Collective Investment: Land, Crypto and Coin Schemes: Regulatory ‘Property’

Abstract

The UK Supreme Court held in Asset Land Plc v. FCA [2016] UKSC 17 that the land-banking scheme required authorization as a collective investment scheme in order to be sold to retail investors. This article critically analyses that decision in light of various approaches to regulating real estate investment trusts (‘REITs’) in the UK, Singapore, and Hong Kong, respectively. It considers the degree to which deference should be given to the views of the regulators, and the ramifications for other areas like initial coin offerings which regulators are increasingly seeing as securities requiring both prospectus disclosure for the token offering and intermediary regulation for its trading. It argues that the regulation of token offerings is both necessary and desirable. Regulation not only helps protect investors from fraudulent token issuers but also helps to fulfil other worthwhile goals, such as providing additional funding for small to medium-sized enterprises and financial inclusion.

1 Introduction

This article sets out the decision of the UK Supreme Court in Asset Land Plc v. FCA¹ where it was held that the land-banking scheme there required authorization as a collective investment scheme in order to be sold to retail investors. The HK definition is quite similar as to how authorization works in practice which is that, without it, schemes can only be offered to sophisticated investors. The position in Singapore, however, was different in that until definitional changes came into effect in October 2018 to remove the concurrent requirements of pooling and external management, land-banking schemes were kept outside the regulatory regime for collective investment schemes altogether. Amongst other things to be discussed are how much deference should be given to the views of the regulators, and its ramifications for other areas like initial coin offerings which regulators are increasingly seeing as securities requiring both prospectus disclosure for the token offering and intermediary regulation for its trading. It appears, however, that interest in initial coin offering (‘ICO’) has waned slightly and it is suggested here that this was because regulators failed to respond quickly enough to prevent a lemons story arising. The Singapore experience with real estate investment trusts (‘REITs’) was that although they did not fall within the definition of a collective investment scheme in 2002, the definition was amended to bring them into the regulatory sphere and REITs now form 10% of the market capitalization of the Singapore Exchange. We will argue that regulation is necessary not only for investor protection but creates a form of regulatory ‘property’. Tokens can still be seen as representing a technology interest which even if not fully proprietary is a form of intermediate interest lying between contract and property. But this requires disclosure or some standardization of rules in order to decrease information costs associated with them. Regulation as a strategy can also fulfil other worthwhile goals, such as funding small and medium-sized enterprises (‘SMEs’) and financial inclusion, even if at the expense of some investor protection.

2 Land-Banking Schemes

In Singapore, the Monetary Authority of Singapore traditionally held that land-banking schemes were not considered collective investment schemes and consequently fall outside the scope of the Securities and Futures Act (‘SFA’).² For example, in July 2010, the Monetary Authority of Singapore (‘MAS’) stated that:

Land-banking investments involve investors acquiring direct interests in real estate rather than in securities related to real estate and, as such, fall outside the scope of the SFA and [Financial Advisers Act].³

Being outside the collective investment scheme regime, like timeshares and club memberships, these could be sold generally to the investing public without the need to comply with the prospectus requirements of the SFA. Nor did the sales of such investments require any capital markets services licence to deal in capital markets products. This, however, kept these various schemes in the nether world, lacking regulation but also legitimacy with only some consumer protection legislation covering them. These difficulties have largely been removed by an amendment to the definition of collective investment schemes introduced by the Securities and Futures (Amendment) Act 2017 (‘SFAA’). This was passed in January 2017 but only came into force in October 2018. In particular, it amended the definition of ‘collective investment scheme’ to address the land-banking problem so that it is closer to the UK definition in the Financial Services and Markets Act 2000 in that it can have the characteristic of having contributions from and profits of investors pooled (which may not initially be the case with some ‘land-banking’ schemes since the attempt there is to separately allocate notional parcels of undivided land to the investor), or alternatively, the property managed by a manager for the investors (and not both as previously required under the Securities and Futures Act).⁴ Put differently, the collective investment scheme does not have to be as collective as it previously was.

Presently, with some minor differences, a ‘collective investment scheme’ (‘CIS’) in the UK, HK and Singapore in effect means an arrangement in respect of any property under which the participants do not have day to day control over the management of the property, where either the property is managed as a whole by or on behalf of a manager or the contributions of the participants and the profits or income out of which payments are to be made to them are pooled; and which provide the participants with economic benefits. Scheme managers thus attempt to reduce their involvement and at the same time try to confer more day to day management powers to investors in order to maintain the idea that investors acquire direct interests in real estate. It is unlikely that any artificial technical specifications in the scheme in order to avoid regulation will work to allow these schemes to be sold to retail investors without authorization.

In Asset Land Investment Plc v. FCA,⁵ the UK Supreme Court unanimously upheld the decisions of the courts below in finding an archetypal land-banking scheme in the UK (where individual plots are sold to investors despite the land being restricted and on the understanding that a developer would be found to buy up the plots as a whole) as falling within the meaning of a CIS in section 235 of the Financial Services and Markets Act 2000. This meant that Asset Land was operating a regulated activity which required authorization under section 19 of the same Act. Asset Land had represented that they would seek planning permission to rezone the land for residential use during the sales process. In subsequent written documentation, such as the contract of sale, however, there was a representations clause which attempted to remove any prior misrepresentations. A letter would also be sent before the payment of the initial deposit which stated that Asset Land was not responsible for pursuing re-zoning or planning permission (this also appeared as a disclaimer in the contract of sale). At first instance,⁶ Smith J held that the non-reliance clause only covered statements of fact and not of future intention. The Court of Appeal⁷ did not address this point but instead held that the definition of a CIS captured ‘arrangements’ which did not have to be formal nor legally binding, and in particular that there was no need for a mutual understanding. Such an arrangement arose at the point when the oral representations were made, even if that arrangement ended or was varied by the subsequent documentation.

