[*PG253]COERCING COOPERATION FROM OFFSHORE FINANCIAL CENTERS: IDENTITY AND COINCIDENCE OF INTERNATIONAL OBLIGATIONS AGAINST MONEY LAUNDERING AND HARMFUL TAX COMPETITION

Benjamin R. Hartman*

Abstract:  International initiatives concerning the global financial system traditionally have been implemented through the building of consensus among affected states to identify problems and then set forth the means to deal with those problems. Recently, however, under the title Actions Against Abuse of the Global Financial System, the G7 nations began a campaign that uses the threat of sanctions to coerce cooperation from offshore financial centers in the areas of money laundering and tax competition. The use of sanctions to force compliance is problematic because, although sanctions would be available to remedy a violation of an international obligation, there has been no attempt to identify any such obligation for the Object States. Instead the Financial Action Task Force (FATF) and the Organization for Economic Cooperation and Development (OECD), in accordance with the agenda of the G7, have set forth self-referential criteria that assess compliance of the offshore financial centers with presumed international “standards” against money laundering and harmful tax competition. The legal enforceability of the sanctions threatened for non-compliance, however, depends upon the existence of legal bases for both sets of criteria. This Article attempts to determine whether such bases exist.

[*PG254]Introduction

The summer of 2000 was hard on many of the classic “offshore” financial centers (OFCs); they found themselves in the center of a multi-front attack. On May 25th they were labeled “non-cooperative” by the Financial Stability Forum (FSF) in the context of global financial stability, then again on June 22nd, “non-cooperative” by the FATF in the context of money laundering, and finally, on June 26th, “Tax Havens” by the OECD in the context of tax competition. A total of nine countries enjoyed the misfortune of being thrice labeled in those two months: three in the Caribbean (the Bahamas, St. Kitts-Nevis, and St. Vincent) and four in the Pacific (Niue, Nauru, and the Cook and Marshall Islands) as well as Panama and Liechtenstein.1

It is no accident that OFCs have found themselves the objects of such timely criticisms, but rather it is the manifestation of the efforts of the Group of 7 Finance Ministers to the Heads of State and Government (G7), all of which are influential members of the FATF, OECD, and FSF. The G7 collects the initiatives of the three organizations under the title Actions Against Abuse of the Global Financial System while it acknowledges the initiatives separately2 under the sub-headings of Money Laundering, Tax Havens and Other Harmful Tax Practices, and Offshore Financial Centers.3 The G7 threatens the use of sanctions4 against jurisdictions that “demonstrate failure to meet certain standards and are not committed to enhancing their level of compliance with international standards.”5 While international initiatives concerning the global financial system traditionally have stressed consensus building, the concept of identifying, isolating, and punishing countries for non-compliance with international standards has be[*PG255]come more common.6 The idea first originated in 1988 in the context of money laundering.7 Since then some unilateral examples have demonstrated the effectiveness of this approach.8

The effectiveness of the novel approach of threatening sanctions has incited some heavy criticisms from OFCs. The Member States of the Caribbean Community have characterized the three initiatives as “continued attacks on the Region’s offshore sector by the G7,” and have even gone so far as to describe the FATF, OECD, and FSF as “agencies” of the G7.9 The Cayman Islands—which had just hosted a separate initiative organized under the UN by the United Nations Office for Drug Control and Crime Prevention (UNODCCP) Global Programme on Money Laundering10—said it was astonished about being included on the FATF list and that “the decision was made without due process.”11 These criticisms are not altogether unfair in light of the methods that the G7 and the other three organizations have adopted.

The threat of sanctions is problematic in several ways.12 The most significant problem is that while the sanctions are threatened for non-[*PG256]compliance with certain “standards,”13 only the vaguest attempt has been made to define what those standards might be, and no attempt has been made to identify any source of obligation which would require—as opposed to encourage—compliance. The FATF lists a number of standards “for instance14 but does not clearly define what requirements actually are standard in the international context. Moreover, the OECD, in a self-referential manner, demands compliance with its own standards but identifies no source, outside itself, which would make them obligatory standards.15

It is the goal of this Article to address these complaints. This Article, however, is not concerned with whether the G7 is abusing its disproportionate political power. Neither is it concerned with any political or moral arguments for or against the current state of the international financial system. Consequently, I will offer neither apocalyptic nor utopic descriptions of an unregulated international financial system. Instead, this Article is an attempt to address the OFCs’ complaints about the “attack” coordinated against them by the G7 through its agents.

Necessary to this end is a determination of the legal bases of both the initiatives and the standards the G7 attempts to impose, and the consequent enforceability of the sanctions threatened for non-compliance. This attempt begins with an introduction to the three initiatives, the organizations relevant to them, the states that those [*PG257]organizations comprise, and the states that are the objects of the initiatives. In the second part, I attempt to identify the sources of the obligations that could bind the OFCs. Finally, the third part of the analysis assesses, based upon the obligations established in section two, to what extent sanctions would be available to the G7, its “agencies,” and their member states against OFCs for non-compliance with the goals set by the initiatives.16

Although the FSF initiative is included in the first section to demonstrate the degree of overlap among the identified states, it is excluded from the latter two sections for the following two reasons. First, its policies are inherently less suspect than those of the other two organizations. The goal of the FSF’s initiative is to assess which OFCs might have implications for the International Monetary Fund’s (IMF) work on the assessment of financial stability in general, and for the joint IMF-World Bank Financial Sector Assessment Program in particular.17 With 182 member states, the IMF is a much more representative body than the FATF and OECD.18 Its policies, which are built upon the consensus of those member states, enjoy greater consensus within the international community as a whole, and these policies are, consequently, less suspect. Second, although the efforts of the G7 to tie the three initiatives together have met with some success, the FSF has resisted becoming the simple agent of the G7. Unlike the FATF, the FSF expressly has dissociated itself from the OECD’s initiative by specifically excluding the issue of harmful tax competition from its agenda.19 These two factors effectively insulate the FSF initiative from much of the criticism born by the other two organizations, especially the OECD. Therefore, it is excluded from the analysis below.

[*PG258]I.  Coincidence and Identity

A.  Players

The states relevant to the inquiry of the existence of legal obligations can be divided into two groups: those attempting to impose the obligations (Subject States), and those against whom the obligations are being imposed (Object States). This distinction is made more real by the fact that there is no overlap between the membership of the two groups.

