The significant EMT regime under MiCAR: a means to an end?

1. Introduction

With the adoption of the Markets in Crypto Assets Regulation (MiCAR)¹ in 2023 and its application date at the end of 2024, the European financial markets are one pillar of regulation richer. Contrary to the other regulatory pillars before it, MiCAR provides for a unique combination of capital requirements, behavioural rules, transparency aspects and market abuse provisions in one directly applicable legislative instrument. As such, MiCAR purports to achieve several regulatory goals, some of which may not always be achieved to similar degrees, or occasionally, even outright conflicting with one another.

One crypto activity caught under MiCAR particularly presents a confluence of such diverging regulatory objectives, the issuance of stablecoins. Whilst technically there are two forms of stablecoins regulated under MiCAR, here stablecoins are understood as those crypto-assets that truly have a (chance) to be stable, the so-called electronic money tokens (hereinafter: EMTs). EMTs attempt to maintain a stable value and are referenced to a single official currency.² Contrarily, asset referenced tokens (hereinafter: ARTs) are a type of crypto-asset that is not an EMT and that purports to maintain a stable value by referencing another value or right or a combination thereof, including one or more official currencies.³ There is an increasing amount of research, and consensus in the market, that claims that EMTs are the only real possibility for a crypto-asset to deliver on its stability characteristic. Other stablecoins, probably also including ARTs, are considered to be unstable due to a variety of reasons and ultimately prone to collapse. Such collapse, however, ought to be less likely for EMTs as MiCAR provides for a strict, very strict, set of rules and requirements for EMT issuers that need to be adhered to at all times.

Regardless, EMTs are not 100 per cent fail-safe nor does MiCAR purport to make them so. Better yet, it is clear from the preparatory documentation of MiCAR, and the international discussion, that the failure of stablecoins and the risks ensuing therefrom drive the regulating effort in the EU and abroad. Specifically, regulators have been largely preoccupied with exploring the financial stability risks of stablecoin failure and the effect thereof on the traditional financial sector. This, the prevention of financial stability risks, can be clearly identified as an important regulatory objective of MiCAR. A second regulatory objective of MiCAR is investor protection, which is particularly relevant in the wild west crypto-asset markets, as has been evidenced by a plethora of crypto-asset intermediary failures, foremost of which is the FTX drama. A third regulatory objective of MiCAR is the fostering of a crypto-asset industry in Europe, and a fourth objective is legal certainty. Within the regime applicable to EMTs and even more so in the regime applicable to significant EMTs, MiCAR amalgamates these four objectives.⁴

It seems, however, that, not completely surprisingly, the financial stability objective has taken the upper hand in MiCAR and has strongly coloured the significance regime of EMTs. Under MiCAR, EMTs are solely to be issued by incumbent financial sector entities, that is, banks or electronic money institutions (hereinafter: EMIs), whereby EMTs provide for a de iure transmission channel between the crypto-asset markets and the traditional financial markets.⁵ Therefore, (partly) breaking with the sectoral based legislative architecture of EU financial law, MiCAR provides for additional prudential rules for EMIs issuing EMTs and fully new rules for EMIs issuing significant EMTs. The need of such supplementary regulation has been thoroughly discussed in the run-up to MiCAR and is broadly accepted. However, limited research has been undertaken into the proportionality of the significant EMT framework, though the first research—and the broad market sentiment—argues that it may not be.⁶ Some individuals are even carefully suggesting that the significant EMT regime is meant to kill—or at least severely limit—the stablecoin market before it can ever arise.

This article argues that the proportionality of the regime must be seen in the light of the regulatory objectives served by MiCAR’s significant EMT regime. Or more specifically, in light of the financial stability objective and the interpretation given thereto by the European legislator. It appears that MiCAR’s significant EMT regime primarily serves the financial stability mandate over any other regulatory mandate and that, secondarily, MiCAR seeks to safeguard financial stability through the prevention of any spillover effects from the EMT-programmes into the traditional financial markets.⁷ In that sense, the preservation of financial stability is regulated by MiCAR in a restricted sense, that is, not necessarily extending to the stability of the whole crypto-markets, but a mere segment of it.

