OECD Releases Updated CRS Commentary and FAQs (Part 2)

Peter Cotorceanu Blog, CRS Commentary and FAQs, OECD

As described in our recent blogpost, OECD Releases Updated CRS Commentary and FAQs (Part 1), the OECD announced the release of a series of new FAQs on 6 April. The prior blogpost examined the ones of particular relevance to the fiduciary industry. In this blogpost we summarize the remaining nine FAQs. These other FAQs cover a variety of topics, spanning the niche to the general, as set forth in summary below, segregated per section in the FAQs:

  • Section I: General Reporting Requirements
  • Question 10: Both the owner and the beneficiary of a usufrucht (a type of arrangement available under certain civil law codes) should be treated as joint account holders and Controlling Persons of a trust for due diligence and reporting purposes.
  • Sections II-VII: Due Diligence Requirements
  • Question 6: If an FI must collect additional information on account holders due to revisions in local AML/ KYC rules, such information must be considered for CRS change-in-circumstance purposes to the extent it conflicts with existing classifications, consistent with the current change-in-circumstance rules.
  • Question 23: Addressing the specific context of trust-like collective investment vehicles (CIVs), look-through requirements for the identification of Controlling Persons need not exceed local KYC/AML requirements (provided that reliance on such KYC/AML requirements is otherwise valid).
  • Question 24: Where an FI elects to apply new account procedures to a population of pre-existing account holders, a formal Relationship Manager (RM) inquiry is no longer required. To the extent, however, that the RM knows or has reason to know that information provided on the self-certification form is inaccurate, the self-certification form cannot be validated.
  • Section VIII: Definitions
  • Question A4: An entity classified as an Investment Entity type FI under a FATCA Model 1 IGA may not automatically adopt this classification for CRS purposes without applying the CRS classification rules themselves to see that they lead to the same result.
  • Question A8: E-money providers do not qualify as per se Depository Institution type FIs, but need to qualify under the generally applicable entity classification principles.
  • Question C7: E-money accounts are not entitled to special privileges for classification as low-risk excluded accounts. Under an implementing jurisdiction’s local CRS laws and regulations, they should be afforded such status only if they are properly understood as unlikely vehicles for tax evasion.
  • Question C8: In the case of publicly-traded CIVs organized as trusts, the equity interest owners are the registered unit holders (even in circumstances where such registered unit holders may be acting as intermediaries).
  • Question D4: In order to qualify as a publicly-traded Active NFE (or as an Active NFE related to a publicly-traded entity), the publicly-traded entity must be formed as a corporation because the term “stock” will not cover other forms of equity interests, such as publicly-traded partnership interests, for example.

OECD Releases Updated CRS Commentary and FAQs (Part 1)

Peter Cotorceanu Blog, CRS Commentary and FAQs, OECD

Following months of rumor and prognostication about a materially amended Commentary to the Common Reporting Standard (CRS Commentary), the OECD released the second edition of the Standard for Automatic Exchange of Financial Account Information in Tax Matters on 6 April, available here. In contrast to the predictions of forthcoming seismic changes, the actual changes (at least in current form) are of minimal import to the fiduciary industry (and likely other industries), limited to the following alterations to reporting logistics:

  • Expansion of the final part on the CRS XML Schema User Guide;
  • Provision of additional technical guidance on the handling of corrections and cancellations within the CRS XML Schema; and
  • Revision to and development of the existing set of reporting correction examples.

The other sections remain substantively unchanged from the first edition of the CRS Commentary issued in 2014.

Of further interest, nothing indicates that further substantive edits to the CRS Commentary are imminent. Accordingly, the OECD may prefer to modify CRS rules (or the interpretation of CRS rules) by means of revisions to the CRS FAQs or to its Handbook (readers of this blog though know well my position on the local enforceability of certain key provisions in the Handbook).

As preliminary evidence that the former approach may be favoured, simultaneous with the release of the updated Commentary, the OECD announced the release of a dozen new CRS FAQs, available here.

Amongst these FAQs, the following stand out for their relevance to the fiduciary industry:

  • Question 5, Sections II-VII: Due Diligence Requirements, resolves some lingering (or rather perhaps hopeful) uncertainty around the need to identify Controlling Persons who are indirectly holding their interests in a passive NFE via one or more Reporting FI intermediaries. In response, the OECD makes plain that there are no so-called “blocker entities” for these purposes: The CRS status of any intermediary entities between a qualifying Controlling Person and the passive NFE for which such persons must be identified is irrelevant.
  • For example, a trust that is (or is treated as) a passive NFE will have a corporate settlor owner looked-through for Controlling Persons, irrespective of such corporate settlor’s CRS status.
  • Such an outcome was not unexpected. Proponents of the alternative interpretation might contend that the Controlling Persons are invariably financial account holders of the Reporting FI and thus will be reported by such Reporting FI with far more accurate and useful financial information. Presumably, however, the OECD found the counter-arguments more persuasive, namely the risk of non-reporting due to different reporting jurisdictions and an expectation of more reliable reporting from the banks looking through the passive NFE account holder than from the generally less sophisticated entities owning them.
  • Question 25, Sections II-VII: Due Diligence Requirements, addresses the subject of individual account holders without a tax residence (commonly referred to a “perpetual travellers”). The CRS Commentary seemingly mandates that an individual self-certification form must contain a jurisdiction of tax residence. The inflexibility of this requirement generated an irreconcilable conflict for financial account holders with no tax residence, who evidently could not be documented for CRS purposes. In response, the OECD adopted a common-sense approach: Treat self-certification forms missing a tax residence as unreasonable and thus presumptively invalid, but permit the account holder to rebut the presumption by providing a reasonable explanation and documentation, as appropriate. In my experiences, some banks had already devised a similar approach by requiring written confirmation from a bona fide specialist that the individual account holder in question is indeed not tax resident anywhere.
  • Question C9, Section VIII: Definitions, confirms that Investment Entities that maintain no financial accounts, but nonetheless qualify as FIs due to advisory and/or asset management services, will not need to treat their equity and debt interest holders (i.e. their owners and creditors) as financial account holders unless such accounts are intended to thwart CRS reporting.