At its core, a CIS is a ‘collective investment where there is pooling of (a) contributions and income/profits and/or (b) collective management’.⁸ More specifically, section 235 states that participants in the scheme must not have day to day control over the management of the property, and that the property must be managed as a whole by or on behalf of the operator of the scheme. Although the definition was challenged with respect to its various constituent parts, the main argument raised by Asset Land in the Supreme Court was that the relevant property here was the individual plots of land, over which the investors had control, and not the sites as a whole. This was rejected by Lord Carnwath who held that an arrangement did not have to be viewed objectively just from the operator’s viewpoint; that the relevant property for consideration was the entire site rather than each individual plot aggregated together as in the case of a block of flats; and that management concerned the operation of the scheme rather than looking after the individual plots as a managing agent would. Earlier in the judgment, he had approved of Smith J’s approach,⁹ that management activity was seen as context-specific to the scheme—here it involved the promise to obtain planning permission, and then to find a developer for the entire site.

Lord Sumption preferred a narrower approach, even though he also agreed with Lord Carnwath (the other three judges agreed with both). He went through the history of collective investment scheme regulation and drew the distinction between unregulated activity involving sales of physical property like land, and regulated schemes involving not just land sales themselves (even with subsequent professional services provided to rezone the land, which would not be a Financial Services and Markets Act 2000 regulated activity) but arrangements which fall within the definition of a collective investment scheme. He found this to be present in Asset Land but only on the basis that the scheme was a collective investment scheme at the point when it was constituted (and not just later on because of the parties’ actions) because although the investors legally owned their own plots, the Andrew Smith J at first instance had found that their dominion was illusory as they could not sell their plots individually as this would have defeated the purpose of the scheme even as planned (value would be lost throughout by all the parties). Implicitly this accepted that a non-reliance clause cannot change the course of history, but only serves as an estoppel (contractual) to prevent an action for misrepresentation later on, and possibly an exclusion clause.¹⁰

As we have seen, Singapore has amended its definition of a CIS to look more like section 235 of the Financial Services and Markets Act 2000 in that pooling and collective management have become alternative requirements as in the UK. This would be consistent with Lord Sumption’s analysis of the CIS regime in the UK, where he thought that pooling and having someone else manage the property covered the same mischief where an operator manages other people’s money—they were ‘functionally equivalent’.¹¹ This is important as it means that most funds cannot escape CIS regulation, which may be a better way to police the industry than to regulate the underlying products that a CIS invests in.

MAS has also said, however, that collective investment schemes that require authorization will be restricted to investments in securities or other assets that are liquid, such as precious metals, or which have stable income, such as completed real estate.¹² That will continue to distinguish regulated CISs from unregulated schemes involving timeshares and club memberships which are only subject to consumer protection rules. But given the slightly more liberal philosophical bent with regulators in Singapore, it may be that the position in HK is stricter even with a similar CIS wording. The Securities and Futures Commission of Hong Kong (‘SFC’) has said that all property funds are largely included and it is really only the sale of individual units without more that is outside the CIS definition. In 2013, the attempted Apex Horizon hotel room sale by Cheung Kong (Holdings) had to be withdrawn (with 360 buyers having their deposits refunded) as the SFC saw that as an unauthorized CIS given that the operator continued to allocate guests to the rooms.¹³ Day to day management was still in the operator’s hands, and not the individual buyers. It may, however, be that the views of specific regulators should generally prevail regardless of how similar the definitions are due to the administrative deference accorded to them. It has been argued elsewhere that there is no need for courts to be cautious in reading modern financial statutes even if there are criminal consequences attending their breaches.¹⁴ This is especially so if regulators want to restrict schemes structured to defraud, but just as true if regulators in fact want to legitimize schemes by authorizing them when they may strictly speaking not fall within the definition of a collective investment scheme. The latter was experienced in Singapore with REITs.

3 Collective Investment Schemes and Singapore REITs

A few things need to be said about the Singapore definition at the outset. First, the CIS may need to provide economic benefits measured objectively in financial or monetary terms only. Second, closed-end funds are prima facie excluded from the definition of a collective investment scheme, which started out just covering unit trusts or mutual funds investing in underlying securities or futures contracts. Third, if it is a CIS and unauthorized, it cannot be offered to retail investors, although it can be a restricted scheme that can only be offered to accredited investors. These are subject to minimal regulation by the MAS.

The collective investment scheme definition that was first introduced in 2002 with the Securities and Futures Act is much tighter than the previous definition of ‘interests’ in the Companies Act, which used alternative criteria that involved more open-ended terms like ‘common enterprise’ and ‘investment contract’. Although these terms were not fully tested in the local courts,¹⁵ the width of it can be seen in the US, where in the leading case of SEC v. Howey¹⁶ it was held that an arrangement created by a land sale contract, warranty deed and service contract for a citrus grove development with contracts for cultivation, marketing and remitting of net proceeds was considered a security.¹⁷ Registration of the offered securities was required and the offer was prohibited as there was no registration statement. It is the element of it being an investment contract whereby a person invests money in a common enterprise and expects to make a profit, predominantly through the efforts of a promoter or third party,¹⁸ which renders offers of such securities registrable, even if the underlying asset that is managed is not a security itself, and the units issued are non-redeemable. That the authorities in Singapore wanted to disassociate themselves from the possible development of the meaning of securities via investment contracts and still intend this was confirmed recently by the November 2018 update to the MAS Guide to Digital Token Offerings which said that ‘(t)he treatment of a token under the Howey Test is not a consideration for deciding whether a token is a product regulated under the SFA’.¹⁹

The definition of a collective investment scheme, which is somewhat similar to the definition in Schedule 1 of Hong Kong’s Securities and Futures Ordinance, as well as the meaning of ‘collective investment schemes’ under the UK’s Financial Services and Markets Act 2000, has concurrent requirements that require delegation to a manager or the pooling of monetary contributions and profits (thus effectively excluding timeshares and club memberships),²⁰ and the sharing of what appears to be profits in pecuniary form (‘the profits or income from which payments are to be made to them are pooled’ and ‘profits, income, or other payments or returns’) rather than ‘profits, rent or interest’ as was the case in the definition of ‘interests’ under the now repealed section 107 of the Companies Act). The long list of exclusions also relate to schemes that generate such financial returns. It does appear from the beginning that the authorities wanted to really only regulate unit trusts as CISs, which is why there was a problem when the first REITs actually desired classification as a CIS in 2002 (as opposed to being completely unregulated even when sold to retail investors) as these fell within the express exclusion for ‘a closed-end fund constituted either as an entity or trust’.