To this end, the article focusses its attention on EMIs issuing significant EMTs, seeing as EMIs issuing EMTs are exclusively subject to the additional MiCAR prudential requirements. Banks issuing EMTs are largely exempted from the prudential requirements of MiCAR, on the grounds that the already applicable banking regulatory framework provides for sufficient prudency.⁸ Thus, for EMIs the already applicable framework of the revised Electronic Money Directive (EMD2)⁹ was considered too weak to serve the financial stability objective and the supplementary mandatory capital and liquidity requirements imposed by MiCAR provide a clear case for consideration.¹⁰ What this article hopes to clarify in its evaluation is that the significant EMT regime apparently is configured in a manner that (almost) actively seeks to prevent large EMTs from developing, protecting financial stability in the traditional financial markets. Perhaps as a sought after side-effect, such inhibitive working would also ensure the predominance of the euro as means of payment and account and the subsequent bolstering of the digital euro project.¹¹

Below, first some essential aspects of EMTs are discussed in order to clarify the differences and similarities between electronic money and EMTs, exploring why EMIs are the entities that are to issue EMTs. Second, it will be briefly evaluated to what extent and why EMTs resemble another financial product, namely money market funds (hereinafter: MMFs) as regulated under the Money Market Fund Regulation (MMFR).¹² Third, building on these comparisons and differentiations, the article will touch on the financial stability concerns and aspects of EMTs/stablecoins and the manner in which these have defined the framework of significant EMTs. Fourth, one of the concrete prudential requirements applicable to significant EMTs is addressed, the reserve of assets, analysing where MiCAR overshoots the mere financial stability of the crypto-asset markets for the benefit of the stability of the traditional financial markets. Fifth, the own funds requirements of significant EMT issuers are considered, being a litmus test of the increased severity of financial stability concerns of the regulator as compared to the EMD2 regime. Sixth and last, the article draws some conclusions as to the (perceived) efficacy of MiCAR in ensuring financial stability and the prevention of spillover effects, concluding that if nothing else, MiCAR in any case provides for a significant regulatory market entry barrier that not many businesses will (want to) pass.

2. Electronic money versus electronic money tokens

MiCAR defines an EMT as a type of crypto-asset¹³ that purports to maintain a stable value by referencing the value of one official currency, for example, a token that purports to maintain a stable value of EUR 1 or USD 1.¹⁴ Such crypto-assets, purporting to maintain a stable value are known outside of the European Union (hereinafter: Union) and in the crypto-realm as stablecoins. The literature differentiates stablecoins in those stablecoins that retain their stable value by means of a reserve of assets, that is, by means of retaining collateral, and those stablecoins that use a technical configuration, such as algorithms, to maintain their ‘peg’ to the reference value.¹⁵ EMTs as regulated under MiCAR are part of the former group, being crypto-assets backed by a collateral of traditional financial assets and referencing a sole fiat currency.¹⁶ Algorithmic stablecoins, arguably an oxymoron, shall not be eligible as EMTs as they are not considered to be stable nor have proven to be stable.¹⁷ It is stablecoins like EMTs, that is, backed by adequate collateral and redeemable at par value for funds, that are considered to have the largest potential for large scale uptake by the economy, possibly even obtaining the status of a means of payment if the transactions costs are sufficiently low.¹⁸

Foreseeing such use, MiCAR designates EMTs as electronic money and requiring their non-bank issuers to be authorized as EMIs pursuant to EMD2.¹⁹ In doing so, MiCAR makes a parting in the realm of stablecoins, relocating the more speculative ‘stablecoins’ to the ART regime and the more reliable EMTs to a class of traditional funds, one that could actually be used for payments in the real economy. Importantly, MiCAR must not be understood to simply see EMTs as tokenized electronic money, rather MiCAR equated EMTs to electronic money due to a high degree of similarity between the use of both products.²⁰ To understand why MiCAR was considered necessary on top of the EMD2,²¹ we take a step back to the genesis of electronic money, briefly recalling what electronic money is, what the functionality of EMTs is and why EMTs are equated with electronic money.

