Ron Pol details his research revealing that anti-money laundering rules are almost completely ineffective

By Ron Pol*

A new study published in an international peer-reviewed journal extends the anti-money laundering industry’s open secret. Despite nearly three decades, and global ubiquity, money laundering controls (and associated supporting measures) scarcely have the impact of a rounding error on criminal accounts.

New study tests the resplendence of Emperor AML’s cloak

The anti-money laundering industry is based on a series of assumptions, seldom critiqued, let alone tested by practitioners and governments. For example, that money laundering threatens economies, that money laundering controls are effective, that extending anti-money laundering (AML) obligations to new industries will have a big effect, and that a new global ‘effectiveness’ methodology based on specified outcomes is an adequate measure of effectiveness.

The latest study, just published, tests the second and third of those assumptions. A separate study tests the fourth. (Many elements of the first premise have been tested elsewhere in the academic literature).

Not yet in its thirties, the anti-money laundering industry is marked by global reach and prodigious endeavour. Millions of businesses spend billions of dollars each year on compliance staff, risk assessments, consultants, training and software. A core goal is to identify suspicious financial activities; to help detect and dismantle serious profit-motivated crime. The intent is laudable. But, is it working?

Outcome effectiveness

Published in the peer-reviewed Journal of Financial Crime, “Uncomfortable truths? ML=BS and AML=BS2” combines money laundering research with policy effectiveness; disciplines with surprisingly few connections in research and practice.

This line of enquiry is concerned less whether rules exist, or meet specified standards, or whether countries implement the rules, or firms comply with them. Policy or outcomes effectiveness explores whether the rules work. Do they produce intended outcomes? And if not, what might?

In motoring terms, compliance professionals and industry consultants are like driving instructors, helping firms follow traffic rules established by authorities.

Irrespective the differences and complexities in the road rules in each jurisdiction, policy effectiveness specialists are concerned whether firms’ and countries’ AML vehicles can reach their intended destinations. Like mechanics or technicians, outcomes effectiveness analysis examines the AML cars themselves of firms and countries. This involves lifting the hood, conducting diagnostic tests, and calibrating the navigation system. Is it capable of reaching its (crime detection and prevention) destination, or is it more likely to become mired in a series of spaghetti junctions liberally interspersed with the AML industry’s proliferation of toll booths?

The industry seems already to know the answer to that question.

AML industry’s open secret: Crime pays, and serious crime pays seriously well

It’s well known in AML circles that the industry’s huge cost and earnest endeavour has limited effect on criminal finances. Yet the industry narrative continues, seemingly unaffected, churning out much the same model each year, with slight modifications; continuously urging the incremental expansion of AML obligations.

I don’t profess to know why knowledge of what seems to have hallmarks of policy failure has not yet materially affected the industry narrative or practice.

Maybe it has something to do with the fact that simply flipping the telescope produces a more comforting view.

Changing the perspective from ‘crime’ to ‘criminals’ enables officials and politicians to serve a steady stream of good news stories. (Especially in an environment when a non-reflective ‘tough on crime’ mantra prevails, it would be a brave politician to reveal what really lies beneath the ‘tough on crime’ rug).

The ‘good’ view is a positive one. The effect on individual criminals and criminal networks identified by money laundering controls truly is significant. Authorities rightly claim success when suspicious activity reports lead to more arrests and seizures of criminal assets. In many countries, the police do this well. ‘Follow the money’ policing often results in mounting arrest and forfeiture figures. These are, indeed, good results. And they may be sufficient to please a non-reflective ‘tough on crime’ constituency.

But, in policy effectiveness parlance, they are merely ‘output’ measures. Illustrated in the following chart, inputs (such as police salaries), activities (dawn raids) and outputs (arrests and assets seizures) may be necessary precursors towards but are not the ultimate (crime prevention) outcomes intended by good policing.

Looking at broader ‘outcome’ measures, and reflecting on the impact and effect of AML and its associated regulations on criminal finances and criminal activity reveals a less comfortable view.

