¹ See MJ Casey, ‘World's “Unbanked” En Route to Financial Inclusion With Mobile Money’ (Money Beat, Wall Street Journal, 5 November 2014) <http://blogs.wsj.com/moneybeat/2014/11/05/worlds-unbanked-en-route-to-financial-inclusion-with-mobile-money/>.

² The Consultative Group to Assist the Poor, ‘Financial Inclusion’ (Financial Inclusion) <http://www.cgap.org/topics/financial-inclusion>.

³ ibid.

⁴ See Alliance for Financial Inclusion, ‘Mobile Financial Services: Basic Terminology’ (2013) 1.

⁵ See World Bank, ‘Kenya Economic Update: Kenya at the Tipping Point? With a Special Focus on the ICT Revolution and Mobile Money’ (2010) 23.

⁶ Financial inclusion means affordable access to financial services. See The Consultative Group to Assist the Poor (n 2).

⁷ See C Scharwatt et al., ‘State of the Industry 2014: Mobile Financial Services for the Unbanked’ (GSMA 2015) 26.

⁸ In 2014, 53 per cent of live mobile money services were in sub-Saharan Africa. However, half of all new launches occurred outside that region, most notably in Latin America and the Caribbean, East Asia and the Pacific and South Asia. Competition is also increasing: see ibid 14–16.

⁹ See Greenacre, J and Buckley, RP, ‘Using Trusts to Protect Mobile Money Customers’ (2014) 2014 SingJLS 59Google Scholar.

¹⁰ Excepting Belize and Guyana, all countries in Latin America have a civil law tradition, as do the majority of countries in Africa and South East Asia. See M Lupoi, Trusts: A Comparative Study (Cambridge University Press 2000) ch 5.

¹¹ See eg F Barrière, ‘The French Fiducie, or the Chaotic Awakening of a Sleeping Beauty’ in LD Smith (ed), Re-imagining the Trust: Trusts in Civil Law (Cambridge University Press 2012); Figueroa, D, ‘Civil Law Trusts in Latin America: Is the Lack of Trusts an Impediment for Expanding Business Opportunities in Latin America?’ (2007) 24 ArizJIntl&CompL 701Google Scholar.

¹² The question of how to fit trusts in civil law jurisdictions has been extensively debated. See eg de Waal, MJ, ‘In Search of a Model for the Introduction of Trusts into a Civilian Context’ (2001) 12 StellenboschLRevGoogle Scholar; T Honoré, ‘On Fitting Trusts into Civil Law Jurisdictions’ (2008) Oxford Legal Studies Research Paper No 27/2008 <http://ssrn.com/abstract=1270179>.

¹³ We have selected developed countries that represent different approaches to the accommodation of common law trusts and developing countries where regulations of e-money services already exist.

¹⁴ See Directive 2009/110/EC of 16 September 2009 on the taking up, pursuit and prudential supervision of the business of electronic money institutions (hereinafter, the ‘E-Money Directive’), art 2.

¹⁵ Based on M Tarazi and P Breloff, ‘Nonbank E-Money Issuers: Regulatory Approaches to Protecting Customer Funds’ (CGAP Focus Note 2010) 2.

¹⁶ In many developing countries, bank branches are not widespread. They can use the agency network of MNOs to reach out to potential customers.

¹⁷ eg in the case of Colombia, see Ley de Inclusión Financiera de 4 de junio de 2014, art 1.

¹⁸ This is the case in Colombia and Mexico. See Alliance for Financial Inclusion, ‘Regulatory Approaches to Mobile Financial Services in Latin America’ (2014) 5–7.

¹⁹ This excludes Internet-based payment systems such as PayPal, which require bank accounts.

²⁰ eg Paraguay: see Resolución n° 6 del Banco Central de Paraguay (BCP), de 13 de marzo de 2014, which establishes the Reglamento de Medios de Pago Electrónicos (hereinafter, ‘Paraguayan E-Payments Regulation’).

²¹ This description is based on the highly successful M-PESA service in Kenya. See M Klein and C Mayer, ‘Mobile Banking and Financial Inclusion: The Regulatory Lessons’ (World Bank Policy Research Working Paper 5664 2011) 7–10.