Electronic money and its core aspects

In the late nineties, electronic money was hailed as a cost-effective alternative to cash for small value transactions and as a convenient medium to pay over the internet. The wildest claims posited electronic money as a potential replacement of cash altogether, prompting the deputy Governor of the Bank of England to foreshadow the end of central banks.²² As such, following a string of reports from the Bank of International Settlements (hereinafter: BIS)²³ and the European Central Bank (hereinafter: ECB)²⁴ the EU developed a first Electronic Money Directive (hereinafter: EMD).²⁵ This Directive regulated the issuance, distribution and use of electronic money by entities other than banks, that is, EMIs, a practice which had been considered controversial as early voices were against the issuance of electronic money by anyone other than banks.²⁶ In light of the review of EMD in 2007, it was decided that the regulatory framework had to be significantly overhauled as EMD was considered to have prevented the electronic money market from developing to its full potential due to too stringent prudential rules. Therefore, the Commission set out to apply a more proportional prudential regime to the issuance of electronic money, giving us the currently applicable EMD2 dating to 2009.²⁷

EMD2 specifies that electronic money is issued on receival of funds on either physical payment cards or in an administrative form as recorded on the internal ledger of the issuer, that is, situations where the electronic money is not stored on a physical payment instrument.²⁸ Furthermore, electronic money needs to be redeemable at par value, in order to protect the confidence of the electronic money holder that it may regain access to its funds at a 1:1 ratio.²⁹ A policy decision that was made in the period after the adoption of EMD1 was that electronic money was solely to serve as a means of payment and that any use as a store of value would be prohibited. To that end, since EMD2 electronic money cannot receive nor generate interests for its holders, making it unappealing as a store of value.³⁰

Already at the adoption and preparation of EMD2 and all the more now, it can safely be said that electronic money did not replace cash nor at any point in time posed a large risk to financial stability. Instead, electronic money mainly became a payment instrument for small (anonymous) value transactions, for example, for public transport, with attempts to introduce mainstream electronic money based payment cards being largely unsuccessful.³¹ Despite the fact that recently electronic money seems to be on the rise again, it is reasonably not to be expected that electronic money may ever present a direct threat to the dominant position of (central) bank money as a unit of account or means of payment.

EMD already considered the issuance of electronic money not as a deposit taking (ie banking) activity, thus creating a sui generis licence regime for EMIs.³² In respect of the prudential considerations, however, there was initial debate as to who should be able to issue electronic money. The ECB and its predecessor considered it suitable that only banks would be able to take up such activities and EMD1, as a compromise, largely applied the banking prudential framework to EMIs, albeit in a somewhat lighter form.³³ This regime, sometimes called a bank-light regime, was found to be much too burdensome for EMIs, with the prudential requirements not being commensurate to the risks of EMIs.³⁴ As a result, the prudential requirements were subsequently lowered in EMD2, although the principal requirement of EMD1 to hold a minimum amount of own funds equal to 2 per cent of outstanding electronic money, remained retained.³⁵ The consideration included in EMD1 that the issuance of electronic money ‘may affect the stability of the financial system’ was consequently removed in EMD2.³⁶

The functionality of EMTs

EMTs, and more specifically stablecoins, have been surrounded by similar enthusiastic claims as electronic money was at the end of the last century, with its main believers hailing EMTs as the replacement of cash and ending the traditional banking system.³⁷ On a fundamental level, EMTs are functionally similar to electronic money, namely they serve as a means of payment that is stable in value, referenced to a single currency and obtained in exchange for funds.³⁸ When successfully used as a means of payment, crypto-assets can present opportunities in terms of cheaper, faster and more efficient payments. This in particular on a cross-border basis, limiting the number of required financial intermediaries involved in payments. As for electronic money, a clear policy choice has been made for EMTs that they will only be able to serve as a means of payment and not as a store of value, notwithstanding the ECB’s remark that such function alone could already possibly lead to financial stability risks.³⁹ Thus, just like electronic money, EMTs may not be remunerated, that is, in course of staking, by their issuers or other potential third parties and need to be redeemable at par value at any time and against no additional cost.⁴⁰ Through such legal arrangements, MiCAR principally addresses the core critique of stablecoins—namely that stablecoins are not stable.⁴¹ By giving EMT holders an absolute redemption right at par against the issuer MiCAR moves the investment risk from the token holders to the token issuer, as the latter will always be required to pay out an amount of funds equal to the outstanding balance of EMTs, regardless of the investment result realized by the issuer, as is done for electronic money under EMD2.⁴² Moreover, EMIs issuing EMTs have to safeguard the funds received in exchange for the EMTs and invest them in accordance with certain investment rules, which are discussed further below. Summarizing, it can be remarked that MiCAR clearly delivers on one of its regulatory objectives when it comes to EMTs, that of legal certainty.⁴³ The legal position of EMTs has been clearly circumscribed and delimited, with EMTs being equated to electronic money and EMTs not being treated as deposits under MiCAR.⁴⁴