Next week we’ll look at the other new OECD FAQs. To quote Bob Dylan: “the times they are a-changing.”

CRS and FATCA expert

URGENT: Do Fiduciaries Have to Register Their FATCA Sponsored Entities by 1 January 2017?

Peter Cotorceanu Blog, FATCA, FATCA Sponsored Entities

Many fiduciaries have recently received letters from banks and other financial institutions about their sponsored entities under FATCA. These letters generally take one of two forms. The first either implies or outright states—incorrectly—that all sponsored entities must register with the IRS and obtain their own Global Intermediary Identification Numbers (GIINs) by 1 January 2017. The second type of letter says that sponsored entities may have to register and get their own GIINs by that deadline. Whilst this second type of letter is accurate, it is unhelpful because it doesn’t say which sponsored entities have to register and which don’t.

Having received these letters, lots of fiduciaries are now scratching their heads and wondering whether they need to register all their sponsored entities and, if not, which ones. This blog clarifies all. If you can’t wait, click here to scroll down to the Table, which has the bottom-line answers.

First, though, a bit of background.

With rare exceptions, entities that are “Foreign Financial Institutions” (FFIs) under FATCA must either (i) register with the IRS and, as a result of that process, obtain their own GIIN, or (ii) be “sponsored” by another entity. Being sponsored in essence means that all of the sponsored FFI’s FATCA due diligence, reporting, and other obligations are fulfilled by the sponsoring entity on the sponsored FFI’s behalf.

A sponsor must meet certain requirements, amongst which is registering as a sponsor on the IRS website and getting a special sponsor’s GIIN. You can always tell a sponsor’s GIIN because it has the letters “SP” in it. This registration is in addition to any other registration the sponsoring entity may be required to make in its own separate non-sponsor’s capacity, e.g., as a Participating FFI, Reporting Model 1 or Model 2 FFI, etc. Thus, most sponsors will have registered twice and received two different GIINs: Their own regular GIIN and their sponsor’s GIIN with an “SP” in it.

A W-8BEN-E, W-8IMY, or other FATCA self-certification form given by an FFI to another FFI where the former holds an account must include a GIIN. Which GIIN depends on whether the FFI providing the form is sponsored or not. Until now, all sponsored entities have been able to put their sponsor’s GIIN on the form. This will change come 1 January for certain sponsored entities. After the New Year, sponsored entities that do have to register and get their own GIINs (see the below Table – click here) will have to put those GIINs on the forms. In addition, if they have ticked the box on the old (February 2014) version of the W-8BEN-E saying “Sponsored FFI that has not obtained a GIIN”, they will have to file the current (April 2016) version of W-8BEN-E and tick the box that says simply “Sponsored FFI”. (Note: This category is only applicable to Sponsored Investment Entities not covered by an IGA—it does not apply to Sponsored Investment Entities covered by an IGA or to any Sponsored Closely Held Investment Vehicles, whether covered by an IGA or not.)

Although registering and getting a GIIN is a relatively simple process, multiply that process over perhaps hundreds or even thousands of FFI client entities and, well, you get the picture–registration can be a real pain. All other things being equal, then, it is better to have one or two entities register as sponsors rather than have your whole portfolio of client structures register separately.

So what exactly is changing on 1 January 2017? As mentioned, some have been wrongly told that all sponsored FFIs must register with the IRS and get their own GIINs by that date. However, the truth is that only certain types of sponsored FFIs must do so. Knowing the difference is key. Only some Sponsored Investment Entities (SIEs) have to register and get GIINs by 1 January, specifically:

  • SIEs not covered by an IGA
  • SIEs covered by a Model 2 IGA, and
  • SIEs covered by a Model 1 IGA, but only if and when they identify a “U.S. Reportable Account”

(The official list of all IGA countries indicating which are Model 1s and which are Model 2s is available here.)

In contrast, the other type of sponsored entity, a so-called “Sponsored Closely Held Investment Vehicle” (SCHIV), never has to register or get its own GIIN, not now and not beginning 1 January. This is regardless of whether the SCHIV is in an IGA country and, if so, whether the IGA is Model 1 or Model 2, and regardless of whether the SCHIV identifies a U.S. Reportable Account.