Uncomfortable truths long hidden in plain sight

Academic warnings about limited effectiveness and poor outcomes began sounding at least as early as 1994 (Gold/Levi/UK Police Foundation appraisal of suspicion-based reporting). In recent times, the United Nations examined what it called the ‘success rate’ of money laundering controls and in 2011 starkly announced that just 0.2% (one fifth of one percent) of criminal funds is successfully intercepted by authorities worldwide.

In 2016, Europol frankly admitted likewise. In Europe, a 1.1% interception rate means that “98.9 percent of estimated criminal profits are not confiscated and remain at the disposal of criminals.” Plus, of course, the accumulated wealth from previous years. Compounded annually.

The newly published study extends this evidence-base, indicating that money laundering controls seem “almost completely ineffective” in disrupting the proceeds and funding of serious crime.

New study uses policy effectiveness lens

Using countries’ own data, the research found interception rates (ie the United Nation’s ‘success rate’, as the proportion of criminal proceeds seized or confiscated by authorities) between a barely perceptible 0.1 percent and 3.3 percent. (And the highest rate, in New Zealand, was based on figures excluding key areas of criminal funds, so its ‘real’ rate will be lower).

Illustrated in the following graph, “the quantum of criminal proceeds interdicted by authorities is little more than a rounding error in the respective country accounts, and globally, of ‘Criminals, Inc’.”


A brief interlude: Addressing the immediate rejoinder

In my experience, this is typically when someone says, “that may be so, but money laundering controls are designed to combat laundering, not crime.”

I usually resist the temptation to point to an uncomfortable dichotomy. It would be churlish to retort that the industry is replete with claims for its crime detection and prevention capacity, except, it seems, when tested.

Instead, I try to address the issue directly. As expressed elsewhere, if the AML/CFT (anti-money laundering countering financing of terrorism) system focuses only on money laundering, and not crime and terrorism prevention, “what is the point of the complexity and expense of a vast global AML/CFT industry?” If, for example, “the number of money laundering prosecutions matter more than genuine outcome measures reflecting their effect and impact on crime and terrorism prevention objectives, the global AML/CFT complex, and the Financial Action Task Force (FATF), arguably serves little purpose.”

That is because, absent the artificial construct of criminalization, money laundering itself is devoid of concepts of good or bad. Anti-money laundering would be largely meaningless but for its capacity to mitigate, reduce and prevent the economic and social harm wrought by serious profit-motivated crime.

So, if we agree that money laundering controls, and FATF, serve a useful purpose, we might return to AML/CFT outcomes.

The least effective policy endeavour, ever, anywhere?

In 2015, having then completed the initial analysis ultimately resulting in the present study (and finding impact figures below one percent), I remarked to a journalist that the modern AML endeavour is arguably “the least effective of any anti-crime measure, anywhere”. In December 2017, that quote was used to open Congressional testimony, and in February in the context of the Mueller investigation. The now-published study is the research those statements were based on. The study itself concludes that “the standard AML/CFT model is arguably a contender for the least effective policy/regulatory/enforcement intervention, ever, anywhere.”

These findings do not criticise the genuine intent of industry participants. In my experience, most practitioners, regulators and policymakers want their efforts to have a significant impact on crime.

Beyond good intent and rules compliance

The industry narrative is marked by good intent and earnest endeavour. Others have noted likewise. For example, Mr Gold and Professor Levi observed that “many citizens, law enforcement officials, and politicians are searching desperately for signs of success in fighting crime, and most of the bankers we have met would like to feel that they are playing a positive part rather than setting up expensive systems for merely ritual compliance with the demands of law.”

Much like mine, their observation is unremarkable. The genuineness of intent is clear, strong, and self-affirming, particularly in banks and government agencies. What is most striking about Mr Gold and Professor Levi’s observation is when it was made. In 1994.

If today’s equally good intent and now globally ubiquitous compliance framework still scarcely has the impact of a rounding error on criminal accounts, it may be timely to test some of the core assumptions underpinning the modern AML framework.

Time to face, and frankly address, the gap between intention and results?

There are signs that practitioners and policymakers are willing to test elements of the industry narrative.

Finding a “lack of clear and compelling evidence…of results” a Canadian Senate committee, for example, concluded that the time for continuous incremental change “is at an end”. The time for “examination of fundamental issues has arrived.”