That regulatory mindset linking CIS to open-end unit trusts has created a great deal of difficulty but could, however, have been maintained in other ways. In the UK, the Financial Services and Markets Act 2000 essentially forbids the public offering of unit or trading trusts that are not open-ended in nature, and these can only be marketed privately and to sophisticated investors.²¹ Closed-end companies are treated as shares, and closed-end trusts cannot be authorized or offered to the public.²² But redemption there was given a wider meaning—and it is stated under section 243(11) of the Financial Services and Markets Act 2000 in relation to the redemption formula in section 243(10):

But a scheme shall be treated as complying with subsection (10) if it requires the manager to ensure that a participant is able to sell his units on an investment exchange at a price not significantly different from that mentioned in that subsection.

The position in the UK, which the Securities and Futures Act collective investment scheme regime modelled itself on, was therefore both stricter and more flexible at the same time. A closed-end trust, though still considered a collective investment scheme as defined in the UK, would not be authorized as such, the consequence of which is that it cannot be offered to the public but only to a set of sophisticated investors. However, the meaning of redeemability is wider than is usually understood, and permits that requirement to be met by a listing on the exchange. This could then allow the UK Financial Conduct Authority to treat the trust or company as open-ended,²³ which would then permit its authorization and offer to the public.

In Singapore, however, closed-end trusts or companies were kept out of the definition of collective investment schemes altogether. While the latter is regulated when it offers shares to the public, the former was outside the regulatory reach of the Securities and Futures Act entirely, and could be offered to the public completely unregulated (and not to just accredited investors had it been an unauthorized CIS). This created some difficulties, particularly with the introduction of REITs into the capital markets. In Singapore, these are structured as trusts (mainly for tax reasons), and due to the illiquid nature of the underlying assets, the units that are issued or sold to investors are in effect not redeemable (the trust deeds generally state that they cannot be redeemed while the REIT is listed on an exchange). Given the exclusion of closed-end funds from the definition of a collective investment scheme, they should have fallen outside the regulatory regime, and would not have required MAS’s approval as a collective investment scheme before they could be sold to the public. However, most issuers of REITs, and the investment bankers associated with them, in fact desired the regulatory safeguards as that increased investor confidence in the units being sold or marketed to the public. This is something that is true in perhaps more developed financial jurisdictions where institutional investors can only invest in regulated products (even where they are only lightly regulated).²⁴ So with REITs, they were registered as CISs even though redemption was suspended and so could not legally have been considered open-end funds. Crucially, this was largely due to the efforts of the REITs (more accurately the persons who set them up).

What happened was that the definition of a ‘closed-end fund’, itself excepted from the CIS definition, was seen in practice to exclude these REITs, and this was then progressively amended since the mid-2000s so that it now reads:

‘closed-end fund’ means an arrangement referred to in paragraph (a) or (b) of the definition of ‘collective investment scheme’ under which units that are issued are exclusively or primarily non-redeemable at the election of the holders of units, but does not include—

(a) an arrangement referred to in paragraph (a) of that definition—

• (i) which is a trust;

• (ii) which invests primarily in real estate and real estate-related assets specified by the Authority in the Code on Collective Investment Schemes; and

• (iii) all or any units of which are listed for quotation on an approved exchange;

But there is a new subsection (aa) inserted by the SFAA 2017 that effectively states that all funds operating for economic gain are not to be seen as closed-ended and hence will be collective investment schemes. This was actually introduced in a 2013 amendment to the CIS definition by the Securities and Futures (Closed-End Fund) (Excluded Arrangements) Notification 2013. This means that the CIS exclusion for a ‘closed-end fund’ is itself subject to further exclusion if the fund is not carrying on an active business (i.e., holding passive investments). Specifically, MAS has prescribed the following criteria to determine whether a closed-end fund is deemed to be a CIS (which operated prospectively from July 1, 2013):

• all or most of the units issued under the arrangement cannot be redeemed upon the election of the holders of the units;

• the entity operates in accordance with an investment policy under which investments are made for the purpose of giving participants in the arrangement the benefit of the results of the investments, and not for the purpose of operating a business; and

• the arrangement has certain prescribed characteristics.

What now seems to have happened is that the CIS exclusion for closed-end funds highlighted above does not really exist anymore except for very few investment funds. This actually brings us to the position in HK where the SFC has made clear that it does not matter whether a CIS is open or closed-ended.²⁵ Still, the influence in Singapore of that previous philosophy may explain why even security tokens, which are usually non-redeemable, may not be seen as a CIS in Singapore.

The creeping nature of regulation shows the careful approach of Singapore regulators towards market development. There is a fear of overregulation until they see clear opportunistic behaviour. This has been the case with land-banking schemes. However, with REITs the story is inverted in that while they were not subject to regulation at the start, they desired it and MAS was then willing to provide it immediately to help grow a nascent industry. There is thus also a strategy of regulating early in order to create regulatory ‘property’ in the sense of legitimating a deliberate practice through signaling.²⁶ This also indirectly helped overcome the fact that REITs in Singapore are probably non-charitable purpose trusts that traditional English law would have seen as invalid. Courts have slowly recognized the REIT as a separate legal entity as is the case with US business trusts, and this was partly due to regulation and the application of those regulations in, for example, the restructuring context where parallels are drawn with corporate schemes of arrangement.²⁷ REITs now form 10% of the market capitalization of the Singapore stock exchange. The counter-example is of holiday timeshares which still remain in a shadowy state outside the definition of a collective investment scheme altogether as MAS sees it. In contrast, it has been argued that US holiday timeshares have acquired some form of property status through a mix of state securities regulation, urban law and consumer legislation,²⁸ unlike, say, timeshares in watches.²⁹ However, should so much discretion be given to regulators and what would happen if they are challenged in court were they to attempt to provide property or quasi-property status through statutory intervention to a form of technology interest?³⁰