Similarly, Trustee-Documented Trusts never have to register or get their own GIINs. Although TDTs are not technically “sponsored” as such, they nevertheless share many similarities with SIEs and SCHIVs. One is that another entity (in the case of TDTs, the trustee) fulfils all of the “sponsored” entity’s (i.e., the trust’s) FATCA obligations on the sponsored entity’s behalf.

A TDT puts its trustee’s GIIN on its W-8BEN-E or other self-certification. Weirdly, though, the IRS instructions for the W-8BEN-E in effect say that a TDT should put its trustee’s regular GIIN, not its trustee’s sponsor’s GIIN, on the W-8BEN-E. This instruction is the exact opposite of what’s required for SIEs and SCHIVs. It makes no sense at all. After all, why require a Trustee of a TDT to register as a sponsor and get a sponsor’s GIIN in the first place, something the rules clearly require, if that GIIN isn’t going to be used on its W-8BEN-E? But that’s what instructions say to do.

Another difference between TDTs, on the one hand, and SCHIVs and SIEs, on the other, is that TDTs are only allowed under IGAs. A trust not covered by an IGA cannot be a TDT because TDTs are not provided for in the U.S. FATCA Regulations, which are what govern in the absence of an applicable IGA.

The following table summarises the registration requirements for SIEs, SCHIVs, and TDTs as of 1 January 2017.

Type of SponsorshipRequired to Register and Get Own GIIN by 1 January 2017?
SIE Not Covered by an IGAYes
SIE Covered by a Model 2 IGAYes
SIE Covered by a Model 1 IGANo unless and until it identifies a “U.S. Reportable Account”
Trustee-Documented TrustNo

Whilst the above rules about which sponsored entities must register and get their own GIINs by 1 January are simple enough, they do raise a number of questions. Amongst these questions are the following:

  • What’s the difference between SIEs and SCHIVs?
  • Can an SIE switch to a SCHIV or TDT?
  • What is meant by a “U.S. Reportable Account”?
  • What if your country’s IGA doesn’t include SIEs and SCHIVs?

All of these questions are answered on gatcaandtrusts.com. Free 14-day trial subscriptions are available here. Once you’ve signed up, you’ll find the answers to these questions—and much more—under Subscriber Resources (the answers to these questions are in the Subscriber Resources under “Sponsorship” in the FATCA materials—scroll down the page till you see FATCA Sponsorship FAQs).

Paid Subscribers to gatcaandtrusts.com also have access to downloadable (i) sponsorship agreements, (ii) corporate sponsorship resolutions that work for both SIEs and SCHIVs, and (iii) sample W-8BEN-Es and W-8IMYs for all categories of entities, both sponsored and non-sponsored (the latter are found under “Self Certifications”). Coming soon is a Table showing the pluses and minuses of SIEs, SCHIVs, and TDTs, as well as so-called ODFFIs (Owner-Documented FFIs) and recommendations as to which status to choose depending on whether the FFI is in an IGA country (if so, Model 1 or Model 2), is an underlying or parent entity, has outstanding loans, etc.

Why, and When, the OECD’s CRS Implementation Handbook and FAQs Should be Ignored.

Peter Cotorceanu Blog, CRS, OECD

In my last blog, I showed why the OECD’s FAQ (Available here) saying that all Protectors are “Account Holders” in FI trusts doesn’t make sense. Worse than that: Why that FAQ is flatly wrong and directly contradicts the OECD’s very own definition of “Account Holders” in the Common Reporting Standard itself. I won’t re-tread that discussion here. If you haven’t read that blog, I’d encourage you to do so. (You can find it here)

Despite how dead wrong this FAQ is, many fiduciaries feel compelled to follow it. They also feel compelled to follow all the other CRS FAQs plus the CRS Implementation Handbook (available here). After all, or so the theory goes, the OECD came up with CRS, FIs in CRS participating jurisdictions have been mandated to implement CRS, and the FAQs and Implementation Handbook are part of that regime. End of story, no?


With rare exceptions, fiduciaries are not required to follow the OECD’s FAQs and CRS Implementation Handbook. I’ll go even further: With rare exceptions, fiduciaries are not even allowed to follow any parts of the OECD’s FAQs and CRS Implementation Handbook that directly conflict with the Standard for Automatic Exchange of Financial Account Information in Tax Matters (available here). (For simplicity’s sake, I’m going to refer to this latter book in the rest of this blog as the “Standard” to avoid confusion with the “true” Common Reporting Standard, which is but 32 pages of that 300-plus-page document.)

This is an extremely important point. It goes far beyond the narrow issue of how Protectors of FI trusts should be treated. It goes to the very heart of the OECD’s legal authority. That authority is exactly—wait for it–zero.

There is a tendency to think of the OECD as a sort of super-legislature imposing laws from on high on vassal states. Many fiduciaries (and not a few advisors) seem to view the OECD in this light. The OECD says jump; people ask “How high?”. The OECD FAQ says all Protectors are Account Holders in FI trusts? Well, then they are.

If you’re a trust company taking this approach, you’ll report all Protectors and, if you’re an advisor taking this approach, you’ll advise trust companies to report all Protectors. Except you’d be wrong to do so in most countries. Going along to get along may seem like the safest course. It’s not. Quite the contrary: It could land you in a world of hurt. Here’s why.