²² J Greenacre and R Buckley, ‘Trust Law Protections for E-Money Customers: Lessons and a Model Trust Deed Arising from Mobile Money Deployments in the Pacific Islands’ (2013) 8 <http://www.afi-global.org/sites/default/files/publications/piwg_knowledge_product_e-money_trust_and_model_trust_deed.pdf>.

²³ The term ‘fund isolation’ is often used in the context of mobile money services. See Alliance for Financial Inclusion (n 4) 3.

²⁴ We can expect MNOs to hold securities in an account with a securities intermediary that will have an account at the central securities depository.

²⁵ See Greenacre and Buckley (n 9) 63–5. There have been recent concerns with the possibility of customers’ funds being stolen while stored with the Provider. In 2012, workers at the telecommunications provider MTN in Uganda stole 15 billion Ugandan shillings. See J Mbanga, ‘How MTN Lost Mobile Billions’ (The Observer, 24 May 2012) <http://www.observer.ug/index.php?option=com_content&view=article&id=18921:how-mtn-lost-mobile-billlions>.

²⁶ See Greenacre and Buckley (n 22) 10.

²⁷ For a thorough analysis of the general principles of trusts law, see GW Thomas and A Hudson, The Law of Trusts (2nd edn, Oxford University Press 2010).

²⁸ See Greenacre and Buckley (n 9) 64.

²⁹ Trusts law has an important distinction between legal ownership and beneficial ownership that should be carefully considered.

³⁰ See Greenacre and Buckley (n 9) 68.

³¹ See ibid 67.

³² See ibid 70.

³³ LD Smith, ‘The Re-imagined Trust’ in LD Smith (ed), Re-imagining the Trust: Trusts in Civil Law (Cambridge University Press 2012) 258.

³⁴ Some scholars argue the differences are relatively unimportant and that there are ways of adapting trusts to civil law jurisdictions, as has occurred in Scotland and South Africa. For an analysis of the suitability of trusts in civil law jurisdictions, see eg de Waal (n 12) 66.

³⁵ In France and in many Latin American jurisdictions, these approximations aim at improving the appeal of certain jurisdictions to foreign investors, particularly from England and the US. See Barrière (n 11); Figueroa (n 11).

³⁶ The most relevant civil law country to have ratified the Convention is Italy. See M Graziadei, ‘Recognition of Common Law Trusts in Civil Law Jurisdictions under the Hague Trusts Convention with Particular Regard to the Italian Experience’ in LD Smith (ed), Re-imagining the Trust: Trusts in Civil Law (Cambridge University Press 2012).

³⁷ Default rules fill the gaps in incomplete contracts. They govern unless parties contract around them by prior agreement. See Ayres, I and Gertner, R, ‘Filling Gaps in Incomplete Contracts: An Economic Theory of Default Rules’ (1989) 99 YaleLJ 87Google Scholar, 87.

³⁸ See eg arts 1281–1288 of the Spanish Civil code.

³⁹ Civil codes provide complete sets of ‘default rules’. These default rules may be contained in (1) the general part of the Law of Contracts (as in France or Spain, and the codes inspired by them) and/or in the general part on the Law of Obligations (as in Germany); or (2) in the parts of the codes dedicated to the specific ‘types’ of contracts (sales, services, mandate, etc). This makes the ‘classification’ or ‘qualification’ of the contract a previous first step in order to determine which default rules are applicable. From a practical perspective, Spanish courts constantly resort to this technique to determine the default rules applicable to atypical contracts. As a recent example, see the decision of the Spanish Supreme Court of 24 October 2014 (RoJ 4811/2014) to determine the ‘nature’ of an escrow account contract. From an academic perspective, the technique is quite marked in the French tradition. See eg X Henry, La Technique des Qualifications Contractuelles (Thèse de doctorat, Université de Nancy, 1992) 50–76, who puts this technique in relation with general concepts, such as the objet, or the cause of the contract. See also Overstake, JF, ‘Essai de classification des contrats spéciaux’ (1969) LGDJ 12ffGoogle Scholar, or Frechette, P, ‘La qualification des contrats: aspects théoriques’ (2010) 51(1) Les Cahiers de Droit 117–58Google Scholar, who also studies the approach under the Laws of Quebec. In Germany, the specific rules for each type of contract are construed as default rules, which require a previous classification. See B Markesinis, H Unberath and A Johnson, The German Law of Contract: A Comparative Treatise (2nd edn, Hart Publishing 2006) 138–44, 162–4.