4 Judicial Deference to Financial Regulators

The Financial Conduct Authority (‘FCA’), and its predecessor the Financial Services Authority, has by all accounts been very interventionist since the Global Financial Crisis,³¹ and it was said at first instance in Asset Land that ‘its bona fides cannot be questioned’.³² However, for this to work, courts may have to give regulators (at least the more qualified ones) some leeway, and there are growing signs of this. In Financial Conduct Authority v. Capital Alternatives Ltd,³³ which the English Court of Appeal decided after Asset Land (but permission to appeal was refused by the Supreme Court), Christopher Clarke LJ thought that ‘Courts are reluctant to disturb a settled interpretation and the practice based on it, and there was a powerful presumption that the meaning that had been given to a phrase in issue by a regulatory body was the correct one’.³⁴ This added a further gloss to the speech of Lord Phillips in Bloomsbury International Ltd v. Sea Fish Authority³⁵ that Clarke LJ relied on where his Lordship only stated that there is ‘a powerful presumption that the meaning that has customarily been given to the phrase in issue is the correct one’.³⁶ After a detailed examination of various guidance letters of the FCA that were referred to by defence counsel, however, Clarke LJ thought that they concerned different structures from the one under consideration there.³⁷ In Asset Land,³⁸ Gloster LJ may have gone even further in finding helpful a paper published by the Financial Markets Law Committee dated July 2008, ‘Issue 86—Operating a Collective Investment Scheme’, which had said that the definition of ‘arrangements’ is very wide and intended to be so, and which in that case guided her in finding such an arrangement that resulted in the establishment of a collective investment scheme. The Supreme Court while referring to this paper did not, however, comment on it and it may be that the approach in the UK is still more conservative as opposed to other developed jurisdictions, including other European countries.³⁹ This may explain why the UK was one of the first countries in the world to adopt the financial regulatory sandbox, which in a sense bypasses Parliament. Singapore and Hong Kong followed not long after in their attempt at encouraging financial innovation without being hamstrung by extant legislation in ways in which perhaps some other jurisdictions are not.

In Singapore, courts may also be slightly jealous of their custodial role of interpreting statutes, which is the traditional approach. This may explain why section 321 of the SFA had to be expressly introduced in 2002 to grant MAS the power to issue practice notes and no-action letters. It is, however, also acknowledged such no-action letters do not bind the public prosecutor should it choose to bring a criminal prosecution, and the fact that the provision was thought necessary shows that the MAS could not otherwise issue no-action letters, which are frequently used in the US and Australia. Similarly, in HK, it was said in a securities licensing appeal from the Securities and Futures Appeals Tribunal in Ng Chiu Mui v. Securities and Futures Commission⁴⁰ that:

It was submitted that the SFC’s interpretation was the correct one and the Tribunal was wrong to have dismissed that interpretation as well as the footnoted commentary in the consultation paper presented to the Bills Committee as ‘straws in the interpretative wind’.

But the SFC’s view can be of no relevance as a matter of law unless it is a tool of statutory interpretation. Since Mr Grossman accepts that it is not such a tool, the Tribunal’s approach plainly was correct.

By contrast, in the context of judicial review, courts in the US,⁴¹ Canada,⁴² and to a lesser extent in Australia,⁴³ accord regulators a zone of discretion in the exercise of their administrative powers. The role of the court is to ensure that they stay within that zone, but not to attempt to second-guess their intentions.⁴⁴ It may be that this is the only way that financial markets can properly be regulated given the pace at which things change and the fact that the various participants in the market are actively trying to avoid regulation. There may be a balancing exercise for which courts are ill-suited where the application of open-ended standards may require a cost–benefit analysis.⁴⁵ While the tide may have turned when it comes to seeing whether an investment arrangement involving underlying land or real property is a collective investment scheme, this may, however, not be because courts are prepared to defer to the regulatory expertise of the FCA/SFC/MAS. Rather, there has now evolved enough formal material to aid the courts in statutory interpretation when a land-banking matter comes up against highly technical objections raised by those with less of a stake in its proper regulation.

It is not clear if that is the case with cryptocurrency funds and initial coin offerings due to their more recent nature. Further, regulators themselves are ambivalent about how and how far to regulate what are in effect virtual property schemes given their possibly greater contribution to the economy.⁴⁶ At the moment, it is likely that some are issuing no-action letters in order to help guide Fintech participants, while others, like the US, are more stringent with ICOs. Without formal regulation, however, these tokens will find it hard to obtain general acceptance as an asset class. Courts have provided a threshold test before recognising something as property, requiring it to be ‘definable, identifiable by third parties, capable in its nature of assumption by third parties, and have some degree of permanence or stability’.⁴⁷ Unlike tangible property, however, it has been said that intangible property ‘only exists because the law says it does […] and, in principle, it is open to the parties to establish the extent of the debtor’s rights under the contract in any way they wish’.⁴⁸ In addition, as Merrill and Smith have pointed out, a crucial distinction between in personam contract rights and in rem property rights is that contract law typically permits contractual parties to customize their rights and duties while property law requires that parties ‘adopt one of a limited number of standard forms that define the legal dimensions of their relationship’.⁴⁹ Many types of legal relationships lie somewhere in between this contract/property interface.⁵⁰ A possible example of such intermediate relationships is the one between cryptocurrency/token issuers and investors where the relationship is mediated by financial regulators.⁵¹ The key question remains whether regulators should step into regulate what might otherwise be a purely contractual relationship between issuers and investors. At the moment, the ICO story appears to parallel more what has happened with timeshares in Singapore as opposed to REITs. While it was correct to take a wait and see approach to token regulation, the time is perhaps right to regulate them, possibly as collective investment schemes, in order to fulfil goals of helping SMEs and financial inclusion that did not fully succeed with the earlier crowdfunding story.⁵² But token issuers, as was the case with Singapore REITs, must take the lead and work towards being regulated.

