With rigorous economic research and practical policy solutions, we focus on the issues and institutions that are critical to global development. Explore our core themes and topics to learn more about our work.
In timely and incisive analysis, our experts parse the latest development news and devise practical solutions to new and emerging challenges. Our events convene the top thinkers and doers in global development.
The Working Group on the Unintended Consequences of Anti–Money Laundering Policies examined how rich countries might rebalance their policies to continue to protect against money laundering and terrorism financing without hindering the ability of people from poor countries to conduct business and transfer money across borders. The working group's report provides policy recommendations for multilateral organisations and the US, UK, and Australian governments.
In 2014, migrants sent over $400 billion of remittances home through formal systems and at least an additional $130 billion through informal channels. In addition, businesses in poor countries engage in cross-border transactions to export goods and import key inputs. But banks in rich countries, under pressure from anti–money laundering and counterterrorism enforcement efforts, seem to be exiting entire sectors and regions in a process that has come to be known as "de-risking." A high-profile example of de-risking is the widespread denial of bank accounts to money transfer organisations. Following high-profile de-banking episodes in the UK, US and Australia, in some markets only the very largest players have had access to bank accounts, and many smaller players have been forced to close down or become agents of larger businesses.
What are some of the consequences of these actions? A reduction in competition within the money transfer sector could lead to higher remittance costs. Global remittances (the money migrants send home) are worth at least three times the total amount that is sent as aid to developing countries. Non-profit organisations also report damaging effects of de-risking on their ability to conduct humanitarian aid operations or recruit staff overseas.
Lower-profile but potentially even more significant is the drying up of correspondent banking services. Major banks across rich countries, including in the UK and the US, seem to be less and less willing to operate correspondent banking services for corresponding banks in developing countries. This reduces those banks’ access to the global financial system and damages important cross-border services such as trade financing. Recently, Bank of England governor Mark Carney referred to this worrying trend as "financial abandonment."
In addition to these cost concerns, de-risking may be undermining the effectiveness of the AML system by pushing transactions into less transparent channels. For example, some UK MTOs that were previously using bank transfers are now using bulk currency exchanges, rendering flows of money less transparent than before the de-banking. Similarly, the collapse of simple bilateral correspondent banking relations may mean more complicated routing for trade finance via less transparent jurisdictions like Dubai.
The working group is exploring the extent to which these concerns are legitimate, and determining the most appropriate policy response. The first working group meeting was on January 28 in London, with the second meeting in September in Washington. The final report will be available in November 2015.
Under the international regulatory framework for anti-money laundering and countering the financing of terrorism (AML/CFT), banks are assigned significant responsibilities for detecting and preventing illicit financial flows. These responsibilities include performing due diligence on their customers, monitoring accounts and transactions for suspicious activity, and reporting suspicious activities to the government.
The “de-risking” problem
In recent years, regulators have raised their expectations for what counts as adequate AML/CFT compliance. At the same time, they have cracked down on institutions that have fallen short. While arguably necessary, this more stringent enforcement has produced some unintended side effects. In particular, it has put pressure on banks’ ability and willingness to deliver certain types of services, notably correspondent banking services.
Correspondent banking—the provision of financial services by one bank (the correspondent bank) to another bank (the respondent bank)—is vulnerable to illicit finance abuse. A correspondent bank generally does not have a direct relationship with the respondent bank’s customers. Often, the only information it has access to is the originator and beneficiary information contained in the payments messages themselves. Therefore, it can be a challenge for the correspondent bank to properly assess the illicit finance risk that such transactions pose. While regulators have clarified that as a general rule, banks are not expected to know their customers’ customers (KYCC), many correspondent banks nonetheless find these types of risk difficult to manage in a cost-effective way. In addition, correspondent banking has traditionally been a high-volume, low-margin business.
However, there are two new technologies that may help to solve the problem: big data and machine learning.
Big data refers to datasets that are high volume, high velocity, and high variety. These datasets necessitate different hardware, software, and analytical solutions than do traditional data sets. Banks generate enormous volumes of data in a wide variety of formats. Big data systems can help banks to analyze these data in order to identify abuse while preserving relationships with trustworthy customers.