1.  Subject States

The members of the G7, the FATF, and the OECD comprise the Subject States. The Group of 7, which are members of both organizations, is central to the initiatives of these two organizations. The G7 includes Canada, France, Germany, Italy, Japan, the United Kingdom, and the United States (US). The FATF was convened by the G7 in July 1989 specifically to address the problem of money laundering and has become the primary international forum focusing on combating money laundering.20 It has thirty-one members: twenty-nine member countries and two international organizations.21 The OECD, on the other hand, was formed to provide “governments a setting in which to discuss, develop and perfect economic and social policy.”22 There are twenty-nine members of the OECD.23

In addition to the FATF and the OECD, there are other international organizations that attempt to deal with the issues here raised. [*PG259]In the area of money laundering, there are several regional efforts to deal with the relevant problems: Asian Pacific Group on Money Laundering (APG),24 the Caribbean Financial Action Task Force (CFATF),25 the Council of Europe Select Committee of Experts on the Evaluation of Anti-Money Laundering Measures (PC-R-EV Committee),26 and the Organization of American States (OAS).27 In the area of tax competition, there is much less in the way of organizations attempting initiatives analogous to the OECD’s. The most analogous multilateral effort is that initiated by the European Union, the Code of Conduct, but this initiative does not address Tax Havens and therefore is not relevant.28 Even though these additional organizations might be implementing parallel initiatives, in no instance are they implementing the initiatives in question. This factor is important because their membership comprises many of the Object States.

2.  Objects States

Despite the differences in the lists and their sources, there still is substantial overlap among the countries that are the objects of the three lists. The FSF identifies thirty-four states as problematic.29 The [*PG260]FATF identifies fifteen states as non-cooperative.30 It also indicates that five of the jurisdictions examined but not included on the list have introduced relevant legislation since 1999.31 Without this late compliance, it is possible that these five might have been included on the list. The OECD list identifies thirty-five tax haven countries.32 Another six were left off the list because they made a “full commitment to eliminate harmful tax practices by the end of 2005.”33 Technically the OECD does not intend the List of Tax Haven Jurisdictions to be a basis for sanctions;34 ultimately, the OECD plans to publish a List of Uncooperative Tax Havens. The List of Tax Havens Jurisdictions, however, still is useful for two reasons: (1) the OECD acknowledges that members are free to use the list as a basis for unilateral sanctions;35 and (2) it is strongly predictive of the ultimate list because any jurisdiction on it that, by July 31, 2001, has not made a commitment to eliminating harmful tax practices “would automatically be included in the List of Uncooperative Tax Havens.”36 In addition, even those that do commit, including the six that have committed already, but that do not meet the deadlines set by the OECD or that fail to fulfill their commitment in good faith would be placed back on the list of uncooperative countries.37

The coincidence of Object States identified by the three lists is strong. Of the fifteen FATF non-cooperative, only three do not appear on one of the other lists: Israel, the Philippines, and Russia; that is to [*PG261]say that 80% of the states that are threatened with sanctions by the FATF are also threatened by either the OECD or the FSF. And as stated above, there are nine countries that have been identified on all three lists.38

These nine can represent a sample of Object States since they all can be used to assess to what extent the standards can be imposed by both initiatives upon Object States. Since I am here interested in only the OECD and FATF initiatives, I add to this sample Dominica, which was not on the FSF list, but was on the other two. Additionally, I include the Cayman Islands in the sample because it was on the FATF list of non-cooperative countries (NCC List), and would have been on the Tax Haven List had it not made an advanced commitment to implement the OECD’s standards.39 In total there are eleven Sample Object States to which I generally will limit my analysis—the Bahamas, the Cayman Islands, the Cook Islands, Dominica, the Marshall Islands, Niue, Nauru, Panama, Liechtenstein St. Kitts-Nevis, and St. Vincent and the Grenadines.

B.  Initiatives

Even though both the FATF NCC List and the OECD List of Tax Haven Countries, have been identified by the G7 as Actions Against Abuse of the Global Financial System, and they both share the same or substantially overlapping Subject and Object States, they are separate in their legal inception and development. The discreet nature of the initiatives must be followed rigorously in order to distinguish between political rhetoric and legal obligations. While political goals and obligations can be relatively malleable, legal obligations are necessarily more rigid and, therefore, more difficult to form or to change.

1.  FATF: Review to Identify Non-Cooperative Countries or Territories

On June 22, 2000, the FATF published its Review to Identify Non-Cooperative Countries or Territories, which contained the above mentioned list of non-cooperative OFCs.40 The countries that were determined to be non-cooperative were found so based upon the [*PG262]twenty-five criteria (25 Criteria) promulgated by the FATF in its Report on Non-cooperative Countries and Territories. These criteria were set forth “to identify detrimental rules and practice which impede international co-operation in the fight against money laundering.”41 The criteria are said to be “consistent with the FATF Forty Recommendations.”42 The Forty Recommendations were first promulgated by the FATF in 1990 and then revised in 1996 to take into account the lessons of six years of experience.43 These recommendations are said to “have been established as the international standard for effective anti-money laundering measures.”44 Moreover, the Forty Recommendations are built upon the foundations laid by the 1988 UN convention, the Basle Committee on Banking Regulations and Supervisory Practices,45 and make reference to the standards promulgated by the International Organization of Securities Commissions, the International Association of Insurance Supervisors, the International Accounting Standards Committee.46

The 25 Criteria are subdivided into four categories: (1) loopholes in financial regulations; (2) impediments set by other regulatory requirements; (3) obstacles to international cooperation; and (4) inadequate resources for preventing, detecting and repressing money laundering activities.47

2.  OECD: Report on Progress in Identifying and Eliminating Harmful Tax Practices

On June 26, 2000, the OECD published its Report on Progress in Identifying and Eliminating Harmful Tax Practices,48 which provided the names of the Tax Havens listed above. This report represents the [*PG263]first step in the work of the Forum on Harmful Tax Practices. The Forum was created under the mandate of the 1998 OECD49 as a subsidiary body of the Committee on Fiscal Affairs (CFA).50 The CFA “was established in 1971 to provide a forum for senior tax policy-makers and administrators from OECD Member countries and non-OECD countries to discuss international and domestic tax issues.”51 The Forum is “responsible for monitoring the implementation of the Guidelines and Recommendations set out” in the OECD Report on Harmful Tax Competition.52 The OECD Report was prepared, under the joint Chairmanship of France and Japan, by the “Special Sessions on Tax Competition” in response to the Ministerial Communiqué of May 1996 that called upon the OECD to “[d]evelop measures to counter the distorting effects of harmful tax competition on investment and financing decisions and the consequences for national tax bases.”53 The Special Sessions was created by the Committee on Fiscal Affairs specifically to address this request.54 The Harmful Tax Competition Report identified four criteria: (1) no or only nominal taxes; (2) lack of transparency; (3) no substantial activities; and (4) lack of effective exchange of information.

