In late 2020, anticorruption and transparency advocates scored a major victory: the passage of the U.S. Corporate Transparency Act (CTA), which requires U.S. corporations, limited liability companies, and “other similar entities” to disclose the identities of their true beneficial owners to the U.S. Treasury Department’s Financial Crimes Enforcement Network (FinCEN). FinCEN is currently in the process of drafting regulations to implement the CTA. One of the key questions FinCEN is considering concerns the scope of the CTA’s coverage—in particular whether trusts should be considered “similar entities” to which the CTA’s disclosure obligations apply.
The answer ought to be a resounding yes. As the recent revelations from the International Consortium of Investigative Journalists (ICIJ) stories on the so-called Pandora Papers has made all too clear, trusts are prime vehicles for kleptocrats, organized crime groups, and others who want to hide their illicit assets. To be sure, trusts have legitimate uses, such as estate planning, charitable giving, and certain (lawful) strategic business purposes. But the potential for abuse means that it is essential to increase transparency and oversight of trusts.
To understand the role that trusts play in illicit finance, and why application of the CTA’s disclosure requirements to trusts is so crucial, it’s important first to understand the basics of how trusts operate. At the most fundamental level, a trust involves three parties: (1) the settlor creates the trust and provides its assets (which might be money, real estate, shares in a company, or something else); (2) the trustee holds legal title to the trust and is responsible for managing its assets; (3) the beneficiary receives the benefits of the trust, often though not always in the form of periodic distributions of income. For example, a grandmother (the settlor) might set up a trust, with her son (the trustee) managing the trust’s assets for the benefit of her grandchild (the beneficiary), who would receive an allowance every month.
Unfortunately, the trust mechanism is also attractive to would-be money launderers. Trusts are not only extremely flexible, but they also offer a substantial degree of secrecy. For one thing, in contrast to corporations and limited liability companies, the formation of a trust typically does not require filing incorporation documents with the state. Even without the CTA’s beneficial ownership disclosure requirement, the incorporation process typically reveals useful information about a company, including the (alleged) purpose of its business, its location, and its legal address. By contrast, a trust is a contractual agreement between the settlor and the trustee, and the formation of a trust in most jurisdictions generally does not require the filing of any documents with the state or indeed any action by the government.
Furthermore, the legal separation between ownership of the trust (which rests in the trustee) and the trust’s settlor and beneficiaries creates an additional barrier to figuring out who really created, controls, or benefits from a trust. Even when a trustee’s identity is available, trustees may be bound by confidentiality arrangements—which some jurisdictions permit—that bar the trustee from revealing the trust’s settlors or beneficiaries. This secrecy problem is exacerbated when the trustee itself is a company: Because some trust companies act as trustees for hundreds or even thousands of trusts, it can be difficult or impossible to infer a trust’s settlor or beneficiary from the identity of the trustee. Similarly, even when it is possible to identify a trust’s beneficiary, that beneficiary might be another corporate vehicle—possibly another trust—which adds another layer of opacity to the trust’s ownership structure. And if the trust is created by an attorney, the attorney-client privilege creates an additional barrier to transparency.
To illustrate why trusts are an attractive tool for money laundering, consider a stylized example in which a politician (P) wants to solicit a $10 million bribe from a construction company (C) in return for ensuring that the company receives a lucrative government contract. How would P and C arrange this transaction? The simplest way would be for C to pay P in cash, but that would be difficult given the large sum involved. C could transfer the money to P’s bank account, but the banks, which are subject to suspicious activity reporting requirements, would likely flag the transaction, and might even decline to process it. Setting up a trust is an attractive alternative: P can first contact a trust company to establish a trust (T), for which the trust company will act as trustee. P might be the settlor, but P might also have the trust company supply a “dummy settlor”—someone within the company willing to act as the settlor of the trust. (P might also make use of certain trust arrangements—such as use of a protector or a letter of wishes—that allow P to maintain a degree of control over how T’’s assets are used, without claiming legal ownership.) The beneficiaries of this trust would be P’s spouse and children. The assets of this trust would be shares in a dummy company (D) set up by P (or her associates). Meanwhile, C sets up a special purpose vehicle (S) to hold the $10 million. (As far as an outside observer can tell, these assets were generated by C’s past business transactions.) C then executes a phony “consulting” contract with D (the shares of which are held by T), and C arranges to transfer funds from S to D. This consultancy contract appears legitimate, so C’s bank transfers the funds. But transferring the money to D is actually a transfer of that money to T, which owns all the shares in D. And because P’s family are the beneficiaries of T, they can receive that money as distributions from the T, though those payouts could be delayed, and T could have many other assets in addition to shares in D.