The OECD is not a government. It can’t make rules. All it can do is propose them. Every sovereign nation in the world is then free to accept or reject the OECD’s proposals as it sees fit. To be sure, most countries feel enormous pressure to adopt the Standard, and over 100 countries have now done so. How? By enacting all or part of that document into local legislation and regulations. It is that local implementing law—and that law alone–that binds fiduciaries. Not the Standard itself, but the Standard as implemented locally.

Naturally, CRS implementing rules differ from country to country. However, they’re broadly similar when it comes to the mechanism by which they incorporate CRS into local law. Whether contained in legislation or regulations, CRS implementing rules typically refer to the “Common Reporting Standard” and define it to mean the Standard, i.e., the OECD’s 300-plus-page book mentioned earlier. The legislation or regulations will then typically incorporate parts of that document into local law by reference or by repeating large swathes of it verbatim.

The CRS implementing rules in some countries even purport to incorporate any future amendments of CRS into local law as well. For example, the Canada and BVI legislation do this. I am not a constitutional lawyer, but I question the constitutionality of allowing an external body such as the OECD to amend local law in this fashion. It’s one thing for a legislature to incorporate by reference an existing external document into local law. An existing document is a known quantity. However, it’s another matter entirely to say that any future amendments to an existing document will automatically become local law as well. That amounts to giving an unelected external body carte blanche to change local law as it sees fit. If that’s not an unconstitutional delegation of sovereign legislative authority, I don’t know what is. Sure smells bad to me. But I’ll leave that debate to the constitutional experts.

Constitutional issues aside, no CRS implementing legislation or regulations that I have seen so much as mention the FAQs or Implementation Handbook. Might those documents nonetheless be incorporated through the back door in countries like Canada and the BVI that incorporate CRS as amended from time to time? Not in my view. Neither document should be seen as an amendment to CRS. FAQs are meant to explain open questions, not change settled ones. To the extent they contradict the document they purport to explain, they should be treated as a misinterpretation, not an amendment. As for the CRS Implementation Handbook, it isn’t even addressed to fiduciaries. On the contrary, the Handbook itself says that its purpose is to “assist government officials in the implementation of the Standard . . .”. The Handbook goes on to say that it is to “assist in the understanding and implementation of the Standard and should not be seen as supplementing or expanding on the Standard itself.” Straight from the horse’s mouth.

So, if a fiduciary is to follow local law, it will follow whatever documents local law says to follow, which will generally be limited to Standard itself. A fiduciary may also choose to follow an FAQ or a part of the Implementation Handbook if those documents merely explain finer points of CRS. But in no circumstances should a fiduciary follow an FAQ–like the one on Protectors—or any part of the Implementation Handbook that directly contradicts CRS unless local law tells them to do so, which will rarely if ever be the case. If they do, they will be elevating the OECD’s views above local law. I have heard one person defend this positon by calling the Implementation Handbook and FAQs “soft” law. Let’s be very clear about one thing: There is no such thing as “soft” law. There is “law” and “not law”. The OECD’s Standard, Implementation Handbook, and FAQs are “not law”. They are only “law” to the extent local implementing rules make them such.

What about local CRS Guidance Notes? Several refer to the FAQs and Implementation Handbook and hint or even purport to mandate that FIs should follow them. For example:

  • The Cayman Islands Guidance Notes (12 April 2016 version) says that FIs should “consult” the FAQs and Implementation Handbook
  • The Guernsey Guidance Notes (1 March 2016 version) say that they (i.e., the Guernsey Guidance Notes themselves) are “supplemental” to the OECD Implementation Handbook and guidance notes. (Since the OECD doesn’t have “guidance notes” as such, this latter reference is presumably to the FAQs.)
  • The second draft of the Jersey CRS Guidance Notes (February 2016 version) calls the CRS Implementation Handbook a “helpful publication in understanding” CRS, but stops short of saying FIs should follow it.
  • The Isle of Man’s draft CRS Guidance Notes (24 December 2015 version) goes further. It says that FIs “must” apply the Isle of Man CRS regulations with reference to “the CRS itself, the OECD’s guidance and lastly this, the Isle of Man’s own published guidance.” Earlier, the Isle of Man Guidance Notes say that “guidance on practical aspects and the operation of the CRS can also be found in the OECD’s CRS Implementation Handbook”. Presumably, therefore, the reference to the OECD’s “guidance” is to the Handbook though arguably not to any FAQs (such as the one on Protectors) that are not included in the Handbook.

It is clear, therefore, that regulators in most jurisdictions will expect fiduciaries to follow the CRS FAQs and Implementation Handbook. This is true even though local implementing legislation or regulations may not refer to those documents at all. Should fiduciaries therefore follow the FAQs and Implementation Handbook? Sure. Provided they don’t contradict CRS itself. Bear in mind that local Guidance Notes themselves are almost never binding. (Check your local Guidance to be sure.) Regulators, just like fiduciaries, are subject to the rule of law. Therefore, if a regulator’s (non-binding) Guidance Notes say that a fiduciary may (or, in the Isle of Man’s case, must) follow an OECD publication, compare that to what your local law says. If local law says to follow the Standard and non-binding Guidance Notes tell you to follow subsequent OECD publications that directly contradict the Standard, the Guidance Notes should—no, must–be ignored.