3.  The Ultimatum: Compliance or Sanctions

Both the OECD and the FATF conceive of using coercive action to gain cooperation from OFCs. The OECD in its 2000 Report lists numerous measures, such as termination of treaties, to be taken against OFCs which refuse to comply.55 The Report also suggests that “non-tax measures” should be considered.56 The FATF speaks of restricting or even prohibiting financial transactions with non-cooperative jurisdictions, “as an ultimate recourse should a country or territory have decided to preserve laws or practices that are particularly damaging for the fight against money laundering.”57 The fact [*PG264]that sanctions are an ultimate threat should not obscure the fact that they are still a threat.

These measures clearly can be harmful to OFCs. The Economist notes that “[e]ven mild measures can have surprising bite. Last year, Antigua suffered a financial drought after America and Britain issued an ‘advisory’ to their financial institutions, recommending ‘enhanced scrutiny’ for transactions there.”58 But more extreme measures also are being considered; for example, France has indicated that it could be ready to “cease all financial relations of whatever type” with offending countries.59 If an advisory can cause a financial drought then an absolute prohibition on financial transactions with an OFC would be devastating.

Several of the measures suggested potentially would constitute violations by the Subject States of their obligations toward OFCs under international law. For example, the implementation of trade obstacles, such as barring the provision of cross-border financial services as the FATF suggests, could violate obligations to the WTO.60 Unilateral termination of a treaty, as the OECD suggests, also could constitute a breach under Article 54 of the Vienna Convention on the Law of Treaties (Vienna Convention).61 Although there is no certainty that these measures would be illegal, the potential for illegality identifies, in this context, the obligations that Subject States owe to Object States.62

The potential wrongfulness of a breach of the Subject States’ obligations, however, would be obviated if the measures were taken to remedy a prior breach by the Object States. As Article 30 of the International Law Commission’s (ILC) Draft Articles on State Responsibil[*PG265]ity states, “The wrongfulness of an act of a State not in conformity with an obligation of that State towards another State is precluded if the act constitutes a measure legitimate under international law against that other state, in consequence of an internationally wrongful act of that other State.”63 This statement would make the suggested measures legal to the extent that they are a legitimate response to a prior breach by the Object States.

C.  Brief Consideration of Obligations: Incurrence, Breach, and Sanctions

1.  Preliminary Considerations

Although much of the rhetoric surrounding these initiatives relies upon the criminal element of the relevant acts, this is inadequate as a legal basis upon which to rest these initiatives. Both initiatives threaten sanctions for the refusal to remedy the potential for harm, not actual harm.64 Some parties, such as UNODCCP, say that inherent in the concept of a state’s sovereignty is the principle that no other state “should assist citizens or residents of another State in the violation of the laws of their home country.”65 However, even if we accept the principle that State A is injured by State B when B allows A’s citizens to break A’s laws, the obligation that would come therefrom would be that B must not allow A’s citizens to break A’s laws. It does not implicitly follow that B would be obliged to remove even the potential for A’s citizens to break A’s laws as both the FATF and OECD initiatives require. Furthermore, some of the obligations the OECD attempts to impose have no connection with criminal acts. The OECD includes low taxes along with tax evasion under its concern about erosion of tax base. There is, however, no necessary connection between low taxes and tax evasion. There is, therefore, no basis to claim that offering low taxes facilitates tax crimes. For sanctions to be available to Subject States, therefore, the breaching Object States must be bound under an obligation from some other source to comply with the criteria of the OECD and FATF.

[*PG266]2.  Sources of Obligation

These potential sources of obligations can be divided into three categories: (1) Consent, (2) Coincidence, and (3) Consensus. Consent and Consensus66 correspond with the traditional divisions of sources of law under treaties and customs as they are identified in Article 38(1) of the Statute of the International Court of Justice (I.C.J.).67 Coincidence refers to the possibility that obligations under parallel instruments might exist or that behavior of the Object States might estop them from complaining.

To be enforceable, the 25 Criteria either can define the legal obligations themselves or simply assess compliance with other legal obligations.68 If, and to the extent that, they define the legal obligations—which would be possible if the 25 Criteria themselves are legally binding—then they are enforceable. However, if they simply assess compliance with other legal obligations, then they are enforceable only to the extent that they parallel those other actual obligations.

It should be noted, though, that while “there is a wide range of views on the basis of obligation—natural law, consent, principles anterior to the legal system itself, consensus, reciprocity—it is interesting that they all exclude imposed obligation by the enforcement of sanctions.”69 This fact reinforces the principle that, to the extent that the measures constitute sanctions absent some breach of a prior obligation, the unilateral, or even multilateral, imposition of sanctions does not create a legal basis for compliance.

The category of Consent embodies the traditional concept that a state can agree to create obligations to another state. An international agreement is “an agreement between two or more states or international organizations that is intended to be legally binding and is governed by international law.”70 Such an agreement creates obligations [*PG267]binding between the parties under international law.71 An agreement enters into force once the parties have expressed their consent to be bound.72 A state’s consent to be bound by an international agreement may be expressed in a number of ways including signature, exchange of instruments, ratification, acceptance, approval, accession, or by any other means agreed upon.73 But absent such consent, there exists no agreement and, therefore, a state would not be bound.

Coincidence74 can be divided into two sub-categories, obligations incurred either through: (1) parallel agreements or (2) through unilateral declarations. Subject States could argue that Object States, which have expressed an intent—even if in some other forum—to perform in a manner that parallels the expectations contained in one of the respective initiatives, would be estopped from claiming freedom from compliance with that parallel obligation.

The general rule is that international agreements do not create obligations to third party states.75 Thus, when a state consents to obligations under an agreement, that state is not obligated to states which were not parties to that agreement. However, where Object States have consented to obligations to Subject States, which parallel those demanded by Subject States under another initiative, the OFCs have no basis to complain that there are third party states.76

Furthermore, even where there is no parallel obligation to Subject States, Subject States might constructively enjoy rights where the OFCs have made unilateral declarations that demonstrate an intention to perform in accordance with principles the Subject States are attempting to enforce. As the I.C.J. has expressed:

It is well recognized that declarations made by way of unilateral acts, concerning legal or factual situations, may have the effect of creating legal obligations . . . . When it is the intention of the State making the declaration that it should become bound according to its terms, that intention confers on the declaration the character of a legal undertaking, the [*PG268]State being thenceforth legally required to follow a course of conduct consistent with the declaration. An undertaking of this kind, if given publicly, and with an intent to be bound, even though not made within the context of international negotiations, is binding.77

While a unilateral statement is not an agreement, it may have legal consequences or may become a source of rights and obligations on principles analogous to estoppel.78 The doctrine of judicial estoppel prevents a party from asserting a position in legal proceedings that is contrary to the position previously taken by him in the same or some earlier legal proceeding.79 The I.C.J found that such an analogous principle bound France by its various public statements that demonstrated an intent to cease atmospheric tests of nuclear weapons.80 This Article refers to such obligations as estoppel obligations.