Big data systems can help compliance staff organize and make sense of large volumes of information. Banks’ compliance staff can utilize data from a wide variety of internal and external sources such as transactions metadata, open-source information (such as negative news stories), and government information (such as sanctions lists, arrest warrants, and so on). Traditionally, these data have been siloed and consequently hard to access. Big data systems can reduce the time compliance staff spend searching for and consolidating information. These systems are typically paired with advanced analytics engines, such as machine learning algorithms, which can help identify patterns and relationships in the data that might have otherwise gone undetected by human investigators.
Machine learning is a type of artificial intelligence which enables computers to learn without being explicitly programmed. There are two broad types of machine learning—supervised learning and unsupervised learning. With supervised learning, the machine learning algorithm analyzes a dataset in order to build a model that predicts a pre-defined output. For example, a supervised machine learning algorithm may be presented with transactions labeled “suspicious” and “not suspicious” and instructed to develop a model that best categorizes transactions as one or the other, based on the available data. With unsupervised learning, the machine learning algorithm explores the features of a dataset, looking for patterns and relationships without attempting to predict a pre-defined output.
Machine learning algorithms are already being used to tackle money laundering and the financing of terror. The application of machine learning to customer segmentation and transactions monitoring has the potential to greatly reduce both false negatives and false positives in identifying suspicious activities. Clustering, a type of unsupervised learning, can be used to develop much more sophisticated customer typologies than traditional methods. This can help banks to gain a better understanding of their customers’ financial behavior. In addition, classification algorithms, a type of supervised learning, can be used to identify suspicious transactions. These algorithms can be trained so that, over time, their accuracy improves. Recently, HSBC has partnered with Silicon Valley-based artificial intelligence firm Ayasdi to automate some of its compliance functions. Another American company, QuantaVerse, is helping several large international banks to fight money laundering and other financial crimes.
In early October, we will be discussing the scope of new technologies to address de-risking at RegTech 2017 in Brooklyn, NY. We will also be publishing a report, Technology Innovations to Address De-Risking, in which we will examine this topic in detail. While there have been many positive actions on the regulatory side, our view is that technologies that have emerged over the past few years present very real opportunities to solve the complex problem of de-risking.
Regulatory pressure on international banks to fight money laundering (ML) and terrorist financing (TF) increased substantially in the past decade. We find countries that have been added to a high-risk greylist face up to a 10% decline in the number of cross border payments received from other jurisdictions, but no change in the number sent. We also find that a greylisted country is more likely to see a decline in payments from other countries with weak AML/CFT institutions. We find limited evidence that these effects manifest in cross border trade or other flows. Given that countries that are placed on these lists tend to be poorer on average, these impacts are likely to be more strongly felt in developing countries.
Distributed ledger technology, like Bitcoin’s blockchain, has the potential to transform cross-border payments, boost financial inclusion, and lessen the unintended consequences of anti-money laundering enforcement. Ripple, a fintech company using distributed ledger technology, made headlines recently, as did the appearance of a new cryptocurrency, Zcash. If you’ve gotten swept up in the enthusiasm around emerging financial technologies (fintech), you may think that the creaking system of international transfers in fiat currencies, and the problems of global financial exclusion associated with it, will soon come to an end. However, as we’ve said before, these innovations may not have as much of an impact as you expect.
Ripple and Zcash are making waves
Ripple has announced that a consortium of 12 large global banks are trialling their currency-neutral clearing system based on a distributed ledger-secured digital asset, XRP, to facilitate cross border payments at the international level. We’ve blogged before about the potential of Ripple’s technology, which could make international transactions cheaper, more secure, and almost instantaneous. This trial is an important step in the right direction and a good example of how banks are seeking to integrate fintech into their existing business models.
The arrival of Zcash is similarly exciting. This bitcoin-inspired cryptocurrency tweaks the Bitcoin model by layering on additional levels of privacy, which should make it more amenable to global financial services companies. To date, banks have been hesitant to embrace the Bitcoin platform in part because it relies on a public ledger that prevents users from keeping information about their trades confidential. Zcash solves this by using ‘zero-knowledge proofs’ that can verify transactions without revealing underlying data. However, the enhanced privacy afforded by Zcash poses a challenge for financial regulators, as would-be criminals seeking to obscure their transactions may also find it appealing.