Professors Levi, Reuter and Halliday recently remarked that the modern anti-money laundering system is “highly cost-inefficient” and has failed “to produce credible evidence of [its] effectiveness.” And the latest study also tests elements of the modern AML system’s effectiveness.

But, the study ends positively. If there is appetite frankly to address the gap between intention and outcomes, the system may not be broken, simply less effective than intended. If so, although FATF’s first iteration of an effectiveness framework may not be all it seems, FATF remains well placed to lead the development of new pathways to the destination intended by the G7 nations, substantially to cut profit motivated crime and its social and economic harms. And potentially enabling the prospect of demonstrably better outcomes from AML rules, beyond compliance with them.


*Former lawyer Ron Pol ( is a legal management consultant and principal at His PhD thesis was supervised by Professor Jason Sharman and is entitled “Effective sentinels or unwitting money launderers? The policy effectiveness of combating illicit financial flows through professional facilitators (lawyers, accountants and real estate agents)”.

There's a fuller author profile here.

The full study

Pol, R. F., Uncomfortable truths? ML=BS and AML=BS2 Journal of Financial Crime, 2018, Vol 25 No 2, pp294-308. NB: Depending on your organization’s access rights to scientific publications, a charge may apply. The author receives no part of any such charge.

*‘Crime pays’ chart

The United States was not part of the study. Data source: FATF/APG Mutual Evaluation: United States (2016), pp5 & 78-79. $4.4-4.6b forfeited in 2014 (pp78-79) divided by estimated $364b proceeds of crime annually from financial crime (2010) and drugs (p5). The former figure fluctuates year by year. The latter excludes criminal proceeds from tax evasion, transnational organized crime, human smuggling and public corruption (domestic and foreign), so the ‘real’ interception rate may be less.

We welcome your help to improve our coverage of this issue. Any examples or experiences to relate? Any links to other news, data or research to shed more light on this? Any insight or views on what might happen next or what should happen next? Any errors to correct?

We welcome your comments below. If you are not already registered, please register to comment or click on the "Register" link below a comment.

Remember we welcome robust, respectful and insightful debate. We don't welcome abusive or defamatory comments and will de-register those repeatedly making such comments. Our current Comment policy is here.


I work in the aml/cft space and I, unfortunately, have to agree with Ron Pol on this. As a professional body we spend a great deal of time and money and the results seem negligible at best. I don't know the answer but what we are doing now isn't working and worldwide billions is spent each year. It would have to be the lest effective policy in history for sure.

I suspect the global war on drugs can probably compete for most completely futile, result free policy. After more than 40 years, probably trillions spent, and same measurable outcomes that best efforts make zero impact on supply.

Sadly have to agree that is another reasonable contender, for reasons even better known than for the AML effectiveness gap. And worse than unknown solutions, known solutions not applied.

More like you NoFax, and we'll have a good conversation in the AML industry and maybe in the political arena. Not that I'm saying I'm 'right' (though I think there's reasonable evidence to have this discussion), but I have had many fascinating if disconcerting discussions with very smart people, equally passionate about the same intended outcomes, opposed even to the concept of testing established norms. With the inevitable knowledge/evidence gaps used to 'disprove' the entire evidence base questioning those norms. I have no idea which might prevail.

Highly interesting, Ron Pol. What is curious is the ability of the US to enforce political/economic sanctions through its effective influence on much of global banking, and yet its inability to employ any such influence when it comes to combatting criminality. That is, the banks can be used to constrain international political adversaries, but not, it appears, international criminal activity.

Perhaps it is simply so much easier to define international political adversaries than to define international financial crime - a distinction seemingly confirmed by the recent UK parliamentary report, 'Moscow's Gold', which, as you'll know, looks at Russian 'dirty money' flowing through the City of London. (In a relevant aside, it is interesting that renewal of Roman Abramovich's UK visa may depend in part on his satisfactorily explaining the sources of his wealth. The same questions, of course, might also be put to The Trump Organisation.)

Only difficult to define of those defining are participants.

Pol is well versed on the issue and has written an academic thesis on the topic .

The problem is he offers no solutions .

My view is that at least we are doing something , and the "obvious" should be easy to recognise, but ascertaining the source of funds coming into a bank account is never and has never been easy .