As with Consent, however, the intent of Object States to be bound legally is essential to Coincidence. Agreements or declarations made by states are “in many cases bereft of legal (though not political) significance, as the states may regard it a policy manoeuvre and not as setting up juridical relations between themselves.”81 Agreements and declarations of this sort are sometimes described as “soft law.”82 It is important to note, however, that such “soft law” is not legally binding.83

While both Consent and Coincidence rely upon the intent of the Object State to be bound, Consensus can produce obligations for States which have not expressed any such intent.84 Customary International Law is the principle means through which states are legally bound absent consent.85 The two elements necessary to demonstrate the existence of a norm in customary international law are state prac[*PG269]tice and opinio juris.86 As the I.C.J. expressed it in the North Sea Continental Shelf Case:

Not only must the acts concerned amount to a settled practice, but they must also be such, or be carried out in such a way, as to be evidence of a belief that this practice is rendered obligatory by the existence of a rule of law requiring it. The need for such a belief, i.e., the existence of a subjective element, is implicit in the very notion of the opinio juris sive necessitatis. The States concerned must therefore feel that they are conforming to what amounts to a legal obligation.87

The Court goes on to explain that state practice including “that of States whose interests are specially affected,” should have been both extensive and virtually uniform.88

State practice can be manifested in many ways including what states do in or through international organizations,89 or their participation in international agreements. Express statements by participating states are not necessary to demonstrate opinio juris; rather, it can be inferred from acts or omissions.90 Although custom can be built on acquiescence, broad dissent from a practice should preclude that practice from becoming custom. Furthermore, even where dissent does not bar the development of a custom, “a state that indicates its dissent from a practice while the law is still in the process of development is not bound by that rule even after it matures.”91

III.  Obligations and Their Sources

A.  Preliminary Considerations

It is important to note that there is substantial overlap between the evidence of the sources of obligations. For example, the potential distinctions between the sources are blurred by the fact that state [*PG270]practice includes the practice of the most interested states in order to form a new customary law. Because state practice can be manifest in the treaty obligations, there would be substantial overlap between the international agreements of the Object States and the customs which the state practice of those states supports.

Furthermore, despite the differences in the sources of obligations, all three of the sources depend upon the presence of a subjective element. Both Consent and Coincidence (including both parallel obligations and unilateral declarations) require that the Object State intended to bind itself by the obligation.92 Consensus, on the other hand, requires that the state practice manifest a belief that there is an obligation to act so, i.e. opinio juris.93 Therefore, although the first two demand an intent to create obligations and the third a recognition of obligations already existing, regardless of which is the source of an obligation, the absence of the required subjective element precludes the existence of that obligation.

The categorization of evidence might, consequently, be somewhat arbitrary, but a more thorough presentation would be redundant. The evidence of participation in conventional law therefore will be presented once either under the heading of Consent or Coincidence, and evidence of state practice outside multinational acts will be presented under the heading of Consensus with only brief mention of evidence offered by international agreements.

Regardless of these overlaps, in order for the Subject States to have sanctions available, it is absolutely essential that they be able to demonstrate an independent source, whether Consent, Consensus, or Coincidence, of the obligation for OFCs to comply with either the FATF’s 25 Criteria or the OECD’s 4 Criteria.

B.  25 Criteria

1.  Consent

Since there is no evidence that OFCs have consented directly to the 25 Criteria, there is no basis for the Subject States to claim that the Object States have agreed to be bound by that criteria.94 None of [*PG271]the OFCs identified have representation in the FATF.95 In fact, the Caribbean Community has specifically criticized the implementation of these criteria based on the fact that they were set unilaterally by FATF members, and Caribbean OFCs had no input into their creation.96 The lack of consent to assessment by the FATF according to the 25 Criteria also can be implied by the fact that Object States have consented to an alternate mechanism to assess compliance with the goal of preventing money laundering, namely mutual evaluation.97 External review by the FATF is a means of evaluation previously not considered by the Object States. Absent any evidence of consent, there is no basis to claim that OFCs have agreed to be bound by the 25 Criteria.

2.  Coincidence

A stronger argument can be made for requiring compliance with the 25 Criteria under coinciding legal obligations created either through parallel obligations or unilateral declarations. Parallel obligations between the Subject and Object States can derive from a number of sources. The most relevant ones include the United Nations Convention Against Illicit Traffic in Narcotic Drugs and Psychotropic Substances (UN Convention), the Model Regulations Concerning Laundering Offences Connected to Illicit Drug Trafficking and Other Serious Offenses promulgated by the OAS (OAS Model Regulations), the Council of Europe’s Convention on Laundering, Search and Confiscation of the Proceeds from Crime, and membership in the Egmont Group.98 Unilateral declarations include the statements of states in regional bodies that demonstrate their intent to be bound by the Forty Recommendations99 and the Nineteen Recommendations of the CFATF.

[*PG272]a.  Parallel Obligations

The UN Convention is the fundamental source of obligations in the area of money laundering.100 Coming into force in 1990, this convention represents the first global attempt to provide the law enforcement community with the necessary tools to undermine the financial power of the cartels and other groups in a manner sensitive to the requirements of international cooperation.101 Although the 25 Criteria do not expressly invoke the UN Convention, Recommendations 1, 4, 5, and 7 do. As of March 1, 2001, there were 158 member states to the convention.102 Six of the NCCs are not signatories to the UN Convention: the Cayman Islands, the Cook Islands, Liechtenstein, the Marshall Islands, Nauru, and Niue. The remaining nine all are bound by the UN Convention.

Many of the eleven Sample Object States also are bound under regional agreements. In 1990, the Council of Europe drafted the Convention on Laundering, Search and Confiscation of the Proceeds from Crime.103 Only Liechtenstein has signed, although not ratified, this treaty.104 The OAS promulgated its Model Regulations Concerning Laundering Offences in 1992 which were in conformity with the 1988 UN Convention.105 Of the eleven Sample Object States, the Bahamas, Dominica, Panama, St. Kitts and Nevis, and St. Vincent and the Grenadines are members of the OAS.