⁴⁰ Common law judges are also prone to such biases. Law-and-economics scholars suggest that generally courts tend to require parties to have reached a ‘sufficient’ agreement to derogate from default rules, and often use a strict interpretation of the sufficiency of the agreement in a way that turns default rules into something closer to mandatory rules. See Ayres and Gertner (n 37) 120–3; Goetz, CJ and Scott, RE, ‘The Limits of Expanded Choice: An Analysis of the Interactions between Express and Implied Contract Terms’ (1985) 73 CLR 261, 263Google Scholar; Schwartz, A and Scott, RE, ‘Contract Theory and the Limits of Contract Law’ (2003) 113 YaleLJ 541Google Scholar, 564ff. In civil law countries courts may be even more prone to this exercise due to the confluence of ‘specific’ default rules (eg the rules of the different contract types, such as sale, mandate, deposit, etc) and ‘general’ default rules, which may constitute an autonomous source of obligations, such as the duty of good faith. Such open textured provisions can be a source of uncertainty in some civil law countries. For a reference to the German law tradition, see Markesinis, Unberath and Johnson (n 39) 119–33.

⁴¹ Some view the causa as a basic requirement of validity, like common law consideration, though in reality the concept is the pivot to evaluate the validity of the contract pursuant to its finality, or (under some ‘objectivistic’ views of the causa) economic function. See L Díez-Picazo, Fundamentos Del Derecho Civil Patrimonial (vol I, Thomson Civitas 2007) 266–85.

⁴² Under this doctrine, there must be a closed number of rights receiving privileged protection against third parties. A party cannot grant another party privilege over assets (right in rem) without good reason (causa). If the parties create a privilege outside the rights in rem contemplated by the law, or use a right in rem protected by the law in a way incompatible with its function, the arrangement may be held invalid. If the arrangement is held valid, the gaps left by the parties will be filled by the default rules of the specific right in rem. See L Díez-Picazo, Fundamentos Del Derecho Civil Patrimonial (vol III, Thomson Civitas 2008) 131ff.

⁴³ There are seemingly two legal traditions. The ‘Roman fiducia’ limits the power of abuse of the fiduciary by the negative binding obligation. By contrast, in the ‘Germanic fiducia’ any use contrary to the end sought results in the return of the object to the settlor or the settlor's heirs, even if it negatively impacts a third purchaser: S Cámara Lapuente, ‘Trusts in Spanish law’ in M Cantin Cumyn (ed), Trust vs Fiducie in a Business Context (Bruylant 2000) 197. Common law trusts and civil law fiduciary instruments may have a common base in the Roman-Canonical usus. See de Waal (n 12) 65; Lupoi (n 10) 185. However, it is generally understood that the beneficiary under a fiducia is not equivalent to the beneficiary under a trust. For a detailed analysis of the fiducia and the common law trust, see Figueroa (n 11) 4–7, 23–32; R Sánchez Aristi, Property and Trust Law in Spain (2nd edn, Kluwer Law International 2014) para 243.

⁴⁴ For an overview of the differences in different jurisdictions, see Barrière (n 11); Lupoi (n 10) 273–91; M Cantin Cumyn, ‘Reflections Regarding the Diversity of Ways in Which the Trust Has Been Received or Adapted in Civil Law Countries’ in LD Smith (ed), Re-imagining the Trust: Trusts in Civil Law (Cambridge University Press 2012).

⁴⁵ See eg Cámara Lapuente (n 43) 195–8.

⁴⁶ See eg French Civil Code, art 2012.

⁴⁷ See eg Ley 17.703, signed into law 4 November 2003, regulating fideicomiso in Uruguay (hereinafter, ‘Uruguayan Fideicomiso Act’).