This arrangement makes it easier for P and C to disguise the bribe and to launder the illicit payments. Even if law enforcement or other investigators suspect some form of wrongdoing, the secrecy associated with trusts can make it more difficult to follow the money—especially since, as noted above, some jurisdictions offer secrecy protections that make it immensely difficult and sometimes practically impossible to trace the proceeds given to the beneficiaries to the person who really controls the trust. Connecting the $10 million paid by C to the assets received from the trust by P’s family members is even more difficult if the trust company offers to provide a “dummy settlor,” and if the jurisdiction where T is formed allows the identity of the beneficiaries to be kept secret—unknown to anyone but the trustee and the settlor—until ownership of the assets is transferred to them. In short, the secrecy associated with trusts allows the settlor to benefit from illicit funds while obscuring her personal connection to those assets.
In order to prevent the use of trusts to facilitate criminal activities, accurate and adequate beneficial ownership information must be made readily available to government agencies. A trust’s beneficial owners ought to be understood as all natural persons who exert substantial influence over the trust and/or enjoy its benefits, regardless of their technical legal status (settlor, trustee, beneficiary, or none of the above). In the above example, if the trust company, acting as trustee, were obligated to collect and report information on T’s beneficial owner, then investigators would know that P is T’s beneficial owner, and this would make it substantially easier for government investigators to trace the initial bribe payment (from S to D) back to P. (Doing so would be even easier if D is also covered by beneficial ownership reporting requirements, as P would be the beneficial owner of D, in that P is the beneficial owner of T, which owns and therefore controls D.)
FinCEN should therefore follow the Financial Action Task Force’s recommendation that governments require every trustee to obtain the identities of settlor, other trustee(s), and beneficiaries—and, if any of these are legal entities, the identities of the actual human beings who are the true beneficial owners of those entities—as well as any other person who exercises ultimate effective control over the trust. (The EU’s 5th Anti Money Laundering Directive embraces a version of this approach, requiring every party to the trust, as well as any other individual with control, to be registered as a beneficial owner.) In the context of the rulemaking on implementing the CTA, this means that FinCEN should explicitly define “other similar entities” as including trusts, and to treat the beneficial owners of a trust’s settlors, trustees, and beneficiaries all as counting as beneficial owners of the trust itself. And FinCEN should not limit the application of the CTA only to those entities that must register or file documentation with a state or tribal chartering authority (as some commenters have urged, see here and here), because, as noted above, many jurisdictions don’t require the filing of documentation to create a trust.
The purpose of the CTA is to ensure the government can identify individuals using legal entities to conduct illicit activities within the United States. As the Pandora Papers has made abundantly clear, trusts are one such legal entity. FinCEN has the opportunity—and obligation—to implement specific and defined regulations in order to advance the CTA’s core objectives. Kleptocrats (and their lawyers and accountants) are sophisticated and will scour the details of CTA and FinCEN’s regulations in search of potential workarounds. Leaving “other similar entities” either undefined or without explicitly covering trusts (as well as other entities susceptible to abuse, like partnerships and foundations) would therefore be a grave error.
Thanks Devon, great and timely post! I really like your clear illustration of how trusts operate and how they can be misused to hide criminal proceeds. I’m curious if there are any good (or bad) arguments for why we should not include trust in the CTA, or if there would be any practical concerns that might hinder this recommended reform?
Thanks Emma for your question! From my understanding, perhaps one of the biggest reasons for not including trusts under similar entities is the fact that, as mentioned above, trusts are considered to be contractual agreements between the trustee and the settlor and therefore there is technically no “beneficial owner.” In fact, in FinCEN’s final rule on customer due diligence in 2016, FinCEN noted specifically that trusts were not included because identifying a beneficial owner would be impossible. I personally don’t think this argument has much merit, given that various international jurisdictions and organizations have designed a beneficial ownership definition for trusts.
Devon, it is amazing how you clearly described trusts and their function when it comes to receiving bribes and laundering money.
Following up Emma’s question, I am curious about the legislative history of CTA. Do you think Congress was willing to include trusts when the statute says “other similar entities”? About the interpretation of the CTA, which arguments do you think we could use to argue that trusts are similarly situated as corporations and limited liability companies?
Great post, Devon! Your post did a great job of explaining the potential problem with trusts and how expanding the definition of “other covered entities” in the CTA could help alleviate the problem. Do you anticipate other states besides South Dakota adopting similarly secretive trust laws and whether, if FinCEN fails to adopt your proposed broader definition, we may see a ‘race to the bottom’ with trust law in the U.S.? One additional question: the ICIJ report indicated that there were 206 trusts linked to 41 countries holding collectively, over $1 billion in assets, but that less than 30 were linked to individuals accused of bribery, fraud or human rights abuses. Ignoring the fact that individuals not accused could still be guilty, are there specific “touchstones” that these suspicious trusts exhibit that might be targeted? If only a small fraction of trusts are problematic, I, like Emma, wonder if this regulation could impose overly burdensome compliance costs.