That said, a conscientious fiduciary will naturally be reluctant to ignore what it perceives to be its local regulator’s expectations regarding CRS implementation. Therefore, some may err on the side of reporting all Protectors as Account Holders in FI trusts. After all, a regulator is far less likely to complain about over-reporting than under-reporting. At first blush therefore, over-reporting looks like the safer course. But is it?

At the end of every report is a human being. One who might be very upset if a fiduciary reported him or her when it was not required to by law. Upset enough to sue. Over-reporting itself may very well violate local law, including privacy and data protection laws, and give rise to legal liability to the person improperly reported. Therefore, rather than being the safer course, over-reporting will generally be the more perilous one.

If I were a fiduciary, I would think long and hard about whether to follow an FAQ or provision of the Implementation Handbook that contradicts the Standard. Just to be clear: I have no problem whatsoever with fiduciaries following those documents in general—but not when they contradict what local law says fiduciaries should be implementing. Obviously, given that local laws do vary, fiduciaries should get legal advice about their home jurisdiction’s CRS implementing rules. In general, though, fiduciaries are likely to find that the CRS FAQs and Implementation Handbook are not binding.

This blog and the previous one have focused on the FAQ about Protectors. That is by no means the only OECD pronouncement of questionable validity. For example, the Implementation Handbook says that the reference in the Standard’s definition of Account Holders in FI trusts to “other natural person exercising ultimate effective control over the trust” includes trustees. I have serious doubts about that interpretation. Trustees are expressly included in the definition of Controlling Persons of Passive NFE Trusts. However, trustees (along with protectors) were deliberately deleted from that definition when it was amended to become the definition of Account Holders in FI trusts. What was the purpose of that deletion if trustees are nonetheless automatically included in the catch all “other natural person exercising ultimate effective control over the trust”? Presumably the deletion was meant to accomplish something. Indeed, what was the purpose of expressly including trustees in the definition of Controlling Persons of Passive NFE Trusts in the first place if they were already automatically included as “other natural person exercising ultimate effective control over the trust”?

Here’s another significant provision of the Handbook that is, in my view, invalid and should be ignored (subject to what local law says): The part that requires looking through all entities that are equity-interest Account Holders in FI trusts and reporting their Controlling Persons. Nonsense! CRS, just like FATCA, requires looking through Passive NFEs and reporting their Controlling Persons. But CRS, just like FATCA, does not look though FIs and require reporting their Controlling Persons (subject to a narrow exception in CRS for professionally managed Investment Entity FIs located in Non-Participating Jurisdictions.) So just where does the OECD get off saying that trusts—and only trusts, mind you—must look through all equity-interest Account Holders and report their Controlling Persons? This provision of the Implementation Handbook should be ignored to the extent it requires looking through entities other than Passive NFEs and professionally managed Investment Entity FIs in Non-Participating Jurisdictions (again, subject to what local law says to follow).

There are a number of other parts of the FAQs and Implementation Handbook that arguably should also be ignored. However, there are too many to cover in this one blog. Suffice it for now to re-iterate the main point of this blog: You should ignore any of the CRS FAQs and any language in the Implementation Handbook that directly contradicts what local law requires you to implement (generally, only the Standard itself, but check local implementing legislation and regulations). The FAQ on Protectors is just one example. The general principle is much broader.

PS We’ve recently uploaded the Jurisdiction Selection Decision Trees for both FATCA and CRS to the Subscriber Resources page of gatcandtrusts.com. Much more is in the works and will follow soon.

CRS and Protectors: The OECD Has Lost the Plot.

Peter Cotorceanu Blog, CRS, CRS and Protectors, OECD

I’ll go even further: The OECD has gone mad when it comes to Protectors. Stark raving mad. The real question now is whether countries implementing CRS will follow the OECD into the insane asylum. Some might. None ought to.

What on earth am I on about?

Earlier this month, the OECD posted updated CRS FAQs on its website. They are available here. If you scroll down to page 15, Question 5, you will see the following new Q&A:

protectors of a trust

This FAQ does not pass the laugh test. The OECD’s own definition in the Common Reporting Standard (CRS) of an Account Holder in a trust that is a Financial Institution is a person who holds a so-called “equity or debt interest” in the trust. A “debt interest” Account Holder is not defined, and debt interests are in any event not relevant to this blog. But “equity interest” Account Holders are defined in the CRS. Specifically, they are held by “a settlor or beneficiary of all or a portion of the trust, or any other natural person exercising ultimate effective control over the trust.” Since the word “Protector” is not mentioned in that definition, the only way a Protector could—even arguably—be an equity interest Account Holder in a trust is if it satisfied the language at the end of the definition, i.e., if it were a natural person “exercising ultimate effective control over the trust.” But look back at the FAQ—it says a Protector must be treated as an Account Holder “irrespective of whether it has effective control over the trust.” This is truly nuts and directly contrary to the OECD’s own definition of an equity interest Account Holder in a trust.