⁴⁸ Usually banks and other authorized financial institutions. See eg Luxembourg Fiduciary Contracts Act, art 4; art 2015 of the French Civil Code; art 385 of Mexico's General Act on Credit Transactions; or art 5 of Argentinian Act 24.441.

⁴⁹ The fiduciary contract will specify who these beneficiaries are. In some jurisdictions, the settlor or any third party as specified in the contract may be beneficiaries of the fiducia. See Cámara Lapuente (n 43) 194; M Virgós Soriano, El trust y el derecho español (Thomson/Civitas 2006) para 43. In others, even the fiduciary may be designated as beneficiary. See French Civil code, art 2016. This, however, is prohibited in many Latin American jurisdictions. See art 8.1 of Paraguayan Ley 921/96 de Negocios Fiduciarios (hereinafter, ‘Paraguayan Fiduciary Contracts Act’); or art 265.4 of Peruvian Ley 26.702, General del Sistema Financiero y del Sistema de Seguros y Orgánica de la Superintendencia de Banca y Seguros (hereinafter, ‘Peruvian Financial System Act’).

⁵⁰ In Bolivia, eg, see section 5, art 3 of the Resolución Autoridad de Supervisión del Sistema Financiero (ASFI) no 835/2011 (hereinafter, ‘Bolivian Regulation on M-Money Providers’). In Peru, see art 15 of the Resolución de la Superintendencia de Banca y Seguros (SBS) no 6283-2013 that regulates the Reglamento de Operaciones con Dinero Electrónico (hereinafter, ‘Peruvian Regulation on E-Money Transactions’).

⁵¹ This is less common in Latin American jurisdictions. In Uruguay, see Ley no 19.210 de Inclusión Financiera (hereinafter, ‘Uruguayan Financial Inclusion Act’), art 5.

⁵² A patrimony can be broadly defined as an autonomous mass with a set of assets answerable for the set of liabilities. See Barrière (n 11) 251.

⁵³ In France, in the absence of any beneficiaries, the fiduciary assets return to the settlor. See Code Civil Français, arts 2029.2, 2030.

⁵⁴ See eg Code Civil Français, art 2024.

⁵⁵ See Code Civil Français, art 2025.

⁵⁶ See Barrière (n 11) 250–4.

⁵⁷ For an account of some of the problems these solutions pose, see ibid. Other solutions have been proposed, but they would not fit the particularities of e-money customers.

⁵⁸ Cantin Cumyn (n 44) 7–8.

⁵⁹ See Civil Code of Québec, arts 1261, 1265. Unlike in the French fiducie, where the fiduciary has title to the real rights in the property that has been put into the fiducie. See Barrière (n 11) 239. In Peru, the fiduciary patrimony also seems to be an autonomous patrimony from those of the settlor, the fiduciary and the beneficiaries. See Texto Concordado de la Ley General del Sistema Financiero y del Sistema de Seguros y Orgánica de la Superintendencia de Banca y Seguros (hereinafter, ‘Ley no 26702 SBS Peru’), art 241. The language, however, is somehow confusing as it refers to ‘fiduciary ownership’ (dominio fiduciario). In Paraguay, despite the legal reference to an ‘autonomous patrimony’ (patrimonio autónomo), the legal nature and effects of the fiduciary patrimony seem closer to those of a separate patrimony of the fiduciary. An important element in this conclusion is the termination of the fiduciary contract upon the liquidation of the fiduciary. See Paraguayan Fiduciary Contracts Act, art 41.6.

⁶⁰ See Civil Code of Québec, arts 1296 and 1297. A new fiduciary will be appointed according to the terms provided by the settlor in the contract or as determined by the court. See Civil Code of Québec, art 1277. See also Ley 26702 SBS Peru, art 253.

⁶¹ See eg art 71. VII. E) Mexican Act on Business Insolvency; see also art 2024 French Code civil, or art 155 (1) of the Italian Insolvency Act; Bolivian Commercial Code, art 1410; Paraguayan fiduciary Contracts Act, arts 10, 13; Peruvian Financial System Act, arts 241, 254.