The data from ICIJ comes primarily from only one of the universe of trusts, even in North Dakota, as I understand it from reading their article, namely documents from Trident. https://www.icij.org/investigations/pandora-papers/us-trusts-offshore-south-dakota-tax-havens/
as well as others identified in this set of leaks.
And in general all the ICIJ leaks reports, however big each one is, cover a very small fraction of the relevant universe of companies or UHNW individuals involved in the transactions revealed.
There is no way to extrapolate convincingly to the full set of trusts either in North Dakota or in the United States. Any judgement about overly burdensome should be compared to the potential value of adequate transparency through compliance with the regulations. And without at least basic transparency regulations, it would seem that there is no good possibility of obtaining good estimates of even the cost of compliance or the benefits of regulation, unless one could compare North Dakota to a state where there is a positive model of regulation.
Anyone among the readers who is a specialist on trusts who can point to a tax jurisdiction within the United States or anywhere else in the world where there is a good model for regulation of trusts?
Great post, Devon. It shows in a didactic way how trusts can be misused as a perfect workaround to circumvent government controls.
In your opinion, which interest groups would oppose to the idea of implementing transparency requirements on trusts?
I’m also curious about this question! While I can imagine that industry professionals might be opposed to increased regulation, it seems like opposition to these transparency requirements would be significantly harder post-Pandora Papers. Have there been groups that have consistently come out against such recommendations in the past?
Thanks Rafael and Mayze for your comments and questions! I was quite surprised in my research how little opposition there is about including trusts in the definition of similar entities. The American Bankers Association, for example, argued that the similar entity requirement should follow the CDD 2016 rule (which did not include non-statutory trusts), but mostly for consistency reasons. As I noted above to Emma’s comment, FinCEN might be hesitant to expand the similar entities definition to include trusts, given the fact that they explicitly excluded trusts from the CDD 2016 Final Rule (since they argued that figuring out trust beneficial ownership information is impossible).
I would imagine that much of the criticism would come from, as you mention Mayze, industry professionals, specifically trust companies. Small business owners initially opposed the CTA generally because of increased compliance costs. I also wonder whether trust lobbyists in certain states would oppose this, given how much of an impact they’ve had in places like South Dakota and New Hampshire in transforming trusts laws to attract wealth to their states. I would imagine they, too, would oppose increased disclosure requirements. It is certainly noteworthy to me how little I’ve seen on outward substantive criticisms of covering trusts within the CTA.
Thanks Devon. I think you do a terrific job of showing how trusts make it more difficult to follow the proceeds of corruption. It’s not that you can’t tease out the relationships if you really understand who the settlor and the beneficiary are, or if they have the same last name, but often the relationship between the parties might not be obvious. Only if the bank or trust company really investigates the parties can they figure out where the money has come from and where it’s going. Covering trusts with the CTO would make it easier to figure out who really owns the hypothetical dummy corp. in your example, rather than relying on the paperwork provided by the trust company, which may not even have the true settlor of the trust on the paperwork! If the information in the registry is accurate, it’d also have the benefits of lowering the costs of AML compliance because there’d be a lot less duplication of effort in trying to identify settlors and beneficiaries.
Such a fascinating post and a really effective illustration of the mind-boggling labyrinth of trust mechanics! It also really makes me wonder about the history of trusts. How did that common law property right develop in such a way that fosters so much undesirable secrecy? Or perhaps the secrecy aspect is a more recent development, especially with recent AML efforts?
Regardless, I’ve always been fascinated with how trusts have been employed in creative ways to achieve various ends beyond what we would normally expect (often for corrupt ends, as you illustrate). A couple of years ago, I did legal services work in Philadelphia where I would occasionally come across these really unusual arrangements that we called “South Philly Trusts.” With these quasi-trusts, one individual acts as both the grantor and the trustee, with full control over the property for life, including the ability to sell and mortgage the property, which distinguishes it from a life estate. The settlors typically named their family members as the beneficiaries, creating a sort of will substitute. In my office, one of the theories was that these South Philly Trusts were used to protect property from the Mafia. But it seems just as likely they were a tool of convenience to avoid probate and all of the related taxes/fees/headaches!
Devon, you’ve done an excellent job portraying the complexities of money laundering through trust schemes. Although you make a strong argument as to why FinCEN should include trusts in the CTA, I do wonder if this would create excessive overlap in compliance costs. Given that most Trusts hold bank accounts either directly or indirectly through underlying companies, isn’t the reporting obligation of the UBO still covered under the CTA through financial institutions? Also would your reporting recommendation account for Trusts that have foreign corporate or individual Trustees?