Now let’s dig a little deeper.

Under both FATCA and CRS, the whole universe of entities is divided into two groups: “Financial Institutions” (FIs) and “Non-Financial Entities” (NFEs).

Fls and NFE

NFEs are further divided into two groups: “Active” NFEs and “Passive” NFEs.

Active and passive NFE


As relevant to this blog, FIs report two things:

  • their “Account Holders”, and
  • if their Account Holders are Passive NFEs, the Controlling Persons of those Passive NFEs

(The Controlling Persons of Active NFEs are irrelevant.)

Therefore, what’s relevant for FIs is Account Holders, and what’s relevant for Passive NFEs is Controlling Persons:

FIs Passive NFEs
Account HoldersControlling Persons

Now, the FATCA Inter-Governmental Agreements (IGAs) and CRS both define (i) the “Account Holders” of a trust that is an FI, and (ii) the “Controlling Persons” of a trust that is a Passive NFE. Those definitions are different when it comes to Protectors. As we shall see, that difference is fundamental.

For reasons I’ll explain later, my firm view is that a Protector should never be an Account Holder of a trust that is an FI if that trust is properly administered. Some practitioners take a different view, arguing that a Protector can be an Account Holder of an FI trust if the Protector has certain powers especially the right to change the trustee. One thing is indisputable, though: As demonstrated above, it is absolutely nonsensical to argue that all Protectors are Account Holders in FI trusts. And yet this is the very position the OECD takes in the above FAQ!

To understand even further why the FAQ’s position makes no sense, you need to understand the history of the definition of an Account Holder in an FI trust. More specifically, you need to know not just what that definition says but how it relates to the definition of a Controlling Person of a Passive NFE trust. Let’s start with the latter.

CRS defines a Controlling Person of a Passive NFE trust as “the settlor(s), the trustee(s), the protector(s) (if any), the beneficiary(ies) or class(es) of beneficiaries, and any other natural person(s) exercising ultimate effective control over the trust.”

Passive NFE Trust 2

The FATCA IGAs’ definition is identical in all materials respects. As you can see, Protectors are expressly included. There is no question, then, that at least some Protectors are Controlling Persons of Passive NFE trusts.

One can quibble about whether the language at the end of the definition (“other natural person(s) exercising ultimate effective control over the trust”) modifies all the categories that come before it, i.e., whether Controlling Persons include all settlors, trustees, Protectors and beneficiaries or only those exercising ultimate effective control over a trust. The CRS Commentary takes the position that all Protectors are included even if they are not “exercising ultimate effective control over the trust.” I don’t necessarily agree with that interpretation. However, the definition is ambiguous enough that the OECD’s interpretation is certainly plausible.

Thus, under CRS, if you’re a Protector of a trust that is a Passive NFE you’re automatically a Controlling Person. Full stop. And if the Passive NFE trust is an Account Holder of an FI, e.g., it has an account at a bank, the FI (bank) will report the Protector as a Controlling Person. That much is clear. Open issues include how one treats Protector committees, corporate Protectors, and persons with Protector-like powers who are not called Protectors, e.g., “Appointors”. I’ll leave those issues for another day. To keep things simple for this blog, we’ll assume we’re dealing with a Protector who is called by that name, who is an individual, who is not also a settlor or beneficiary, and who is not a member of a Protector committee.

Now, the above definition of a Controlling Person of a Passive NFE trust was not created by the OECD or even by the U.S. when it wrote the FATCA IGAs. No, the definition was copied from the definition of a “beneficial owner” of a trust in the 2012 version of the so-called Financial Action Task Force (FATF) Recommendations, which are anti-money-laundering (AML) rules. Indeed, both CRS and the FATCA IGAs expressly require the definition of Controlling Persons of Passive NFE trusts to be interpreted consistently with the FATF Recommendations.

With that background, let’s turn away from the definition of a Controlling Person in a Passive NFE trust back to what’s directly involved in the OECD’s new FAQ on Protectors, i.e., the definition of an an Account Holder in an FI trust.

Under both CRS and the FATCA IGAs, the definition of an Account Holder of an FI trust uses the definition of a Controlling Person of a Passive NFE Trust as its starting point. But it makes hugely significant changes especially when it comes to Protectors. Take a look at the following table comparing the two definitions and you’ll see what I mean.

Passive NFE Trust

As you can see, neither trustees nor Protectors are mentioned in the definition of an Account Holder in an FI Trust. This is not a mistake. It was an intentional omission.

Remember, the definition of a Controlling Person of a Passive NFE was copied from the FATF Recommendations. But the FATF Recommendations don’t define “Account Holders” in trusts. This is because trusts don’t actually have “Account Holders”. Settlors, yes. Beneficiaries, yes. Trustees, yes. Protectors, maybe. But “Account Holders”? No. Except that both CRS and FATCA say FI trusts have “Account Holders” (I am not making this up!) so they obviously had to define that term. What did the drafters do? As mentioned earlier, they defined an “Account Holder” in an FI trust to be anyone with a “equity or debt” interest in the trust. So what is an “equity” interest”?