The global financial system has even bigger challenges
Ripple claims that, by using its system, banks can save ‘up to 60 percent’ of payment processing costs, and clear in less than 5 seconds transactions that used to take days. If the widespread adoption of Ripple as the global payments’ rail were just around the corner, then, this would be a pretty big deal. The increased speed and lower unit cost of international payments would enable more minor transactions to take place across currency borders, potentially creating new types of international businesses, especially in poor countries. Zcash, if it were widely adopted, could have a similar effect.
However, the global payments system faces bigger challenges than high payments processing costs and slow transaction speeds. In a 2015 CGD report, we detailed the way in which ‘de-risking’ by banks is constraining the global financial system, leading to widespread account closures for money transfer and humanitarian organisations, and a decline in correspondent bank relationships. When banks cite their reasons for ending these relationships they tend to emphasize concerns about general profitability and the higher cost of complying with both anti-money laundering and countering the financing of terrorism (AML/CFT) and heightened prudential regulations rather than the speed or cost of processing payments.
Distributed ledger technology could lessen these problems, but it can’t solve them
Ripple’s system can help lower compliance costs by rendering transactions between accounts simple bilateral affairs, thereby allowing financial institutions to bypass the byzantine network of “nested” correspondent banking and foreign exchange dealing relationships. Settling via a different cryptocurrency could do something similar. This doesn’t help much, though, unless banks can be sure that counterparties at the far end of a transaction can be reliably identified. More broadly, global banks must be certain that their client financial institutions have procedures in place to comply with AML, “Know Your Customer,” and sanctions requirements. In too many cases, global banks lack this confidence, particularly in countries where they do not believe financial supervisors are doing their job.
Solving this problem will require political as well as technical solutions. We identified steps towards these solutions in our report, and have been heartened to see progress recently. As we noted in the report, creating an enabling regulatory environment for new technology is an important part of addressing concerns about de-risking, but more difficult political actions will also be required to solve the problem.
Last November, a CGD working group of experts convened to address the unintended consequences of anti-money laundering (AML) policies for poor countries, where they recommended that the Financial Stability Board (FSB) should take the lead on addressing problematic de-risking by banks. Below, we outline our takeaways on the FSB’s progress thus far.
De-risking has potential negative consequences for senders and receivers of remittances, businesspeople operating across borders, and charities working in conflict zones. Given that the FSB’s mandate is to coordinate and review the work of financial standards-setters and regulators, the CGD working group felt that it was particularly well-suited to address the full range of causes of problematic de-risking behavior by banks and the subsequent decline in correspondent banking relationships between rich and poor countries.
The FSB has in fact taken on the challenge. Late last year, it issued an action plan to undertake the following tasks:
Examine the dimensions and implications of the decline in correspondent banking.
Clarify regulatory expectation relating to AML compliance in correspondent banking relationships.
Build domestic capacity in jurisdictions that are home to affected respondent banks.
Strengthen the due diligence tools available to banks.
In March 2016, the FSB established the Correspondent Banking Coordination Group (CBCG), comprised of senior representatives from international organizations and standard setters and national authorities in the FSB and its Regional Consultative Groups, in order to implement its action plan. Last week, it issued a progress update indicating its work in each of the four key areas.
Process is important. The FSB is a key player in shaping the global financial architecture and its involvement in this area is both timely and significant. Its ability to convene the key players at the national and international level has raised the level of dialogue around the problem of de-risking.
The FSB report’s emphasis on four key areas is exactly right. And we feel compelled to point out that they are very similar to our recommendations! In particular, we like the emphasis on data collection and data sharing.
Significant progress has been made since the FSB began working in this area. Recommendations made by us and by the FSB in November 2015 to enable banks to include Legal Entity Identifiers on payment messages are now well in progress. Other big players are also addressing the issue, with new pieces of work especially from the IMF and the US Treasury.
