembourg, and the United States.

The money reached a Manhattan penthouse, a Beverly Hills mansion, a private jet, Van Gogh and Monet paintings, and eleven wire transfers totalling USD 64 million directly to Red Granite Pictures, the Hollywood production company co-founded by Riza Aziz, the Prime Minister’s stepson. Red Granite produced The Wolf of Wall Street. The film — a story about a man who steals money, hides it offshore, and eludes accountability — grossed over USD 392 million worldwide. The producers were apparently familiar with the territory.

The central figure in the scheme’s financial architecture was Low Taek Jho — known as Jho Low — a Malaysian financier whose name runs throughout both the DOJ complaints and OCCRP’s extensive investigative archive. Jho Low has not been apprehended; he is reportedly living under an assumed identity in Shanghai. Najib was convicted of corruption by a Malaysian court in 2020 and is currently serving a twelve-year sentence.

Meanwhile, at the United Nations

Malaysia ratified UNCAC in 2008, one year before 1MDB was established. Between 2010 and 2015, under the IRM’s first review cycle, Malaysia’s anti-corruption and international cooperation frameworks were formally assessed — precisely the areas most relevant to the multi-jurisdictional money movement happening at the same time.

The review ran. Malaysia’s self-assessment was completed. The peer reviewers — government experts from Kenya and the Philippines — conducted their dialogue and produced their report. That report addressed Malaysia’s anti-corruption legal frameworks and international cooperation mechanisms in considerable detail. It does not mention 1MDB. By 2015, DOJ investigators and the Sarawak Report group had already begun documenting, through publicly available financial records and corporate filings, what those frameworks had failed to prevent. The information existed. The mechanism’s defined scope was pointed elsewhere.

Managing the Record

In 2015, Malaysia amended its Official Secrets Act to restrict domestic reporting on 1MDB. The Attorney-General investigating the fund, Abdul Gani Patail, was controversially removed and replaced. Editors were pressured. Publications were suspended.

A mechanism reliant on state self-reporting, excluding civil society, and making site visits contingent on consent cannot compensate for a state actively managing its own information landscape. This is not a criticism of individual reviewers — they worked within the system they were given. It is a description of what the system structurally permits — and in this case, what it permitted was everything.

The mechanism did not miss 1MDB because it looked in the wrong place. It missed 1MDB because its architecture only permitted it to look where the state had already agreed to be seen.

By Design

The States Wrote Their Own Review

The IRM was negotiated by the Conference of the States Parties to UNCAC — that is, by the states being reviewed, determining the terms of their own review. The decisions to make site visits optional, to anchor the process in self-assessment, to exclude civil society, to default to confidentiality, and to issue non-binding recommendations were not oversights. They were negotiated positions, agreed upon by governments whose interests lay in maintaining the performative functioning of an international anti-corruption framework while preserving maximum autonomy over its findings.

Consider each design choice in terms of who benefits. Optional site visits benefit states where the gap between legal framework and operational reality is widest. Confidential reporting benefits states that would suffer reputationally from published findings. Civil society exclusion benefits states whose compliance gaps are best documented by journalists and NGOs. Self-assessment benefits states that control their own information landscape. Peer review by fellow state parties benefits states whose peers share a structural interest in keeping the mechanism comfortable — because the mechanism reviews them all.

This is not a conspiracy. It is the ordinary logic of international treaty design: states negotiate the minimum accountability architecture necessary to ratify the instrument. The result, reliably, is an architecture that serves the states doing the negotiating.

Participation Is Not Compliance

This distinction matters beyond the 1MDB case. It is the structural logic of international self-regulatory frameworks more broadly — mechanisms in which the regulated parties negotiate the regulatory architecture, the information inputs, and the consequences of non-compliance. UNCAC is an extreme illustration, but it is not an anomaly. The same logic appears in voluntary corporate governance frameworks, industry self-regulation, and bilateral investment treaty arbitration systems, where states and investors jointly determine the rules governing disputes between them.

The 1MDB case is useful precisely because it is extreme — the scale of the fraud, the involvement of the head of government, the multi-jurisdictional money flows, the eventual criminal conviction of a sitting prime minister. All of this makes the gap between what was occurring and what the mechanism found impossible to attribute to bad luck or insufficient diligence. The gap is structural. And a structural gap of this kind does not close on its own.

It Functioned as Designed — and That Is the Problem

There is a phrase familiar to anyone who has watched an institution investigate itself: we looked into the matter and found no evidence of wrongdoing. It travels across industries and jurisdictions — from corporate audits to police complaints to parliamentary inquiries — and, after sufficient repetition, acquires an almost comedic reliability. The finding is always the same. The process is always described as thorough. The gap between the two is never explained.

The UNCAC Implementation Review Mechanism is the international community’s formal version of this arrangement. It exists. It operates. It produces reports, recommendations, and executive summaries. States participate. The process is followed. In this structural sense, it functions perfectly.