As anyone with a passing knowledge of trust law knows, beneficiaries have equitable title to trust assets and trustees have legal title. Protectors, of course, have neither equitable nor legal title. So it made sense to include beneficiaries as “equity interest” Account Holders in trusts, but not to include trustees or Protectors. As for settlors, well, unless they are also beneficiaries, they don’t have equitable title. However, settlors at least come a lot closer to having equity interests in trusts than do trustees and Protectors—after all, settlors are the people who actually create trusts and transfer their assets to those trusts. So, rightly or wrongly, settlors were included in the definition of “equity interest” Account Holders. But, rightly, trustees and Protectors were deleted from the definition, which took as its starting point the definition of Controlling Persons of Passive NFE trusts.

Thus, as pointed out above, given that Protectors (and trustees) are not expressly listed as Account Holders in FI Trusts, the only possible way they could be included is if they fell within the language at the very end of the definition, i.e., if they were natural persons “exercising ultimate effective control over the trust.” There is no other possibility: Unless a Protector is also a beneficiary or settlor (which is not the sort of Protector we’re discussing in this blog), there can be no way a Protector can fall within the definition of an Account Holder in an FI trust.

As mentioned earlier, I don’t believe any Protectors of trusts that are properly administered are Account Holders in FI trusts. In other words, I don’t believe any Protectors exercise “ultimate effective control” over properly administered trusts. The definition doesn’t say “some control”, “much control”, or even “significant leverage”. It says “ultimate effective” control. That’s pretty powerful wording. That describes someone who’s running the show and calling the shots—someone in the driver’s seat.

Now, some practitioners argue that if a Protector has the right to change the trustee, the Protector does run the show and call the shots and is therefore exercising “ultimate effective control” over the trust. I beg to differ. Sure, a Protector with the right to hire and fire the trustee has a good deal of leverage over the trustee. But “ultimate effective control”? To reach that conclusion, you have to assume the trustee is, in effect, the Protector’s puppet—that the trustee is so afraid of being fired that it does whatever the Protector says. If the trustee truly cowers in this fashion, then the trustee is not doing its job. A trustee has fiduciary obligations. If it ignores those obligations and simply bows to the Protector’s every wish to save its job, it is in breach of trust. So, I maintain that a Protector of a trust that is properly administered should never meet the definition of an Account Holder in a FI trust, i.e., should never be treated as exercising “ultimate effective” control over the trust.

Let’s say, though, that you have a trust that’s not properly administered, i.e., where the trustee really does do the Protector’s bidding and cedes “ultimate effective” control over the trust to the Protector. We can all agree that in that case—a case where the trustee is in clear violation of its fiduciary duties—the Protector should be treated as an Account Holder of the trust. However, all sane people can also agree that—provided you apply the actual definition of an Account Holder, not some made-up definition—that a Protector who does not exercise “ultimate effective” control cannot be an Account Holder.

Now let’s go back to the FAQ. It says a Protector “must be treated as an Account Holder irrespective of whether it has effective control over the trust”. That is patent nonsense. Again, an Account Holder is (i) a settlor, (ii) certain beneficiaries, and (iii) any other natural person “exercising ultimate effective control over the trust.” So on what rational basis can anyone—even the OECD—argue that a Protector is an Account holder “irrespective” of whether it has effective control over the trust? Not possible. Except that’s just what the OECD has done in this FAQ. By sheer fiat, it has re-written its own definition of an Account Holder in an FI trust—a definition that appears in its own CRS and Commentary—and reinserted the word Protector. Think about it—the definition of an Account Holder in an FI trust deliberately excludes the word “Protector”. And now the OECD says “oops”, we’re sticking it back in. Not by amending the definition, mind you, but by adopting an FAQ that directly contradicts the definition.

You may well think all of the above is academic—that the OECD makes the rules, and if it wants to change them mid-stream by playing fast and loose with its own definitions, so be it. CRS countries and the trustees in those countries have to follow these rules, even if they’re nonsensical, right? Wrong!

In my next blog, I will address head on whether this new FAQ, indeed whether any OECD FAQ or even the the so-called “CRS Implementation Handbook”, should be followed if they contradict the CRS and its Commentary. For reasons I will explain in that blog, I will show that, unless CRS countries roll over and play dead and require their fiduciaries to do the likewise, anything the OECD says that contradicts the plain language of the CRS and Commentaries themselves should be ignored. I’ll go further: Anything the OECD says, whether in FAQ or the Implementation Handbook, that contradicts the plain language of the CRS and Commentaries must be ignored unless expressly required to be followed by an implementing jurisdiction. Stay tuned.

Reflections from Panama

Peter Cotorceanu Blog, Panama Papers





I was in Panama last week for the Latin Private Wealth Management Summit. The Summit was organised by the Marcus Evans group. Congrats to all concerned for a very well run event.

I’d never been to Panama before. I am definitely going back. Panama City is an interesting mix of new and old—lots of shiny skyscrapers, including the BBA building, which is the coolest one I’ve ever seen but also an amazing old town—walking through it is like stepping back in time to the colonial era. It’s full of neat bars and restaurants and interesting holes-in-the-wall. And the air is filled with salsa music drifting lazily out of almost every one of them. Time seemed to stand still, trapped by the languid heat and humidity. Well worth a visit.