A good point Boatman. These studies point out effectiveness gaps, not solutions. Seems to me we need to have the conversation about effectiveness, in order to get to solutions. The industry narrative is very, very strong, and even with this knowledge since 1994 or 2011, has not properly confronted even that there is a problem. And btw, I am currently working through what a solution might look like. It's not easy, because it's never been done, nor do I have the luxury of a job allowing me to focus on it fulltime, but it is an interesting challenge.

A couple of questions, Ron Pol. Is it not the case that the only really effective international monitoring and controlling mechanism is SWIFT, the network for country-to-country transfers controlled by the US? And that SWIFT is only effective as long trade and transfers are conducted in US dollars?

We see significant impetus behind new methods of trade and financial transfer, such as cryptocurrencies, and we also see China, Russia and others seeking to introduce or develop non-US dollar trade and transfers. In this environment, perhaps the Western political will is to maintain US/Western banking hegemony, and deal occasionally with its criminal accounts, rather than see it undermined by influencing the growth of largely invisible and adversarial financial platforms.

Crypto is interesting. Often viewed as criminal, due to their early adoption, but the future may see blockchain as part of the solution. Thus, as you say, potential competitor/hegemony threat.

In this space the motivators of participants, policymakers, regulators etc appear many and varied, and definitely affect results. I have taken a simple/strong outcomes effectiveness focus in all, thus enabling assessment against that, namely the capacity for serious profit-motivated crime detection/prevention, and reducing the economic/social harms from crime. Viewed with that lens (even absent any moral assertion of that as the 'correct' measure), the manifold other motivations (or at least their manifestation in results) appear more starkly against such poor outcomes on that measure.

The "Vancouver Model" coined by Professor John Langdale from Macquarie University’s Department of Security Studies and Criminology.

"Capital flight from China, he wrote, is facilitated by high rollers using Canadian casinos, with help from junket operators, resulting in investment in Canadian real estate."

Yep, although the 'model' is not at all unique to Vancouver. But, kudos for good branding. Dutch researchers have long studied ML into real estate. A Columbian model was explored by another, identifying a multiplier effect on prices by criminal funds compared with the ordinary income multiplier effect with legitimate earnings. My own empirical research identified new ways to access and assess underexplored real estate transactions in NZ over the past few decades, identifying how crims used lawyers, accountants and real estate agents.

The collateral damage in this drama is huge and continuing, for all those innumerable honest citizens/residents and businesses who have to run through many hoops and blocks to carry on their legitimate activities. The imposed assumption seems to be 'you are suspect, till you prove your innocence'. But the actual big fish/operators know how to get away, as they usually do.

That is so true SmoKey. I was in the bank a few days ago, able to laugh at tick-box data required of me that I knew from empirical evidence does not stop crims. But for everybody else, its no joke if the barriers apply more to ordinary people than crims. To which some reply that's the price we pay for catching crooks. And many crims are indeed caught by ML controls. But it seems to me that the argument would be stronger if the system was much more effective in substantially reducing serious profit-motivated crime than having scarcely the impact of a rounding error on the national and global accounts of 'Criminals Inc'

So its sole function is as a Toll Booth on ordinary people. How wonderful. A case study in people with good but deluded intentions using computers as a tool to increase complexity and decrease productivity. Essentially we are employing our best and brightest to dig holes and fill them up again. Sounds like the Era of New Communism, dreamed up by Them as Knows Best. What is it Nassim Taleb calls them, "intelligent idiots"?

The last time I reported a property related scheme to the Serious Fraud Office I was given the third degree on who I was and how I got the information. The overall impression given to me is that a.) nobody cares unless there is a political or public (media) narrative; b.) it depends on who is involved in the scheme, as in are there any accountant or lawyer friends (or potential friends), that may be collateral damage should an investigation and prosecution eventuate; and c.) the majority of property related schemes are so blindingly obvious that you really have to conclude that there is an element of willful ignorance.


It's as if these laws are just there to control ordinary people and not really about laundering at all. Just like the five eyes network for example. Is the UK using their intelligence to stop terrorism? Not really - they're using it to send people to jail for "hate" crime.

The end goal is a cashless society. If you're a dissenter "they" will cut you off at the bank account and you won't even be able to purchase a bag of oats.