⁶² See eg art 2027 French Code civil.

⁶³ In an e-money transaction customers purchase e-money units issued by the Provider, who gains a proprietary right over the money used to pay for those units. See Section IIA.

⁶⁴ See eg art 386 of Mexico's General Act on Credit Transactions; arts 2447sexies and 2447septies of the Italian Codice civile; art 1259 Quebec Civil code.

⁶⁵ See eg Paraguayan Fiduciary Contracts Act, art 10 and 13; Uruguayan Fideicomiso Act, art 6.

⁶⁶ See eg French Code civil, arts 2021, 2022; Luxembourg Fiduciary Contracts Act, arts 6, 7.

⁶⁷ See eg Uruguayan Fideicomiso Act, art 4.3; French Code civil, arts 2018.6°, 2022, 2026; Luxembourg Fiduciary Contracts Act, art 7(3).

⁶⁸ See eg French Code civil, arts 2018.6°, 2022, 2026; Mexico's General Act on Credit Transactions, art 391. One exception is Argentina: see eg Argentinian Act no 24.441, arts 4.d), 6 and 7. One imaginative solution is that of Luxembourg, where the fiducie is primarily defined on proprietary terms, but a reference is made to the law of mandate contract to fill the gaps in the duties of the fiduciary. See Luxembourg Fiduciary Contracts Act, art 7(1).

⁶⁹ See Thomas and Hudson (n 27) 33.01–33.120. For tracing in general, see L Gullifer (ed), Goode on Legal Problems of Credit and Security (4th edn, Sweet & Maxwell 2008) 1–57, 41.

⁷⁰ Martín Padilla, ‘La formación del concepto de subrogación real’ (1975) 1111; Roca Sastre ‘La Subrogación real’ (1949) 281; Vallet de Goytisolo, ‘Pignus tabernae’ (1953) 483.

⁷¹ Westdeutsche Landesbank v Islington [1996] AC (Lord Browne-Wilkinson).

⁷² See eg Argentinian Act no 24.441, art 7.

⁷³ See eg Luxembourg Fiduciary Contracts Act, art 7(6).

⁷⁴ See Civil Code of Québec, art 1290.

⁷⁵ See Figueroa (n 11) 725–6.

⁷⁶ Some jurisdictions mandate that a fideicomiso is settled for 100 per cent of e-money issued and in circulation. See eg Paraguayan E-Payments Regulation, art 15. Brazil represents an interesting case because it requires the Provider to guarantee the e-money issued following a progressive scale with a 20 per cent yearly increase of the total proportion of e-money guaranteed from 2016 to 2019. By 2019, Providers must guarantee that 100 per cent of the e-money issued is guaranteed. See Circular no. 3681 de 4 de Novembro de 2013 do Banco Central do Brasil, art 12.9. In other jurisdictions, the intermediation of e-money customers’ funds deposited in bank accounts is expressly prohibited. See eg Uruguayan Financial Inclusion Act, art 6.

⁷⁷ It is very common among Latin American regulators to restrict the securities in which e-money customers’ funds can be invested to securities issued by the federal government or central bank. See eg Reglamento de Fideicomiso contenido en la Recopilación de Normas para Bancos y Entidades Financieras (RNBEF), Chapter XVII (hereinafter, ‘Bolivian Regulation on Fideicomiso’), art 12; Circular BC Brasil, no. 3681, art 12.1.II; Peruvian Regulation on E-Money Transactions, art 16.

⁷⁸ Auditing can help ensure the integrity of the system. See Klein and Mayer (n 21) 13.

⁷⁹ If the delegation of supervisory powers were to be challenged, a court could find that some default rules also allow the settlor to delegate those powers. See Section IIIB.

⁸⁰ For example, the Civil Code of Québec allows the settlor or beneficiary to delegate their monitoring powers. See Civil Code of Québec, arts 1287ff. Art 2017 of the French Civil Code gives the settlor similar powers but not the beneficiaries.

⁸¹ See eg Paraguayan Fideicomiso Act, art 25.13.