The Egmont Group, on the other hand, only includes one of the Sample Object States. The purpose of the Egmont Group, with fifty-three members,106 is to foster international cooperation against [*PG273]money-laundering through Financial Intelligence Units (FIUs). Among the states with operational FIUs are seven of the jurisdictions considered by the FATF in its June Report. Of these, only Panama was identified as non-cooperative.107

b.  Unilateral Declarations

In addition to the parallel obligations mentioned above, unilateral declarations can bind the Object States through estoppel. The most obvious source of estoppel obligations are those statements of states in regional bodies, which demonstrate their intent to be bound by the Forty Recommendations. The various regional anti-money laundering organizations—the APG, CFATF, and the Eastern and Southern Africa Anti-Money Laundering Group (ESAAMLG)—and the Council of Europe all have indicated support in varying degrees for the Forty Recommendations.

The use of these statements to identify an intent to be bound, however, has several obstacles. First, the Forty Recommendations were not promulgated as binding obligations. Second, although the regional organizations generally demonstrate a desire to promote these policies, this desire does not manifest itself as an intent to be bound. Third, because the endorsement by the Object States of the Forty Recommendations preceded the promulgation of the 25 Criteria, it cannot operate as an endorsement of the 25 Criteria. Finally, since the Object States have expressly consented to mutual evaluations as a means to assess compliance, there is minimal basis to argue that they intended to consent to evaluation by the FATF.108

The Forty Recommendations were not promulgated as binding obligations, and therefore, it is difficult to argue that states endorsing them demonstrate an intent to be bound by them. As Professor William Gilmore explains, “the 1990 report admits ‘the minimal standard we recommend can be viewed as rather ambitious’. It should also be stressed that, following normal international practice, these recommendations have no binding force as a matter of international law; customary or conventional.”109 The FATF is not a permanent interna[*PG274]tional organization nor is it a body managing a legally binding convention.110 Inherent in the title “The Forty Recommendations” is the concept that the goals therein are recommended, not obligatory. The FATF, when describing the Forty Recommendations, confirms this. The FATF website explicitly says that the Recommendations are “not a binding international convention,”111 and that the commitment of its members to them is “a firm political commitment,”112 not a legal commitment.

Furthermore, while the regional anti-money laundering bodies generally demonstrate a desire to implement the policies contained in the Forty Recommendations, they do not express an intent to bind themselves. In the instances that Object States have expressed support of the Recommendations, they demonstrate the lack of the subjective element on two levels: (1) states have expressed no intent to be bound; and (2) they feel no obligation exists outside their consent. Even as late as 1998, the APG, when expressing support for the Forty Recommendations, set forth the goal of encouraging113 their adoption by the member states while recognizing that “[e]ach jurisdiction [would] decide on the particular steps it [would] take to combat money laundering.”114 The Council of Europe similarly demonstrates support without expressing intent to be bound.115

The CFATF, on the other hand, has expressed an intent to be bound by both the Forty Recommendations and their own Nineteen Recommendations. In order to formalize its members’ efforts against money laundering, the CFATF adopted the Memorandum of Understanding (MOU).116 According to the MOU, “[m]embers agree to endorse and implement the [Forty FATF and the Nineteen CFATF] Recommendations [and] fulfil the obligations expressed in the King[*PG275]ston Declaration.”117 While this might define both the Forty and the Nineteen as obligatory for the CFATF members, the fact that the CFATF felt the need to formalize its commitment reinforces the argument that their is no obligation to the Forty Recommendations outside consent.

Regardless of acceptance of the Forty Recommendations, the endorsement by the Object States preceded the promulgation of the 25 Criteria; therefore, it cannot operate as an endorsement of the 25 Criteria. In 1992, the CFATF members, by virtue of Article 5 of the Kingston Declaration on Money Laundering, agreed to endorse and implement the Forty Recommendations.118 This commitment was reasserted in the 1996 MOU.119 Similarly, the APG expressed an intent to “[e]ncourage the adoption” of the Forty Recommendations in 1998;120 it has not reaffirmed its commitment since the promulgation of the 25 Criteria. The Council of Europe expressed its endorsement in 1997.121 The Cook and the Marshall Islands are Sample States not included in any of these organizations, and thus not party to any potentially binding unilateral declaration.122

3.  Consensus

In order to establish that the NCCs are bound by consensus, the Subject States must be able to show a settled state practice—including at least some of the Object States—which evidences a belief by the actors that the practice is obligatory. Ironically, the best way to assess [*PG276]state practice of the Object States is to simply look to the countries which have been identified as non-cooperative.123

The idea that the 25 Criteria are directly part of customary law can be readily dismissed. They were promulgated by an organization whose membership includes only twenty-nine states and were designed for application only against non-members. Of the twenty-eight countries considered by the report, fifteen were found to be non-cooperative. That means that over half of the interested object states do not manifest state practice in accord with the 25 Criteria. Unlike the Forty Recommendations, the 25 Criteria were not endorsed by other organizations. In fact, as mentioned above, the Caribbean Community has dissented from the application of the 25 Criteria.124 Four of the countries identified by the FATF as non-cooperative are members of the Caribbean Community: the Bahamas, Dominica, St. Kitts and Nevis, and St. Vincent and the Grenadines. Another three countries considered in the June report are also members: Antigua and Barbuda, Belize, and St. Lucia. Furthermore, the Cayman Islands, which is an observer in the organization, was identified as non-cooperative; and the British Virgin Islands, which is an associated member, and Bermuda, which is an observer, were also considered in the report.

The Caribbean Community’s dissent operates to frustrate the general acceptance of the 25 Criteria as customary law on two levels. First, it demonstrates a lack of the consensus that is necessary to form a new customary norm. Second, even if a new norm has evolved, it establishes that these countries are dissenting from the development of that norm. Such an objection should preclude the application of a new norm to the dissenters.125 While there is substantial state practice that is consistent with the goals set forth within the 25 Criteria, it is difficult to argue that these efforts demonstrate acceptance of the 25 Criteria since the vast majority of the efforts were implemented before the promulgation of the 25 Criteria.

While the 25 Criteria arguably could be deemed a codification of custom, a more viable argument is that the Forty Recommendations [*PG277]have become binding customary law and that the 25 Criteria are enforceable indirectly, insofar as they reflect the obligations of the Forty. Without going into an exhaustive analysis of the customary value of the Forty Recommendations,126 several strong foundations upon which such a claim might rest can be identified. Although in 1993 Bruce Zagaris described the Forty Recommendations as “soft law,” he went on to say that, “[a]lthough soft laws are not immediately binding, they are usually precursors to hard law.”127 Since then, the international acceptance that the Forty Recommendations enjoy has broadened substantially. Counting only the members of the FATF, the Council of Europe, the CFATF, the APG, and the ESAAMLG, over one hundred countries have committed to the implementation of the Forty Recommendations. While the majority of these are not FATF members, they still have effected legislation and policy changes to bring the Recommendations into effect. Some have even agreed to mutual evaluation to assess their progress in the implementation.128 All of these factors are evidence that there is an obligation to comply with the Forty Recommendations.