There’s a lot more to be done! Notably, the FSB has talked about the need for systematic data collection and sharing between governments of information relating to the number of correspondent banking relationships between jurisdictions and the types of further customers served by these relationships, like money transfer organizations. So far, there’s been no sign of that sort of systematic data collection and sharing. The emphasis, rather, has been on one-off surveys that are useful right now, but don’t create a system for assessing the effect of AML enforcement on payment flows going forward.
All in all, the FSB is doing a great job of progressing the effort to address the issue of de-risking and the unintended consequences of anti-money laundering. But much more is needed, not just from the FSB and other standards-setters, but also from national governments. That’s why, here at CGD, we are continuing our research and dialogue with key policymakers.
A recent flurry of legislative activity has seen the introduction of four bills that aim to crack down on the financing of terrorism. While it is very important to combat money laundering and the financing of terror, the actions can result in unintended negative consequences for poor countries as well. We like some things in these new bills, but they also leave a lot to be desired.
What are these bills?
In July, Senators Bob Casey (D-PA) and Johnny Isakson (R-GA) introduced legislation (S.3125) creating a new designation for ‘jurisdictions permissive to terrorism financing’ and a set of sanctions to be applied to such jurisdictions. On the other side of the Capitol, three bills have emerged (H.R. 5607, H.R. 5594, and H.R.5469), that are informed by the House Financial Services Committee’s Task Force to Investigate Terrorism Financing. (CGD’s Clay Lowery testified in March to this Committee.) These three bills aim to reorganise how Treasury deals with countering the financing of terror (CFT), require the establishment of a national strategy for CFT, and require the IMF to fund technical assistance to increase CFT capacity, respectively.
What we like
Safeguarding remittance flows: H.R.5607 [sec. 5(a1-2)] proposes that Treasury launch a pilot study to assess whether technical assistance for Money Transfer Organizations (MTOs) servicing the Somali community in the United States enable remittance flows to be more transparent and easily monitored.
Improving and certifying MTO compliance: H.R.5607 [sec. 5(a3)] asks Treasury to study the impact of MTOs sharing their information with depository institutions and credit unions to meet their CFT obligations.
Streamlining bureaucratic complexity: H.R.5594 [sec. 4(a6)] encourages international and intergovernmental cooperation among federal, state, and local entities to combat illicit finance, though there is no description of what this might look like, and no discussion of data sharing.
What we do not like
No evaluation: None of the bills make any mention of an evaluative framework for their legislation. As we documented in our working group report, CFT legislation has had unintended consequences. Continuing to pass laws that pile further layers of complexity onto existing regulations without a framework for evaluating effectiveness and unintended consequences is not the path to an efficient and effective global CFT system. H.R.5594 calls for the new national CFT strategy to include a ‘comprehensive, research-based, long-range, quantifiable discussion of threats, goals, objectives, and priorities.’ This is clearly very useful, but should be accompanied by a request to assess the costs and unintended consequences of the CFT system, including for security. In this way, good, efficient, and effective regulation can be identified and built upon.
No data: In addition to our own, report after report has noted that enhanced international data sharing is essential for a well-targeted CFT system. The United States could and should take the lead on clarifying how that data sharing could take place across the public and private sector while respecting data privacy rights and laws.
What is still needed
A global stock-take:In our report we called for a global stock-take in which the Financial Stability Board, with the support of the governments of the US, UK, and developing country partners like Mexico, takes responsibility for evaluating the unintended consequences of CFT legislation and proposes a path toward a more efficient, effective system. This effort needs to be underpinned by sufficient data collection and data sharing to allow it to be based on fact, not conjecture. We are pleased to note the Financial Stability Board has created a working group to study whether correspondent banking relationships for poor countries are affected by AML/CFT regulations.
Lowering the costs of compliance: H.R.5594 [sec. 4(a14)] calls for an analysis of ‘current and developing ways to leverage technology to improve the effectiveness of the fight against the financing of terror.’ But it should also call for an analysis of how technology can lower the costs of compliance, as these costs partly drive business decisions to terminate services for MTOs and respondent banks in poor countries. The United States should ensure that it maintains a permissive regulatory environment to facilitate experimentation with blockchain and other fintech technologies that could help lower the costs of compliance and create the cheaper, more transparent international payments system needed for financial transactions that will make the United States and the world both safer and more prosperous.