⁸² In Brazil, for example, see Circular no. 3682 de 4 de Novembro de 2013 do Banco Central do Brasil (hereinafter, ‘Circular BC Brasil no 3682’), art 22.

⁸³ See Organisation for Economic Cooperation and Development, ‘PISA 2012 Results: Students and Money: Financial Literacy Skills for the 21st Century (Volume VI)’ (2014) <http://www.oecd.org/pisa/keyfindings/pisa-2012-results-volume-vi.htm>.

⁸⁴ This seems to be the most common situation in Latin American jurisdictions. For example, in Peru, the Banking and Insurance supervisor is in charge of monitoring authorized e-money issuers. See Peruvian E-Money Act, art 6.3. In Paraguay, the Central Bank undertakes those supervisory functions. See Paraguayan E-Payments Regulation, art 20.

⁸⁵ See eg French Civil Code, art 1984ff.

⁸⁶ A replevin action (acción reivindicatoria) will be available to the principal. See Sánchez Aristi (n 43) 252.

⁸⁷ See eg Spanish Commercial code, art 225; Spanish Supreme Court decision of 5 February 1964.

⁸⁸ See eg German Civil code, section 276.

⁸⁹ eg in Paraguay, the Paraguayan Fiduciary Contracts Act distinguishes between those fiduciary contracts under which property is transferred to the fiduciary and those under which property remains with the settlor. The latter are referred to as encargo fiduciario or ‘fiduciary mandate’.

⁹⁰ See Section IIC.

⁹¹ In the Lehman Brothers’ saga, English courts had to make a similar decision on whether customers’ funds should be protected. The outcome was for a period uncertain despite the legal institution used to protect customers’ money being a trust. See In the matter of Lehman Brothers International (Europe) (In Administration), [2012] UKSC 6; on appeal from: [2010] EWCA Civ 917. The first case to be decided was In re Lehman Brothers International (Europe) [2009] EWHC 3228 (Ch). If the legal institution used was a mandate, which has no explicit bankruptcy protection rules, the outcome would have been even more uncertain.

⁹² For a distinction between mandate contracts and trusts, see Figueroa (n 11) 735.

⁹³ The same considerations made regarding the possibility of customers delegating their supervisory powers over the fiduciary's duties upon a third party expert would also apply here. See Section IIIA.

⁹⁴ In the EU, the Directive 2015/2366/EU, of 25 November 2015 (hereafter the ‘Payment Services Directive II’, or PSD II) adopts such functional approach by imposing upon payment systems (1) safeguard measures (art 10(1)); and (2) conduct duties of the payment services provider (arts 38ff). The Directive 2009/110/EC of 16 September 2009 (the ‘E-Money Directive’) requires similar safeguarding mechanisms. For duties applicable to e-money issuers see arts 10 to 13 of the E-Money Directive. The division of competences between the EU and the Member States has helped the EU develop a functional approach, which can be a useful blueprint for countries struggling with how to properly implement the different functions in their civil law codes. It is important to note, however, that under EU law, Directives set goals that all Member States need to achieve. Member States will decide how best to achieve those goals by implementing EU Directives into their legal systems. As a result, although they will all aim at the same objective, Member States might resort to different legal mechanisms and different regulatory instruments in their pursuit. For an analysis of the transposition status of the E-Money Directive in each of the EU Member States, see <http://ec.europa.eu/finance/payments/emoney/transposition/index_en.htm#maincontentSec2 (updated as of May 2012)>.

⁹⁵ See arts 10(1)(a) and (b) PSD II.

⁹⁶ See Payment Services Directive II, art 10.1(b); E-Money Directive, art 7.1.

⁹⁷ Insurance companies require large numbers of clients in order to avoid the risk of facing numerous simultaneous payouts that would deplete its resources in a short period of time.

⁹⁸ Given the financial pressure borne by guaranteeing e-money customers’ funds, one could expect these companies and public entities to have a strong incentive to monitor the Provider of mobile services.

⁹⁹ See Section IIIA.

¹⁰⁰ See eg Peruvian E-Money Act, art 6.1.