The extent of international support for the Forty Recommendations would warrant one of the following conclusions: (1) they have become part of customary international law through broad acceptance by the dominant nations and acquiescence by the rest; (2) they have become particular customary law for the countries that have committed to them; or (3) the countries that have made unilateral declarations accepting the Forty Recommendations would be estopped from claiming they are not bound by those Recommendations. If they are general customary law, the Forty are binding upon all nations. If they are particular customary international law, they still would be binding upon the majority of the countries identified as non-cooperative, i.e., all except the Marshall Islands, Nauru, and Niue.

[*PG278]B.  OECD’s 4 Criteria

As with the FATF initiative, for the OECD members129 to have sanctions available against Tax Haven Countries, these OFCs must be bound by an obligation, either by Consent, or Consensus, not to violate the 4 Criteria.

1.  Consent

Any attempt to claim that the Object States have accepted the 4 Criteria, in and of themselves, as binding obligations suffers the same frailties as the FATF’s 25 Criteria. None of the OFCs identified by the report is a member of the OECD,130 and none of those countries has expressed consent to these 4 Criteria in a forum outside the OECD.131 As with the FATF initiative, the Caribbean Community specifically has criticized the implementation of these 4 Criteria based on the fact that they were set unilaterally by OECD members and the fact that Caribbean OFCs had no input in their creation.132 Admittedly, six OFCs have expressed such consent and might be legally bound—or estopped from complaint—by the 4 Criteria, but this result would not impact their applicability to the thirty-five Tax Haven Countries that have not accepted them.

Even beyond the problem of lack of consent, it is hard to accept that the Tax Haven Guidelines were intended to be legally enforceable generally, let alone against OFCs. Support comes from a separate, but related, OECD publication which focuses on bank secrecy.133 The preface of this report concludes that the OECD members have “agreed to use this Report, whose proposed measures are not in any way binding as is the case with all OECD proposals in the tax area, as a basis for an on-going dialogue.”134 Though the 1998 OECD Harmful Tax [*PG279]Competition Report does not expressly declare that its recommendations are non-binding, it does imply, in its instructions to the CFA, a belief that they are not. First, it instructs the CFA to develop a dialogue with non-members.135 Second, it instructs the CFA to “encourage [non-members] to associate themselves with the recommendations set out in the Report.”136 Finally, the Report describes the endorsement of the Guidelines as a “clear political message.”137 These statements seem to imply that the OECD believed there was, at that time, no existent obligation for non-member states to comply with the recommendations.

Furthermore, the report admits that the guidelines on preferential tax regimes in member states are not binding upon non-member states.138 If the report were binding on non-members but not on members, there would be a double standard, thereby giving credence to the argument that the Tax Haven Guidelines are indeed unilaterally imposed standards that should have no actual legal impact.

2.  Coincidence

The Tax Haven Criteria do not enjoy the same odds for coincidental enforceability as the 25 Criteria do. There is little support for the OECD from parallel fiscal agreements and none of the Object States has made any unilateral declarations. Aside from the agreements promulgated in Europe, there are no conventions which could demonstrate broad acceptance of the principles contained in the 4 Criteria, and since none of the OFCs are signatories to these instruments, there is no basis for Subject States to argue that the Object States are bound by parallel obligations.

However, the OECD has promulgated two relevant Model Tax Conventions: the Convention on Mutual Administrative Assistance in Tax Matters (Model Convention on Assistance in Tax Matters)139 and the Model Tax Convention on Income and Capital (Model Tax Convention).140 Although these conventions are not binding agreements, the Model Tax Convention, due to its broad acceptance, can be analyzed for insights into the general state practice in fiscal matters.

[*PG280]3.  Consensus

It is very difficult to argue the existence of any customary obligations in the area of taxes; there is simply no broad consensus. The only state practice concerning international taxation is almost always restricted to intricate and heavily negotiated bilateral treaties; success in building consensus is limited to the broad acceptance of the Model Tax Convention. This is very different from the FATF initiative, which was born from a ten-year-old, broadly accepted effort; the OECD initiative appears from a void in relevant customary and treaty law. The fact that OECD countries offer termination of treaties as a sanction further demonstrates that they feel no obligations exist outside the treaty framework.

There is very little evidence of consensus in tax matters outside the area of double taxation, and efforts to form consensus with regard to mutual legal assistance (MLA) have proven largely unsuccessful. Generally, MLA instruments specifically exclude criminal tax matters.141 The most progress in the area of multilateral cooperation in tax matters has occurred in Europe. Under the 1978 Additional Protocol to the European Convention on Mutual Assistance in Criminal Matters, members of the Council of Europe abandoned their discretion to refuse assistance in relation to fiscal matters.142 Thirty-one states have ratified this protocol. Of the five Council of Europe states that were at risk of being identified as tax havens, Andorra, Cyprus, Liechtenstein, Malta, and San Marino, only Cyprus has ratified this protocol. Cyprus, Malta, and San Marino all have made an advance commitment to comply with the 1998 OECD Report. Only Liechtenstein and Andorra have no obligations.143 The various successes within the EU and among its members are relevant only on a regional level.144 They enjoy no application outside Europe, and none of the Object States is a member.

[*PG281] Attempts to extend such a uniform approach to legal assistance in tax matters outside Europe have not been as successful. In 1988, the OECD, together with the Council of Europe, drafted the Multilateral Convention on Mutual Administrative Assistance in Tax Matters.145 This convention has not been broadly accepted. While it has finally come into force seven years after it opened for signature, it has been ratified by only eight countries.146

III.  Using the Criteria to Assess Compliance

A.  Preliminary Considerations

The use of sanctions by OECD or FATF members for the breach of the principles contained in the 4 and 25 Criteria can be defended under two potential rationales. First, the criteria themselves can contain legally binding principles, and measures taken to remedy a breach of those principles therefore would be precluded from wrongfulness. Second, OFCs are prevented from complaining of the imposition of countermeasures for non-compliance with the criteria, either by parallel obligations or unilateral declarations, which indicate consent, or by customary obligations.

Because neither group of criteria is inherently obligatory, the category of Consent, as here defined, is not a viable source of obligation for the Object States.147 We are left, therefore, to analyze the extent to which those criteria operate as a means of assessing compliance with other obligations, whether existing by Consensus or created by Coincidence.

It is important to note that, whether the obligations are estoppel or legal obligations, the result is the same: OFCs would be precluded from complaining about the imposition of sanctions. Therefore, the distinction between obligations which might be incurred by unilateral declaration—and so, enforceable by estoppel—and obligations in[*PG282]curred by parallel agreement or consensus—and so, enforceable by law—is not relevant. Because of these factors, it is beneficial to deal with all of the obligations simultaneously in order to assess to what extent there is an overlap with the criteria.