What it does not do is detect the misappropriation of USD 3.5 billion from a sovereign wealth fund by the officials responsible for it, during the review cycle covering the precise legal frameworks relevant to that conduct, in a state that participated throughout in good standing. It does not do this because it cannot — not as a matter of capacity, but as a matter of design. A mechanism that excludes independent verification, marginalises civil society, makes its most consequential elements conditional on state consent, and defaults to confidentiality will find what states choose to report. In Malaysia’s case, that was not 1MDB.

Leslie Caldwell was right that life imitated art. What neither she nor anyone else at that press conference observed was that the UN’s anti-corruption review had spent the intervening years certifying the canvas.

The IRM was not designed to fail. It was designed to avoid the conditions under which failure could be measured. That is a more precise and more damning description. Failure implies an attempt.

What the IRM offers instead is participation — the performance of accountability, conducted according to rules written by its subjects, producing findings that its subjects control, carrying consequences that its subjects can ignore.

The money moved. The review cycle ran. The mechanism found compliance.

Case closed.

All three outcomes were, structurally, entirely predictable.

Next in this series: Found It: Clare Rewcastle Brown and the 1MDB Billions

How one investigative journalist, a whistleblower, and a foreign Department of Justice uncovered what UNCAC’s review mechanism missed entirely. A case study in accountability when systems fail.

The distinction between legal ownership and beneficial ownership is where financial crime lives. A shell company may be registered in the name of a nominee director or another corporate entity. The person actually controlling the structure — directing where the money goes, benefiting from what it holds — may appear nowhere in the public record.

Beneficial ownership information can sometimes be recovered from company registers, though accuracy, completeness, and public accessibility vary significantly by jurisdiction. In many of the cases examined in this publication, the gap between what a register discloses and who actually controls the entity is precisely where the money disappears.

See also: FATF Guidance on Beneficial Ownership Transparency

Definition drawn from FATF Glossary and AML/CFT Library (aml-cft.net)

FATF doesn’t enforce the law. What it does do is more structurally significant: it sets the international standard. Its Recommendations — 40 in total, revised in 2023 — establish the benchmarks for compliant anti-money laundering and counter-terrorist financing systems. Jurisdictions that fail to meet the standard risk being placed on FATF’s grey or black lists, which triggers reputational and financial consequences serious enough that most states treat the Recommendations as effectively binding, regardless of their formal legal status.

This publication will reference FATF frequently. Its definitions form the baseline for financial crime terminology across member jurisdictions, and its mutual evaluation process — through which member states assess each other’s compliance — is itself an accountability mechanism worth examining.

FATF's guidance on beneficial ownership transparency sets the international standard for what disclosure member states are expected to require — and documents how far practice falls short of it.

Definition drawn from the FATF mandate and FATF Recommendations (2023).

This is not inherently criminal. Legitimate uses include holding intellectual property, simplifying corporate structures across jurisdictions, and managing assets during mergers. The problem is structural: because a shell company has no operations, it has no operational paper trail. Nothing to audit, no payroll to examine, no supply chain to trace.

In financial crime, that absence is the point. Each additional shell in a chain adds a jurisdictional layer and another set of records to subpoena across multiple legal systems, each with its own disclosure requirements and timelines. Unwinding the chain requires MLA requests and cooperation from each jurisdiction in sequence. The 1MDB funds moved through shells across Singapore, Switzerland, Luxembourg, and the United States before reaching a Manhattan penthouse, a Picasso, and a Hollywood production company. By the time the legal process reaches the far end of a chain like that, the asset is frequently gone.

Definition drawn from FATF Glossary and AML/CFT Library (aml-cft.net)

In practice, the gap widens with complexity. Financial crime frequently involves conduct that sits at the edges of how different legal systems define criminalisation — clearly criminal in one jurisdiction, differently classified or unlegislated in another. Each gap is a legally sufficient basis for refusal, and assets continue to move while the question is resolved. Where the predicate offence is not recognised as criminal in the requested state, the money laundering charge collapses at the first condition.

The principle exists to protect states from enforcing foreign criminal standards that conflict with their own legal frameworks. That is a legitimate function. In grand corruption cases, it is structurally indistinguishable from obstruction.

Definition drawn from the Basel Institute on Governance (Introduction to MLA)

Some people are satisfied knowing that something happened. Leo has never been one of those people — she needs to know how, and why, and what the system that was supposed to prevent it was doing at the time. She has, in fact, always wanted to open a detective agency. She still does. This publication is the first step toward that: building the body of analytical work that will, eventually, become the practice. For now, this is the office.

The Holistic Detective exists because the most important part of any story is never the headline — it’s the mechanism behind it. The quiet infrastructure of decisions made and not made, laws written and not written, evidence collected and lost and ignored, that made an outcome not just possible but structurally inevitable. Most of the time, the pieces were there. They were just held by people and institutions that couldn’t, or wouldn’t, connect them.