Last November, we released a report on the unintended consequences of anti-money laundering policies for poor countries that focused on remittances, corresponding banking, and the delivery of humanitarian aid. Today, we are pleased to report progress towards reducing the negative, unintended consequences of anti-money laundering (AML) regulation, despite the shadow cast on the international development community by Brexit. One significant policy change from the Financial Action Task Force (FATF) and three new reports give us reasons to celebrate a little, even when there is much work to be done.
CGD research contributes to better FATF Recommendation 8
FATF sets the global standards for anti-money laundering and counters the financing of terrorism (AML/CFT) regulation. FATF’s views are described in a document that lays out 40 Recommendations, which set the standards that guide global regulation. Our report criticized FATF Recommendation 8 and called for it to be changed:
FATF should urgently revise its Recommendation 8 to reflect the fact that non-profit organisations (NPOs) may be vulnerable to terrorist abuse by virtue of their activities, rather than whether they happen to be an NPO or not.
Recently, FATF made precisely this change, clarifying that non-profits are not ‘particularly vulnerable’ to terrorist financing abuse. This will hopefully reduce the chance that non-profits will be excluded from the financial system and should make them better able to deliver humanitarian aid, particularly to conflict zones.
The change to FATF Recommendation 8 demonstrates that sustained advocacy on an issue can be successful—charities and their industry associations, along with think tanks and others, have been lobbying for this change for a long time.
New data and IMF leadership emerge
In addition to the news from FATF, there are three new reports on the unintended consequences of AML/CFT. The IMF has a major new report investigating the widespread withdrawal of correspondent banking accounts—a problem that we also highlighted in our report. The IMF is concerned by the exclusion of developing country banks from the global financial system and issues a call to arms for policymakers. The report lists recommendations very similar to those we championed, including calls for systematic evidence collection that goes beyond surveys, better data sharing, and a more rational application of privacy laws. In a speech to the New York Fed, Christine Lagarde summed up the problem that banks are facing:
Another factor quoted to us has been the considerable uncertainty among banks concerning their regulatory obligations. The possibility of large penalties and reputational risks associated with enforcement on sanctions, tax transparency, and anti-money laundering seems to increase the costs of compliance for global banks in significant ways. Since these are the very issues where regulators have tried to provide clarity, this suggests that both sides still have some work to do in order to reach a better understanding.
The past month has also seen the publication of two more reports about the unintended consequences of AML from the Commonwealth Secretariat and from SWIFT, a global payment messaging provider. Both concern “de-risking,” whereby banks seek to limit their exposure to risk by ceasing activities in a wholesale rather than a case-by-case fashion. As well as demonstrating a growing awareness of de-risking among policy makers and industry representatives, these reports deepen our understanding of the issues. The recommendations from the Commonwealth Secretariat and from SWIFT are consistent with what we propose regarding improving the clarity of guidance to banks.
While the developments of the past month are encouraging, there is much more that needs to be done. Most importantly, ad-hoc surveys and reports are not a sufficient evidence base on which to judge the possible unintended consequences of the global AML regime. We hope that the Financial Stability Board will work with FATF to develop standards for the exchange of information between national Financial Intelligence Units that will enable more rigorous analysis of the state of correspondent banking and money transfer sectors. Here at CGD, we will also be doing more data analysis in the coming year, to provide further insights to policymakers on the impact of AML policies on poor countries.
Late last week, the UK Financial Conduct Authority (FCA) released a report on the drivers and impacts of derisking by banks and other financial institutions. Here at CGD, we have also been focused on the consequences of derisking by financial institutions for money service businesses, non-governmental organizations, and correspondent banking relationships in poor countries. Our report, Unintended Consequences of Anti-Money Laundering Policies for Poor Countries—the product of a working group chaired by Clay Lowery—was released last November. We’re very happy that the FCA report echoes many of our findings. It’s another piece of evidence that suggests regulators are waking up to the fact that banks in wealthy countries are increasingly reluctant to offer services to banks in poor countries and to other businesses that transact with poor countries.