5 Cryptos and ICOs

It is unlikely that cryptocurrencies like Bitcoin and Ethereum, recently accepted by a Singapore court as proprietary interests,⁵³ would be seen as a capital markets product as opposed to a currency or payment system that will be regulated by the new Payment Services Act 2019 in Singapore (passed in January 2019 but not yet in force).⁵⁴ It would be quite different with the Bitcoin futures contract now traded on the Chicago Board Options Exchange which if it had a presence in Singapore would have been seen as a ‘futures contract’ under the Securities and Futures Act and with the coming into effect of the Securities and Futures (Amendment) Act 2017, a ‘derivatives contract’.⁵⁵ Many products can, however, overlap. Indeed many ICOs are today offered in exchange for cryptocurrency, so it is the virtual being built on the virtual. ICO issuers usually are or concern putative technology companies with no immediate working product. They are in this sense very much like a fund that says that it has not identified what companies to invest in which in Exeter Group Limited v. ASC⁵⁶ was found to be insufficient disclosure for there to be a public offering in Australia, a country which Singapore modelled its reasonable investor prospectus disclosure standard from 1999 on. It was not enough for there to be full disclosure regarding the absence of any detailed plans on the part of the management of an investment fund which sought to raise funds from the public (in AUD2000 tranches) as to the types of companies it would invest in. There was nothing misleading in, or omitted from, the prospectus. Despite this, the Australian Securities Commission refused registration on the basis that a higher, not lower, standard of disclosure applied where a prospectus was targeted at small or retail investors.⁵⁷

Consequently, most ICOs avoid or try to avoid prospectus disclosure in Singapore as that is either too costly or they fear they cannot comply with it. The basis for using unregulated coins or tokens was a statement made by the MAS about 2 years ago and Singapore has since become the third largest ICO jurisdiction in the world.⁵⁸ Out of the worldwide 211 ICOs in 2017, 20% of that was in Singapore raising about US$790 million according to the Association of Cryptocurrency Enterprises and Startups, Singapore.

In November 2017, however, MAS issued a Guide to Digital Token Offerings.⁵⁹ This stated that ‘digital tokens that constitute capital markets products’ will have to comply with the offering requirements of the Securities and Futures Act, including the need to prepare a prospectus, although the offerors can avail themselves of the exclusions and exemptions there, the most important of which are offers to accredited investors and the $5 million small offer exception.⁶⁰ This statement concerned offers of digital tokens in the primary market that represent underlying securities, for which a great deal of concern has been voiced recently in terms of their financial risks,⁶¹ as well as their link to illegal activity.⁶²

The Swiss Financial Market Supervisory Authority (FINMA) suggested that there are essentially three types of such tokens, payment tokens, utility tokens and asset tokens,⁶³ with only the last being fully regulated. However, it also suggested that these categories are not mutually exclusive and can be in the form of a hybrid. If it has some characteristics of a capital markets product, that should be enough to require its regulation. Tokens could be caught by the CIS definition where the effect or purpose of the scheme is to obtain economic benefits given that the need for the collective nature of the scheme has been reduced by removing the pooling requirement. MAS has acknowledged this in its update to the Guide to Digital Offerings in November 2018 but has also said that the SEC v. Howey test discussed earlier does not apply in Singapore (this was not in the initial Guide). In contrast, the US Securities and Exchange Commission (SEC) has in fact been enforcing extant securities laws in that regard. Howey was used by the SEC in Munchee’s restaurant review application MUN token to argue that it was a security due to the expectations of a return even though it was argued that it was a utility token to be used within Munchee’s system and not to fund it.⁶⁴ Perhaps this was because an ‘investment contract’⁶⁵ can be interpreted more widely than a CIS, and includes how ‘developed and operational’ and ‘immediately available to use’ the system is when the tokens are sold, and so most tokens in the US are seen as securities when this may not be the case in Singapore. We have, however, seen that the use of the term ‘arrangement’ in the CIS definition can be equally wide in the land-banking context. In HK, the SFC has used the CIS regime to stop the token offering of Black Cell Technology Limited, although it appears that not only were the ICO proceeds to be used to fund the development of a mobile application, but the token itself could be subsequently redeemed for equity in Black Cell.⁶⁶ That later characteristic made the securities link unavoidable in its primary offering. Singapore has accepted this position too which seems to be the extent of their enforcement action thus far.⁶⁷ The further use of the CIS regime to regulate ICOs has not been evident, partly, it was suggested above, because it was in the past linked to redeemable interests. It could, however, also be because the Guide to Digital Offerings also stated that one of the tests for whether an ICO has to be regulated is whether they have been targeted at Singapore investors, and most have been offered to overseas investors only.⁶⁸ Further, as we have seen, in Singapore, the test of economic benefits appears to require the scheme to have the effect or purpose of providing investors with financial or monetary returns and this would exclude most utility tokens. The focus is also not on the investors’ expectations but on what the scheme has objectively promised.

At the same time, however, crypto funds may not be covered if the underlying asset is not securities or derivatives products as the MAS has made it clear that a CIS has to concern such capital markets products or, more recently, real estate. But, as with equity crowdfunding, where the issuer is usually exempted from prospectus requirements, perhaps the focus should be on fund regulation as the problems are usually with platform intermediaries set up to channel investor money to the issuers. The SEC was one of the first regulators to notice this and required special purpose vehicles or platforms set up for equity crowdfunding to be regulated as an investment company.⁶⁹ It is likely that some of these platforms have moved over to become crypto funds investing in non-capital markets tokens where they seek to remain outside the regulatory regime altogether. We have seen, however, that there are funds that desire regulation as they cannot operate without it in Europe, where the focus is also more on their regulation as alternative investment funds.⁷⁰ In November 2018, the SFC also said that investment funds based in Hong Kong that invest more than 10 percent of their gross portfolios in ‘virtual assets’, that are not securities or futures contracts, either directly or indirectly via intermediaries, will have to be licensed and registered to deal in securities and to provide fund management.⁷¹ Only professional investors can invest in these crypto funds unless the portfolio is a collective investment scheme authorized by the SFC.

The expansive approach taken recently in the land-banking cases and previously with REITs to the definition of a collective investment scheme should mean that most crypto funds, however, can be seen as CISs. The approach in the UK and Hong Kong is also that if the fund is not authorized you can still promote them to sophisticated investors. Arguably a similar approach should be taken in Singapore so that the crypto fund, while not invested in underlying securities or derivatives contracts (or real estate), could be a collective investment scheme. While we suggest a lighter regulatory framework for token issuers, we do not suggest the same for crypto funds.

6 Regulatory Caution and Strategy

This is about regulators using strategy rather than the regulated gaming the system: see Vejanovski (2010), pp 99–100.