But the most impressive thing about Panama is the people. I’ve been to a lot of countries. I’ve never met more friendly, smiley, happy people anywhere. They’re warm. They look you in the eye. They engage you kindly. The only people that come close in my experience are the people in Costa Rica and Ghana. Hats off to all of them.

I stayed at the Trump hotel. I was in the Presidential Suite. Nah, kidding. It was called the “Soon to be President” Suite. Kidding about that too. It wasn’t even the “Republican Nominee’s” Suite. But it was nice. On the 28th floor with views of the Pacific and of the skyline. Great hotel. On many levels (pun intended). The bar on the 66th floor of the casino has an incredible view and an infinity pool–yes, in the bar!

One of the most interesting parts of the trip was my visit to Mossack Fonseca’s office. I made a special effort to find the place just so I could have this picture taken.


Mossack Fonseca is, of course, the Panamanian law firm whose files were stolen and are now known euphemistically as “the Panama Papers”. I just call them what they are—the Stolen Client Files. Not as catchy as “the Panama Papers”, for sure, but why use the anodyne name the thief gave them? Ideas—and names—matter.

Look, I am all for rooting out corruption and tax evasion and crime. If the theft of these files reveals such shenanigans—and they certainly will—then some good will have come of this larceny. But that is no justification for what the thief did. He claims in his recently published “Manifesto” (how pretentious is that?) that he wanted to expose “corruption” and “injustice”. (I refer to the thief as “he” because the Manifesto is written under the pseudonym “John Doe”. The “Manifesto” is available here.) The thief rails against income inequality and tax havens. And he paints himself as just another whistleblower doing all in the public good.

Except, of course, that whistleblowers worthy of that name reveal wrongdoing and no more—they don’t wholesale steal files and release all, even those of perfectly innocent people. John Doe claims that Mossack staff should be “prosecuted . . . with no special treatment”. So should he if he is ever caught. The thief also decries “our society’s progressively diseased and decaying moral fabric.” Look in the mirror, John Doe–you are a thief! And a hypocrite to boot. John Doe is all for secrecy—he used encrypted phone lines to communicate with journalists, he refused a face-to-face meeting, he uses a pseudonym. Why? Because, he claims, his life would otherwise be at risk. He’s probably right about that. And no-one can blame a person for wanting to hide if their personal security is at risk. Not even wealthy people living in countries prone to kidnapping and extortion and murder who use offshore structures to prevent criminals from learning about their wealth, eh John Doe? What’s sauce for the goose is certainly sauce for the gander. One person’s secrecy is another person’s privacy.

I will not shed a tear for any corrupt or tax evading person whose dark deeds are revealed as a result of this theft (that does not justify the theft of course—ends don’t justify means). I am, however, distressed about all the innocent people who have been hurt by John Doe’s criminal actions. I am even more distressed by the culture of data theft that seems so pervasive in this day and age. A zeitgeist for sure.

So those are my thoughts after my trip to Panama, aka the belly of the beast. Next blog will be about substantive GATCA and FATCA issues—promise.

The Panama Papers: We’re Being Played for Fools

Peter Cotorceanu Blog, Panama Papers

You’ve got to hand it to the International Consortium of Investigative Journalists (ICIJ). They’ve pulled off an amazing coup. They’ve willingly received over 11 million stolen and highly confidential documents. They’ve trolled through them exhaustively for a year. They’ve cherry-picked for publication the most salacious morsels with the most famous names attached. And they’ve made the whole world—media, politicians, the public at large—believe that this incredibly unrepresentative sample paints an accurate picture of the offshore world–a shadowy world full of tax evaders, money launderers, and other scoundrels.

Ha! The truth—the whole, inconvenient truth–is quite to the contrary. Yes, dodgy deals go down in the offshore world. No surprises there. They go down in the onshore world too. But the overwhelming majority of offshore entities are used for perfectly legitimate reasons by perfectly honourable people.

Which would be plain for all to see if the ICIJ hadn’t hoarded its ill-gotten treasure. Talk about secrecy! If the ICIJ were truly interested in transparency, it would do a complete document dump instead of selectively releasing the most damning data. If everything were revealed, the ICIJ would be allowing others to make up their own minds based on all the facts. But then the ICIJ would lose control of the story. No, the ICIJ—let’s call them the International Consortium of Intransparent Journalists shall we?—has a political agenda: It wants privacy jurisdictions (it calls them “secrecy” jurisdictions) and “tax havens” shut down. The truth be damned.

By the way, isn’t receipt of stolen goods as much a crime as actual theft? Surely data are goods of a sort, aren’t they? Shouldn’t prosecutors be looking at the ICIJ’s actions? Why haven’t they? Could it be that governments love data theft as much as intransparent journalists do when it suits their own purposes and agenda? Just ask the German state of North Rhine Westphalia, which has made a habit of receiving stolen bank information.

Maybe someone should hack the ICIJ and publish all the documents since the ICIJ won’t. I don’t actually advocate that—stolen, confidential data should remain private. It would be ironic, though, wouldn’t it, if someone did manage to hack the ICIJ? Then perhaps the ICIJ’s own manipulative agenda would be revealed for the whole world to see.