The FCA report successfully underscores the lack of clarity of the risk-based approach. The authors write:
There is…no doubt that banks are trying to do what they believe is expected of them under the risk based approach (RBA) to AML/CFT, in reducing the extent to which their services are abused for financial crime purposes, by on occasion exiting relationships that present too high a perceived risk of such abuse, regardless of the costs of compliance. These perceptions of risk stem from their own judgments, in part reflecting the signals emitted (or judged to be emitted) from the range of regulators and prosecutors who are salient to their institutions, and also the global rankings from the commercial agencies involved in risk judgments.
Moreover, the report notes the lack of an agreed upon assessment of financial crime risk, which has contributed to the phenomenon of derisking:
Achieving the perception of legitimacy and fairness of the regulatory system requires consistency and transparency when dealing with each type of customer. Established risk-based approaches to financial crime identify the risk associated with various factors such as sector, occupation, types of business; geography and jurisdiction risk; political risk; distribution channels; and product or services that customer requires or uses. However, by contrast to some other banking risks like consumer credit loss and fraud risks, there is not yet a generally agreed quantitative assessment methodology for assessing financial crime risk and it is difficult to determine to what extent the data are sufficient for this purpose, other than to make a broad subjective assessment.
The authors also point out the varying ability of banks to “score” their customers:
Banks vary in their ability to ‘score’ particular customers, depending on the bank’s size, resources, geographic coverage and other factors. Decisions on what financial crime residual risks fall within acceptable parameters for a particular bank may be taken through an expression of financial crime risk appetite and/or as an output from customer risk assessment tools, using the broad risk factor categories.
Outputs from customer risk assessment tools will group customers into risk categories (e.g., at the simplest level, High, Medium, Low). De-risking can also come about by setting scores from these tools above which the customer is defined to be beyond financial crime risk appetite, or to require special consideration. Although this would be regarded as ‘case-by-case’ derisking by the banks, it almost inevitably means that the customers identified share common characteristics, such as sector, business type and country affiliations. From the point of view of those affected by derisking, this would give the impression of a wholesale process.
And finally, citing the CGD report, the FCA indicates that its findings are consistent with the prevailing narrative:
Our findings are that the most consistent impacts have been in correspondent banking, where all banks report a net reduction and among MSBs (at some banks). This confirms the narrative found in much of the literature on de-risking, which has tended to focus on correspondent banking, MSBs and charities as sectors at risk.
The FCA has stated that it will work with financial institutions to help them make better assessments of risk, among other improvements.
Recently, my former CGD colleague, Robin Kraft (@robinkraft, now working on satellites in the Bay Area) posted this on Facebook:
Venmo was simply trying to comply with current anti-money laundering policies—the matter was quickly resolved and Robin will soon be able to enjoy the Cuba Caribe dance show.
But for many people in poor countries, the unintended consequences of anti-money laundering policies are no laughing matter. In a recent CGD working group report, we highlight several groups who are suffering the consequences of rich countries’ AML/CFT policies. These include recipients of migrants’ remittances, banks that need a correspondent relationship with a bank in a rich country to process cross-border transactions, and non governmental organizations that are trying to deliver services to war-torn and fragile countries. These groups have seen financial services terminated, become more expensive, or simply less transparent as transactions have been pushed underground. Our report makes five recommendations that may help to address these unintended consequences while preserving the security of rich and poor countries alike.
We can’t say whether the Cuba Caribe dance show was worth the $26.50 Robin paid for the ticket, but we are pretty sure our report recommendations will help developing countries get the best out of the money sent back in remittances.
Other recent studies have found relations between banks and certain nongovernmental organizations have grown tense.
“Even large international NGOs are sometimes having trouble operating bank accounts if they’re trying to deliver aid to Syria, Afghanistan or Pakistan,” areas of particularly high risk for terror financing, said Vijaya Ramachandran, an expert with the Center for Global Development, a Washington-based think tank that produced a report on the subject in late 2015.
In February of this year, more than 50 nonprofits asked the U.S. Treasury to publicly affirm that nonprofit organizations aren’t inherently high risk.
Read full article here.