Furthermore, because there is a multiplicity of criteria in each initiative, and because there is no formula for determining which of those criteria determine the inclusion of a jurisdiction on the list(s), demonstrating an absence of a state’s obligation to comply with any of the criteria throws the entire categorization into question.

B.  25 Criteria and Coinciding Obligations

It is most efficient to identify the obligations and the relevant 25 Criteria the FATF have promulgated under the four categories that the FATF established: (1) loopholes in financial regulations; (2) impediments set by other regulatory requirements; (3) obstacles to international cooperation; and (4) inadequate resources for preventing, detecting, and repressing money laundering activities. In this manner, one can identify which, if any, obligations coincide with the 25 Criteria which the FATF claims assess compliance with international standards, and identify which criteria cannot stand on this claim.

If we, for the sake of argument, presume coincidence between these obligations and the 25 Criteria, the problem still remains, however, for the Cook and Marshall Islands, Niue, and Nauru which are not signatories to any of those agreements.148 Furthermore, if we focus only on those countries bound by one or more of the above instruments, there are still a number of points on which the 25 Criteria have no support.

1.  Loopholes in Financial Regulations

Of the four, this category enjoys the most support from a variety of parallel instruments: the UN Convention, OAS Model Regulations, CFATF Nineteen Recommendations, the Council of Europe Initiative, and the Forty Recommendations. Regardless, aspects of some of the criteria in this category have no support from any of these instruments.

Criterion 1 speaks of adequate regulation of financial institutions, whether onshore or offshore. Recommendation 26, which cov[*PG283]ers the same concerns, makes no comment concerning financial institutions offshore. Criterion 7, which deals with obstacles to access to information by administrative and judicial authorities, exceeds Recommendation 12 by including information concerning the beneficial owner(s). Admittedly, Recommendation 11 requires the acquisition of information concerning the “true identity of the persons on whose behalf an account is opened . . . if there are any doubts as to whether these clients or customers are acting on their own behalf.”149 Criterion 7, however, shifts the burden even further by requiring financial institutions to “identify the beneficial owners when the identification of the account-holder is not sufficiently established.”150

Furthermore, neither regulation of offshore institutions nor the identification of beneficial owners is mentioned by the UN Convention, OAS Model Regulations, or the CFATF Nineteen Recommendations. Eight of the eleven Sample Object States have been identified as meeting Criterion 1. Five States meet, and the Cayman Islands partially meets, Criterion 7.

2.  Weaknesses in Commercial Requirements Including the Identification of Beneficial Ownership and the Registration Procedures of Business Entities

Criteria 13 and 14 and the accompanying comments continue the 25 Criteria’s overbreadth concerning beneficial ownership by shifting the focus onto the identification of beneficial owners of legal and business entities. These criteria claim to derive from Recommendations 9 and 25. Although number 25 warns of the potential for abuse of shell corporations, it does not recommend any particular means to address this problem. Specifically, it does not suggest addressing it in terms of customer identification nor does it expressly mention the concept of beneficial ownership. Recommendation 9, which does address customer identification, addresses it in the context of “financial activities as a commercial undertaking by businesses or professions which are not financial institutions.” These recommendations do not make clear the relationship between the identification of customers of non-financial institutions, and the abuse of shell companies. Moreover, the language of both recommendations is even weaker than that of others. Recommendation 9 says that the [*PG284]national authorities “should consider applying” the customer identification policies for non-financial institutions, while Recommendation 25 says that countries should “take notice of the potential for abuse.” These factors make it hard to see Criteria 13 and 14 as simply assessing the obligations under the recommendations.151

Furthermore, the identity of the beneficial owners of business entities is not mentioned by the UN Convention, OAS Model Regulations,152 or the CFATF Nineteen Recommendations. Six of the eleven Sample Object States are denoted as meeting Criterion 13 and seven as meeting 14.153 Altogether, ten of the eleven countries are found to meet one of these criteria, which are not fully covered by the recommendations or the other sources of obligation.

3.  Obstacles to International Cooperation Regarding Both Administrative and Judicial Levels

Criterion 18 stresses the use of the fiscal exception to restrict the level of cooperation between supervisory authorities. The fiscal exception is the name given to the refusal of cooperation on the grounds that such transactions may relate to tax matters.

The comment to Recommendation 15 expresses similar concern for the removal of the fiscal exception. However, this comment was only adopted on July 2, 1999. Consequently, it meets the same obstacles as the 25 Criteria for being promulgated after the Object States endorsed the Forty Recommendations.154 It has not been explicitly endorsed by those states.

Seven of the Sample Object States have been identified as violating Criterion 18.155 Four of the OAS members156 have been assessed as violating Criterion 18, despite the fact that the OAS Model Regula[*PG285]tions, the Nineteen Recommendations, and the UN Convention do not call for the elimination of the fiscal exception.

4.  Inadequate Resources for Preventing, Detecting, and Repressing Money Laundering Activities

Of the four, this category enjoys the least support from parallel instruments; it comprises three criteria. The first two, 23 and 24, deal with the lack of resources in public and private sectors; the third, 25, deals with the absence of a FIU or of an equivalent mechanism. None of these criteria is dealt with in the Forty Recommendations. Eight of the eleven relevant OFCs are denoted as meeting Criterion 23; two as meeting 24; and seven as meeting 25. Altogether, nine of eleven countries are found to meet one of these criteria, which are not even considered in the Forty Recommendations.

Of the parallel sources, only the CFATF’s Nineteen Recommendations expressly reference an obligation that adequate resources be dedicated to fighting money laundering.157 Although the OAS Model Regulations do recommend that the Inter-American Drug Abuse Control Commission assist in the procurement of sufficient resources,158 this is only a recommendation and has no specific binding nature.159

Furthermore, while FIUs might be more efficient than traditional mechanisms for dealing with money laundering, such mechanisms were not considered by the Forty Recommendations. The EGMONT Group of Financial Intelligence Units, which even the FATF identifies as a separate initiative, is the only example of an international organization attempting this specific goal.160 Of the Sample Object States, only Panama is a member of the EGEMONT Group, but since Panama did not meet Criterion 25, this Group is irrelevant to the inquiry.

[*PG286]C. 4 Criteria

OECD identifies these four criteria: (1) no or only nominal taxes; (2) lack of transparency; (3) no substantial activities; and (4) lack of effective exchange of information. Because there is sufficient evidence to demonstrate that neither the first nor fourth criterion can stand, I will elaborate upon the enforceability of only these two. With each of these criterion, we encounter two central problems. First, even to the extent there might be identifiable obligations under either Consensus or Coincidence, the expectations of the OECD exceed those identifiable limits. Second, the OECD members do not meet the criteria they attempt to impose on other states and therefore should be precluded from enforcing them.