While it may appear that MAS seemingly only reacts to overreaching behaviour in the financial markets after they occur, this may in fact be the correct way for regulators to act. The market should be allowed to function as much as possible, and only where it has failed, and subsequently failed to adjust to find its own solutions and remedies, should regulators then step into correct the problem. This is not an excuse for the libertarian refrain that ‘this time is different’⁷³; far from it. Regulators have to act in a temperate way, always looking closely at empirical evidence, and challenging their own views. Although said in an infinitely different context, they may well have to constantly remind themselves of ‘the spirit which is not too sure that it is right’.⁷⁴ The statement by Ravi Menon, MD of MAS, on 15 March 2018 is in this vein. He said that ‘MAS assesses that the nature and scale of crypto token activities in Singapore do not currently pose a significant risk to financial stability. But this situation could change, and so we are closely watching this space’.⁷⁵ William Coen, Secretary General of the Basel Committee on Banking Supervision, in a speech sounding very much like what Alan Greenspan used to say also thought that ‘[o]ur job as regulators and supervisors is not to prevent future crises, but to reduce their likelihood, and perhaps more importantly, their impact on the real economy’.⁷⁶

From an academic perspective, it appears that regulators are cautious about coins and crypto funds but ‘politicians like looser regulation’.⁷⁷ One reason for this may be that they fear missing out on the next big growth story, particularly given the slowdown in economic growth alongside recovery in bank bonuses since the Global Financial Crisis. That there is leftover SME guilt can also be seen in the bank capital exceptions for loans to small and medium-sized enterprises.⁷⁸ The Organisation for Economic Co-operation and Development (OECD) has reported on the potential of ICOs for SME financing although it believes these should be ‘appropriately regulated and supervised’.⁷⁹ The young are also struggling for jobs all over the world as retirement ages have been pushed back, and SMEs, as we shall see, employ disproportionately more people in relation to their position in the economic landscape. Another explanation is that there has been a fundamental shift in long-term business cycles so that the world’s economy has been unable to start afresh partly due to the failure of states to address the over-financialization of their economies.⁸⁰ The problem is that these investments in technology, or more accurately in the hope that technology will lead to tangible growth, may not work.⁸¹ Even if they do not pose financial stability risks, the investors, many of whom may well be from that younger generation, can see that the world does not need more than 1500 cryptocurrencies when there are fewer than 200 real currencies. Still, the jury is out on whether they should be regulated as securities, which is unlike the case with land-banking, where there is almost universal recognition now that the starting point is that they are collective investment schemes. Only countries like Malaysia⁸² have clearly come out to say that all tokens are securities from January 2019 although the practice in the US is also to subject even many utility tokens to securities regulation.

We have so far been too focused on the fraud story. Ideally, securities laws should be able to distinguish industrial-based entities requiring fundraising from financial ones. Where the former may have sound economic reasons for a hands-off or purely disclosure-based approach, or perhaps even less, the latter requires more merit regulation as it creates less real economic benefit given the amount of externalised systemic risk created. In addition, a company running a business has to be in existence for a while before it can raise funds from the public, largely because the latter can often be reached only through a listing on a securities exchange, which would usually have requirements for operating and profit track records. Financial instruments can, however, be sold without much prior structure in place, and the banks and financial institutions can easily reach the public directly and off exchanges. Put differently, regulation could operate on a risk-adjusted basis. The problem with not doing so with tokens and crypto funds is that bad money may well have chased out the good.

MAS’s approach to the CIS and requiring it to be linked to investments in securities, derivatives contracts and real estate before regulating them is perhaps a belated attempt at trying to prevent there being too many of such funds. Other tokens and crypto funds which have a purpose other than as a wealth management device were thus supposed to be helped or encouraged by their lack of prospectus, intermediary and platform regulation. But one important ‘other relevant consideration’ should have been that this purpose be reflected in some kind of existing business or functionality.⁸³

7 SME Financing

It has been argued that investor protection can be somewhat sacrificed for capital formation.⁸⁴ But we cannot be neutral about where we want capital to gravitate to as more recently it has been said that there are good corporate purposes that transcend short-sighted profit maximization.⁸⁵ But for that to happen, there has to be some merit decision taken by the regulator as to how and where it wants capital to be formed. There are SMEs actually making things out there and employing people that require capital. The number of SMEs in Singapore has increased steadily from 242,900 in 2014 to 262,600 in 2018, consistently accounting for 99% of all enterprises in Singapore.⁸⁶ During this period, SMEs continue to employ about 72% of all workers in Singapore and contribute almost half of the nominal added value of all enterprises each year.⁸⁷ Nevertheless, many SMEs appear to experience difficulty in obtaining financing for their business. A study conducted by Visa and Deloitte in 2015 found that 40% of the SMEs were unserved by financial institutions and that another 8% were underserved.⁸⁸ Moreover, 72% of the SMEs ‘require[d] funds to better manage their working capital and mitigate cash flow problems’.⁸⁹ Another SME survey conducted by DP Information in 2017 concluded that financing posed the biggest challenge for SMEs: about 35% of the SMEs surveyed faced finance-related issues, up from 22% in 2016 and 14% in 2015.⁹⁰ The five main finance-related issues reportedly suffered by those SMEs were (1) delayed payment from customers (81% in 2017; 14% in 2016), (2) high interest rates for bank loans (29% in 2017; 46% in 2016), (3) suppliers tightening credit access (22% in 2017; 34% in 2016), (4) the need for more collateral for financing (9% in 2017; 19% in 2016), and (5) inability to renew financing (5% in 2017; 7% in 2016).⁹¹ While another survey conducted by the then SPRING Singapore (now known as Enterprise Singapore) in 2017 painted a rosier picture, it did acknowledge that micro companies (i.e., companies with revenue below S$1 million) faced lower approval rates when they sought debt financing.⁹²