That is the question this publication keeps asking. Not what happened — you already know what happened. But how the systems put in place to prevent, catch, or respond failed to do so. And whether that was a bug or a feature.

The cases here range from financial crime and grand corruption to compromised investigations and institutional misconduct. In every case, the system left behind enough of a record to reconstruct exactly where it went wrong and why.

O.J. Simpson walked because the investigation was catastrophic and the prosecution handed the defence a gift wrapped in a glove. Steven Avery may well be guilty — but the police tried to frame him anyway, clumsily and on camera, and that matters independently of whether he did it. The adjacent Brendan Dassey case stands as the poster child for coerced confession. These things are not separable. The system’s conduct is often obfuscated, but always part of the story.

This publication pulls those structures apart. Not to tell you what to think, but to show you how the machinery works — and, more usefully, how it fails in its stated purpose.

How it’s Organised

Each series focuses on a case or subject with a known name — a scandal, a crime, an institution — and follows the structural failures underneath it. The known case is the door.

Posts are tagged by topic and grouped by series, so you can follow a thread from beginning to end or dip in wherever something catches your attention. Each substantive post is accompanied by glossary entries for key terms — always free, always linked inline in the text. Hyperlinks within articles go to source documents and glossary entries; it is always worth knowing where a link leads.

Paid subscribers get full access to the deep-dive criminal case series. The first instalment of each series is free.

One post per week, plus any accompanying glossary entries. No filler.

A Note on the Name

For those unfamiliar, The Holistic Detective is a Douglas Adams reference — a nod to Dirk Gently and the fundamental interconnectedness of all things. Adams has been a lifelong influence: his ability to notice the absurdity built into the structures people accept without question, and to describe it with the calm of someone who simply cannot understand why no one else is pointing at it, is precisely the lens this publication applies to institutional failure.

Nothing happens in isolation. The mechanisms connect to each other. Following the thread usually leads somewhere more interesting than you expected, and considerably more alarming than anyone officially confirmed.

The ‘42’ is on the logo; if you noticed, you’re in the right place.

About the Author

As a child, Leo grew up reading everything — detective fiction to the classics, true crime to astronomy, whatever was within reach — and developed an early and enduring suspicion that the official explanation was rarely the whole story.

She holds a Master of International Law from Griffith Law School with Distinction and a Bachelor of Criminal Justice and Criminology with High Distinction from Swinburne University. She is certified in OSINT, financial analysis, mutual legal assistance frameworks, and anti-money laundering through the Basel Institute on Governance, and is completing the Certified Fraud Examiner qualification.

Her academic work spans international law, criminology, institutional accountability, global politics, democratic erosion, and the legal architecture of crime. The through line is consistent: accountability gaps are rarely accidental, and the mechanisms that create them are almost always visible to anyone willing to look. The systems fail. Accidentally, on purpose.

Alongside research and writing, Leo spent several years in business, managing a diverse portfolio of companies’ travel expenditure, strict compliance requirements, and bespoke contracts. She also spent time diagnosing and rebuilding failing small businesses: financial analysis, fraud detection, operational overhaul, and the particular satisfaction of making a dysfunctional system functional. The same instincts, it turns out, apply at the institutional scale.

For licensing, republication or correspondence: holistic.detective.leo@gmail.com

House Rules

Evidence is the currency. Analysis here is evidence-based and structurally grounded. If you want to dispute a conclusion, bring counter-evidence. The conversation will be better for it.

Disagreement vs. abuse. Disagreement is welcome; abuse is not. The distinction isn’t complicated. Comments designed to be unpleasant rather than constructive will be removed without discussion. Repeat offenders will be removed without ceremony.

Zero tolerance. Misinformation, harassment, and bad-faith arguments have no place here and will be removed without comment.

Intellectual property. All original content published in The Holistic Detective — including articles, analysis, glossary entries, and series instalments — is the copyright of Leonie Chambers. You are welcome to share and quote this work, provided the source is clearly attributed and a link to the original post is included. Reproduction of substantial portions without permission is not permitted. Licensing enquiries: holistic.detective.leo@gmail.com

The baseline for participation is good faith. Everything else follows from that.

A convention that reviews itself. A mutual legal assistance framework with no mandated timeline. An anti-corruption mechanism whose findings are confidential by design. These aren’t edge cases; they are load-bearing features.

This publication pulls those structures apart — examining the legal architecture behind the headlines, and the exact mechanisms that make these outcomes not just possible, but predictable. Where recent case outcomes illuminate the structural questions, those get covered too. And when a criminal case is simply too strange and revealing to ignore, it goes full true-crime.

Douglas Adams understood that the fundamental interconnectedness of all things was the only honest investigative framework. He was right. He was also considerably funnier about it than I am.

If a piece inspires you to examine how a system actually works — not just the outcome, but the mechanics behind it — that is the right question to ask here, and may be featured in future articles and conversations. Thorough analysis benefits from people thinking out loud.

Thanks for reading, and please abide by the House Rules.

— Leo