1.  No or Only Nominal Taxes

Of the sources identified above, aside from the 4 Criteria, only the European Code of Conduct attempts to impose an obligation to limit, or rather, raise, tax levels. Since none of the Object States is a party to this initiative, none has expressed an intent to be bound. Neither have the Object States expressed any awareness of a prior obligation.

This criterion is concerned with minimizing “tax induced distortions” that derive from the disparities in tax costs between jurisdictions.161 The degree of incentive to shift income from one jurisdiction to another is directly related to the difference between their respective tax levels. There would be as much incentive to shift income from Sweden to the US, with a tax disparity approaching 30%, as there is to shift from the US to one of the Object States, where the tax disparity would also be about 30%.162 In this sense, OECD members demand from Object States more than they demand from each other.

2.  Lack of Effective Exchange of Information

This criterion has a possible overlap with FATF’s 25 Criteria and might, therefore, attempt to borrow any legal strength from that source. Under this title, the OECD specifically discusses issues concerning secrecy and customer identification, both of which the FATF [*PG287]identifies under the title Loopholes in Financial Regulations.163 Any overlap that might exist here should be bypassed by the fiscal excuse. To the extent that the Forty Recommendations are hard law, the fiscal excuse could be seen as obviated.164 But the elimination of this excuse would be largely ineffectual for tax authorities because the use of information, even under the Forty Recommendations, is limited to the enforcement of money laundering crimes. It has no impact on the use of information for tax matters except to the extent that participants have added tax crimes to their lists of “serious offences.”

Furthermore, the OECD denies Object States the option to apply traditional principles that limit international cooperation such as reciprocity, but it allows its members the discretion to determine whether they will participate, and to what extent, in any exchange of information. Perhaps the most obvious example comes from Switzerland, which has an express reservation on the OECD Model Tax Convention Article on the exchange of information.165 A more startling example, however, comes from Britain. As the Commentary to the Model Tax Convention explains:

The United Kingdom takes the view that the Article imposes no obligation on it to carry out enquiries on behalf of a Contracting State in cases where no liability to United Kingdom tax is at issue, since to carry out such enquiries would be contrary to its laws and administrative practice.166

Unless there is a separate claim under English or Scottish revenue law, a UK court will decline jurisdiction over a suit to enforce the revenue [*PG288]law of another country167 and will refuse to enforce a ruling of a foreign court concerning that forum’s revenue law.168 This policy runs exactly counter to the principle described in the report stating, “Harmful tax competition which leads to tax evasion by tax payers of other countries may be encouraged if one country will not enforce the tax claims of another country.”169

The OECD further expects OFCs to abandon the principle of reciprocity as a basis to refuse cooperation. For example, the report says that it is not sufficient for an OFC to exchange information concerning criminal tax matters related to other crimes or concerning tax fraud if that jurisdiction does not assist with collecting information concerning tax avoidance.170 Implicit in this statement is the expectation that an OFC must supply information on tax avoidance even where that OFC might not collect the tax in question or even consider avoidance a crime. Such is the case for the Cayman Islands, which collects neither corporate nor personal income tax.171

This premise runs counter to the principle of reciprocity. According to the OECD’s own Model Tax Convention, “[t]he exchange of information can take place only insofar as the tax concerned is covered by the Convention.”172 Consequently, where a state has no income tax, treaties with that state might not include the exchange of income tax information. In the words of one of the OECD’s own reports:

Under legal reciprocity the requested State is under no obligation to carry out administrative measures that are not permitted under its own laws and practice or to supply items of information that are not obtainable in the normal course of its administration. . . . [Legal reciprocity] establishes a kind of ‘minimum position’ whereby the requested State is obliged to give no more assistance than that which the requesting State is capable of providing at the national level; [*PG289]moreover, the requested State need supply no more than is obtainable under the normal practice for that State. This does not mean that a more extensive assistance is excluded, but that the requested State need not comply with the request.173

Even the further reaching OECD MLA Model provides for limits by domestic law according to principles of legal reciprocity.174

Even within the EU, a region that enjoys relatively harmonized tax policies, member states have no obligation to offer assistance where it would be contrary to their present domestic laws. Of twelve EU countries surveyed–Austria, Belgium, Denmark, Finland, Germany, Ireland, Italy, Netherlands, Portugal, Spain, Sweden, and the United Kingdom–all twelve said their tax authorities are limited when the information is not available under national law or domestic law and when administrative practices prevent inquiries being carried out to provide the information, or prevent the collection and use of such information.175 Eleven of the twelve participants were required to refuse exchanging information when to do so would be contrary to public policy. Only Denmark was not so restrained.

Conclusion

Neither the FATF nor the OECD enjoy firm legal grounds to claim that sanctions would be available against states that refuse to co-operate with their respective initiatives.

Admittedly, the FATF is on better footing than the OECD, but it is far from sure footing. Since the 25 Criteria are not themselves binding, the obstacles come from the disparities between actual binding obligations and the expectations within the 25 Criteria. The first obstacle is the fact that some of the Object States, such as the Marshall Islands, Nauru, and Niue, have not bound themselves to any of the parallel instruments. Ignoring this obstacle, the parallel obligations that do exist do not support all of the criteria. Even the Forty Recommendations, which are themselves a dubious source of binding [*PG290]obligations, do not support all of the demands that the 25 Criteria make. Consequently, it would be a difficult task for the FATF members to argue any use of sanctions is legal.

The same task for OECD members, on the other hand, would be Herculean. There is no identifiable source under which Object States could be said to be bound to comply with the expectations of the 4 Criteria, whether Consent, Coincidence, or Consensus. Furthermore, even the OECD members do not meet those same expectations and therefore should be precluded from attempting to enforce them, even if there were some obligation for the Object States.

Subject States should take heart in two factors: (1) to the extent that the measures they implement do not constitute sanctions under international law, they are not precluded; and (2) there is still room for further negotiation on the subjects of money laundering and harmful tax competition. Unfriendly acts that do not amount to sanctions would be deemed acts of retorsion and are not prohibited under international law.176 Furthermore, most OFCs have indicated only support for the international anti-money laundering efforts. In fact, the Caribbean Community, while criticizing the G7, reaffirmed its position against money laundering.177 There is even hope in the struggle against harmful tax competition since strong arguments can be made that such competition is harmful to OFCs in addition to OECD member states.178 But there is still a long way to go before a comprehensive system of international legal obligations will bind OFCs and others alike.