An important reason why SMEs, especially micro companies, have difficulty obtaining external financing is that they are (and rightly so) perceived as high-risk businesses. According to a study by NUS Enterprise, only about one fifth of young start-ups surveyed were self-sustaining and more than half of them were not cash-positive.⁹³ An estimated seven to eight out of ten new businesses formed in Singapore would cease operations within the year of their conception.⁹⁴ As a result, SMEs, especially start-ups, are not ideal clients for banks and other financial institutions, which are inherently risk-averse. Indeed, one of the reasons why micro companies are less successful at obtaining debt financing is due to their ‘weaker business performance’.⁹⁵ Moreover, debt financing often comes with an obligation to make periodic repayments, which start-ups are likely to struggle with while trying to work out their business model. Even more established SMEs are likely to experience cash flow problems from time to time as a result of delayed payments from customers.⁹⁶ Indeed, concern over a company’s ability to make loan repayments represents a significant reason why financial institutions are reluctant to lend to micro companies.⁹⁷ Additionally, start-ups often do not own many assets (such as land or work products) that can serve as collateral to obtain bank loans.⁹⁸ Without collateral, a company can be expected to pay interest rates as high as 10% per month, which can greatly exacerbate its cash flow problems.⁹⁹ In light of the foregoing, it is not surprising that one expert concludes that start-ups with less than 3 years of operations are simply ‘not within the banks’ risk appetite’.¹⁰⁰

In addition to banks and other financial institutions, SMEs may also obtain funding from a number of government programmes, ranging from start-up grants, co-investment programmes, and government-assisted loans. Nevertheless, as one of the authors has argued elsewhere, these programmes often pursue distinct policy objectives or focus on specific industries (e.g., the high-tech industry) and, as a result, come with various restrictions and eligibility requirements.¹⁰¹ Such programmes therefore cannot and are not intended to serve the needs of all worthwhile start-ups. Similarly, venture capital and private equity firms tend to focus on companies with high-growth potential and typically only invest within their areas of expertise.¹⁰² As some industry watchers have observed, bank loans remain ‘the most popular form of external financing across SMEs of different sizes, industries and stages of development in Singapore’.¹⁰³ And yet it remains elusive.

Alternative financing solutions such as crowdfunding provide additional sources of funding for SMEs. Broadly speaking, there are four types of crowdfunding: donation-based, reward-based, debt and equity crowdfunding.¹⁰⁴ Crowdfunding is a relatively nascent industry in Singapore, the success of which remains to be seen. However, a number of features of the current crowdfunding industry are likely to limit its usefulness to start-ups. Firstly, both debt and equity crowdfunding are regulated by the MAS, which relies to a large extent on crowdfunding platforms to safeguard investor interests. These platforms are required to be licensed and are expected to perform various duties, such as conducting due diligence on fundraisers, instituting policies to handle fundraisers’ defaults, and so on.¹⁰⁵ These platforms in turn pass a considerable amount of these regulatory costs to investors and/or fundraisers. For instance, Funding Societies, one of the leading debt crowdfunding platforms in Singapore, charges a ‘nominal’ service fee of 18% of interest earned to cover its operations costs.¹⁰⁶ As a result, start-ups may not find it considerably easier or cheaper to obtain funding through crowdfunding platforms. For example, the interest rates at Funding Societies can vary from 9% per annum for secured loans to 16% for unsecured ones.¹⁰⁷ Fundnel, another Singapore-based crowdfunding platform registered with the MAS, is reportedly ‘very stringent’ when screening potential fundraisers.¹⁰⁸ Secondly, while equity crowdfunding may be more attractive than debt crowdfunding to start-ups without a proven track record or a steady cash flow, it is problematic from an investor perspective since it provides investors with very limited exit opportunity. Shares in private companies are generally highly illiquid; the chances of a start-up being bought out by a VC/PE firm or being taken public are quite low. In the absence of a secondary market, investors are likely to be stuck with the same start-up for years without any meaningful ways to liquidate their holdings. Thirdly, while reward-based crowdfunding does not involve an offer of securities and is therefore not regulated by the MAS, it comes with its own limitations. It works better as a one-off campaign to attract early adopters to novel products or services, rather than a long-term way to fund the operations costs of a sustained business. Additionally, reward-based crowdfunding campaigns may have difficulty attracting investors who are unable to enjoy the products/services offered due to geographical or other constraints.

Token offering is sometimes considered a form of crowdfunding since it also allows issuers to raise a small amount of money from a large number of people. The main difference is that instead of directly receiving goods/services (in the case of reward-based crowdfunding) or shares in a company (in the case of equity crowdfunding), investors receive tokens, which can represent a wide variety of rights, such as a right to (1) goods/services provided by a start-up (sometimes referred to as ‘utility’ tokens), (2) a percentage of a start-up’s profit (sometimes referred to as ‘asset’ or ‘security’ tokens), (3) charitable services, or a combination of the above.¹⁰⁹

One may argue that token offering has potential to solve some of the problems faced by the current crowdfunding industry and to provide additional sources of funding for SMEs. To begin with, similar to reward-based equity crowdfunding, utility tokens do not amount to capital markets products¹¹⁰ and therefore should not be subject to securities regulation.¹¹¹ More importantly, both utility and security tokens are potentially tradable. Indeed, token exchanges (such as TokenMarket and tZero) have emerged to provide a secondary market for both types of tokens.¹¹² The improved liquidity is likely to render utility tokens more attractive than reward-based crowdfunding to investors who are interested in a start-up’s business but find themselves unable to personally enjoy the company’s goods or services. For the same reason, security tokens may attract a greater number of investors than equity crowdfunding where they provide an exit strategy for investors who would otherwise be stuck with highly illiquid shares in a start-up.