¹⁰¹ This particular approach is likely to negatively impact ordinary creditors of the Provider. Accounting principles should be adapted so that the Provider's financial accounts reflect a true picture of its solvency.

¹⁰² Instead of isolating funds via a specific property right the law would require Providers to create such separate patrimony through a different entity. This is a common approach in many Latin American jurisdictions, where e-money issuers are required to provide e-money services through a separate institution. The Provider will also need to ensure that each entity carries its own accounting books. See eg Reglamento de Fideicomiso contenido en la Recopilación de Normas para Bancos y Entidades Financieras (RNBEF), Chapter XVII, art 7; in Brazil, Circulares BC Brasil no 3682 y 3683; in Peru, Resolución SBS no 6286-2013.

¹⁰³ In some countries the regulation of certain financial services, including e-money, has fostered the development of fiduciary contracts. See de Waal (n 12); Figueroa (n 11); Lupoi (n 10) 275, 285, 291.

¹⁰⁴ These reasons include, among others, political reasons, incompatibility within the country's legal tradition or the broader legal system, implications in other areas beyond mobile money services for which the relevant jurisdiction may not be ready. Additionally, such a deep reform may take a long time to pass and regulators may feel the urgent need to protect e-money customers’ funds.

¹⁰⁵ See section IIIA.

¹⁰⁶ In Paraguay, for example, art 25 of the Paraguayan Fideicomiso Act expressly enumerates a series of obligations of the fiduciary that she cannot delegate.

¹⁰⁷ In the EU, the E-Money Directive has adopted this approach due to the absence of an EU regulatory framework for fiduciary contracts that would guarantee a more or less homogenous regulation of fiduciary contracts in all EU Member States. See EU E-Money Directive, art 7.1.

¹⁰⁸ eg the type of bank where customers’ funds can be deposited and whether each customer's funds would be separated from other customers’ funds, among others.

¹⁰⁹ Two of the multiple options available would be (1) to have a ‘framework’ contract subject to the laws of mandate, followed by a fiduciary contract that regulates the transfer and segregation of funds, or (2) a single contract subject to the laws of fiduciary transactions that operates as the broader contract, and which includes specific operational and conduct duties by the Provider, where a specific reference is made to the laws of mandate.

¹¹⁰ See arts 5–9, 11, 13, 16, or 18 of the PSD II, and art 10 of the E-Money Directive.

¹¹¹ In the case In re Lehman Brothers International (Europe) (In Administration), [2012] UKSC 6 the UK Supreme Court decided that a trust arose as a result of the application of statutory provisions (in transposition of Directive 2004/39/EC (MiFID) only after a lengthy process, and careful argumentation. Furthermore, the decision that, in addition to a trust, the specific recovery procedure envisaged by the rules protected all clients’ moneys was adopted by a 3–2 majority. The two dissenting judges (Walker and Hope) argued that, in spite of statutory intent, there was no way that property law and trust law could be stretched to encompass funds that had not been actually segregated (and, also, that this left all clients without a high degree of protection).

¹¹² In Britain the vast body of trust law is supplemented by very specific rules that regulate financial companies’ duties for handling customers’ money. See eg Financial Conduct Authority (FCA), The FCA's role under the Payment Services Regulations 2009.

¹¹³ See n 52 above.

¹¹⁴ This would not need to occur if the law, for example, requires the rendering of e-money services via a separate entity subject to a special authorization akin to that of banks. MNOs could set up such special subsidiary if the laws on significant shareholdings on financial institutions do not pose any problem. Still, it would arguably increase costs for non-bank institutions.

¹¹⁵ For example, in Italy the patrimony of destination is a device given to companies to create segregated pools of assets. See arts 2447bisff Italian Codice civile. These provisions stipulate the rules for patrimonial separation, but say nothing about the attribution of property/security rights over those assets to specific parties.

¹¹⁶ See n 40.

¹¹⁷ See GSM Association, ‘Why Should Mobile Money Funds Be Ringfenced or Otherwise Safeguarded?’ http://www.gsma.com/mobilefordevelopment/programmes/mobile-money/safeguard-of-customer-money/.