8 Token Regulation and Information Costs

Nevertheless, some form of regulation is both necessary and desirable to enable SMEs to truly benefit from token offerings. Such regulation serves at least three main purposes. Firstly, it helps protect investors from fraudulent token issuers. There is invariably significant information asymmetry between token issuers and investors.¹¹³ A number of ICOs have turned out to be Ponzi schemes in disguise.¹¹⁴ The MAS is clearly aware of and has warned investors of various risks associated with token offerings, including ‘highly speculative valuation, heightened risk of fraud and lack of a proven track record’.¹¹⁵

Secondly, regulation, if properly designed, has potential to help worthwhile businesses to distinguish themselves from the crowd. At the moment, even if a worthy start-up seeks to issue tokens to finance its business, it is likely to be crowded out by too many token issuers with nothing more than a white paper for their putative technology or business plans. While these were not immediately fraudulent, the reality was that the funds raised went towards paying salaries and bonuses to persons who sold hope through ‘puffery’¹¹⁶ until the money dried up. Any existing SME or even putative one with real business plans simply could not show that they were not a lemon. We now see a waning interest in these Fintech instruments around the world from the middle of 2018. Some of this may reflect weak issuers being kept out by increased regulation.¹¹⁷ But it is very likely that stronger ones too are coming into a more sceptical market given past experience. The REIT story in Singapore stands in stark contrast to this where we have seen that closed-end schemes were initially not seen as CISs and could be sold without authorization but needed regulatory sanction for them to in fact work. Something similar should be done with tokens. REITs lobbied the state to provide them with the regulation that they desired, and have reaped what they sowed.

Needless to say, regulation does not and cannot guarantee the success of any token issuer. But imposing some disclosure and reporting requirements can help identify businesses that have genuine business plans, which are likely to contribute to the real economy. It will also provide some confidence to investors even if there is no assurance or even promise of financial gain. As with the case of crowdfunding, there is invariably a balance to be struck: too much regulation can render token offering too costly to benefit SMEs which are often cash-strapped; too little regulation may not achieve the desired effect of identifying worthwhile businesses.

Thirdly, regulation is likely to benefit the token industry by establishing a set of standardized rules that apply to tokens, thereby reducing the information costs that investors and other interested parties have to incur when dealing with tokens. As Merrill and Smith have posited, when a right impinges upon a large and open-ended class of third persons, ‘legal rules must be designed so as to minimize the information-cost burden imposed on a great many persons beyond those who are responsible for setting up the right’.¹¹⁸ This in turn suggests that ‘standardized rights will be strongly encouraged’.¹¹⁹ The rights represented by tokens are likely to affect a large and indefinite number of persons. Since a main benefit of tokens is that they are tradable, numerous people have a potential interest in understanding what precise rights a token represents. These people include: (1) any person who buys and sells a token¹²⁰; (2) any person who provides funding to a token issuer (since he has to know whether he ranks higher or lower than those who hold tokens); (3) any person who deals with token holders (e.g., he might want to know whether tokens are assets over which he can take security¹²¹); or (4) any person who should know that it cannot take, or interfere with the relationships behind, the token.¹²²

Given that a large and indefinite number of people are potentially affected by tokens, some standardization is desirable with respect to the rights that tokens represent. While those people still have to process information about the rights and corresponding duties associated with each token on offer, the information cost that each person must incur can be conserved in a number of ways. Examples of such cost conserving methods suggested by Merrill and Smith include (a) ensuring that the duties relating to a resource in question apply irrespective of the identity of the owner, (b) ‘making the duties uniform’, (c) ‘restricting the duties to a short list of negative obligations, easily defined and understood by all’, and so on.¹²³ It is beyond the scope of this article to provide a comprehensive analysis of how the rights and obligations associated with tokens can or should be standardized, particularly as there is no clear definition of what a token is. But one may argue that information costs can be reduced significantly if we were to have standardized rules regarding (a) a limited combination of rights and duties that can attach to different types of tokens¹²⁴; and (b) the priority rules regarding tokens and other persons having an interest in the assets of a token issuer or token holder.

Finally, as Merrill and Smith have acknowledged, standardized rules can be supplied by either market players or the government.¹²⁵ There have been recent arguments that the industry can itself create codes or governance rules as a form of self-regulation.¹²⁶ Nevertheless, the government may have an advantage in ‘setting up focal points in order for parties desiring conventions to attain them more quickly and cheaply’.¹²⁷ This is particularly the case where the number of parties affected by a right is both large and unspecified, which makes coordination among those parties more difficult. As such, financial regulators such as the MAS might be better positioned to supply such standardized rules. In any case, self-regulation was said to have been an ‘abject failure’ in the Global Financial Crisis.¹²⁸

One possible way to regulate securities tokens is to treat token offerings, token funds, and possibly token exchanges¹²⁹ as collective investment schemes. Consequently, authorization by the MAS is required before such tokens or units in token funds can be subscribed or transferred by the public. Without formal authorization, only sophisticated investors can buy them as a form of restricted CIS. As suggested earlier, there does not have to be a lot of regulation—the CIS regime and the Codes supporting it can be flexible (much as property funds were introduced without the need for, for example, the liquidity requirements mandated for unit trusts and mutual funds). Alternatively, light touch regulation of token offerings may be provided through the Payment Services Act, which can potentially cover both utility and security tokens. As defined, a licensable ‘digital payment token service’ includes ‘dealing in’ digital payment tokens, which includes ‘selling of that digital payment token in exchange for any money or any other digital payment token (whether of the same or a different type)’. While the intention is to cover intermediaries rather than issuers, it may be that with enough continuity in issuing tokens, an issuer could be seen to be dealing in them as well.¹³⁰ Most tokens have a payment, as well as a securities, characteristic, and it is possible with hybrids to subject them to the lighter regulatory regime. As with the Singapore REIT story, here the goal is to find a way to regulate token offerings in order to lend credibility to them, and so pragmatism should drive the process. We did, however, argue above that light touch regulation is more apposite for token issuers than for crypto funds, and certainly not for token exchanges. But regulators across jurisdictions still have the discretion as to the kinds of token issuers that they may want to authorize or permit with some utilizing merit regulation to favour real economy businesses and others not distinguishing such businesses from financialized ones.

9 Conclusion

Ultimately, the best way forward is to regulate tokens or coins as well as crypto funds regardless of the underlying assets they invest in. This may not be the precise moment for all regulators to do so as it might still be possible for the non-interventionist approach to work with tokens and funds that do not involve securities, derivatives contracts or real estate (as is the MAS approach). But it is just around the corner if investors continue to ignore or flee the market. The disclosure of information or the standardization of rules associated with such regulation will reify them somewhat so that they are increasingly seen as intermediate interests even if not fully proprietary. This will allow them to be used as collateral themselves for loans, and also facilitate secondary trading. It will also serve the regulatory strategy of channeling funds into productive uses or even just good social purposes by SMEs and start-ups and also help with financial inclusion.