3. The equation of EMTs to electronic money and their fundamental differences

Considering the above, it is understandable that MiCAR designates EMTs as electronic money, seeing that they share many functional similarities, both (meaning to) serving as a means of payment. MiCAR has eliminated most differences between electronic money and EMTs, as it forces the broader group of stablecoins that qualify as EMTs into the functional regime of electronic money, creating legal certainty.⁴⁵ As such, if accepted by merchants, EMTs may function largely equivalent for a payer as would electronic money. Some form of (payment) instrument is used to initialize the transaction at which point in time the requisite amount of electronic money or EMTs is deducted from the balance of the payer. For the payee, however, a noticeable difference will probably exist. Typically, the payee is not credited with the electronic money balance spent by the payer, but rather the payee is either attributed a claim on the issuer of the electronic money, or it receives a payment in commercial bank money.⁴⁶ As a result, electronic money will usually cease to exist after the transaction has been effectuated, as the balance of the payer is reduced and the payee is credited with a claim in regular money, covered by the countervalue held by the EMI that issued the electronic money.⁴⁷ When a payer uses EMTs to pay, it is typically expected that the payment is settled in EMTs as well, due to the specificities of the blockchain. The payee will typically not receive the monetary value the EMTs represent. Subsequently, the payee may use the EMTs to effectuate a payment itself, and so on.

As a result, the acceptance of EMTs may be much broader than that of traditional electronic money and its use for cross-border payments is significantly larger. Moreover, EMTs may be expected not to be regularly redeemed as users of EMTs do not have a direct incentive to do so, provided the value of the EMT remains stable. Moreover, provided that the transaction costs remain equal, a holder of EMTs or acceptor of EMTs will continue to hold or accept EMTs due to the economic benefits that party initially identified when acquiring the EMTs. Logically therefore, the substitutive power of EMTs for fiat money is (theoretically) much larger than that of electronic money, at least for the ecosystems in which EMTs are used.⁴⁸ It is for this reason of form that regulators fear(ed) that EMTs, or better stablecoins, will be widely adopted which if combined with insufficient backing would pose a significant risk to financial stability. Nevertheless, for this same reason EMTs bear a much larger run risk than electronic money. Whereas electronic money is regularly redeemed, EMTs shall—all else being equal—only be redeemed in significant amounts when it has broken its peg, or generally enters the run zone, triggering a redemption run. This is amplified by the fact that a secondary market exists for EMTs, something that is unfamiliar to electronic money. On such secondary markets, arbitrageurs will trade the fungible EMT around its peg in business-as-usual phases, however, they will also contribute to a rapid redemption run in events of stress, or perceived stress.⁴⁹ Such trade will be heavily depended on the information on the EMT that is available to the market, as has also been shown for already existing stablecoins.⁵⁰ As such, it could be argued that EMTs are in their form more similar to MMFs than electronic money.⁵¹ This is a view to which MiCAR seems to have subscribed.

MMF participations are, like EMTs, traded on a secondary markets and MMFs are subject to strict investment restrictions.⁵² Depending on the type of MMF, they may have a constant or variable net asset value. Here we specifically consider MMFs with a constant net asset value (hereinafter: CNAVs)⁵³ and low volatility net asset value (hereinafter: LVNAV).⁵⁴ Such MMFs resemble EMTs in that they purport to retain a stable value and entitle their investors to a certain redemption right at a stable value. To that end, MMFs, a subset of collective investment unit, have also been subjected to a strict set of prudential rules akin to those applicable to EMTs under MiCAR. The purpose of these rules is partly the prevention of contagion risk and partly investor protection based, both related to redemption runs in the event that MMFs lose their stable value, that is, their peg.⁵⁵ Nevertheless, MMFs have slightly deviating rules as to their asset composition and, most importantly, certain MMFs have an allowance for a small deviation in net asset value before having to adjust the reported net asset value, permitting for a small amount of unincorporated volatility in the constant value of MMFs. Moreover, in the event of liquidity stress, that is, a redemption run, MMFs may impose redemption fees and gates, measures to prevent a first mover advantage in a redemption run.⁵⁶ For EMTs, such redemption restricting measures are strictly prohibited, bar from recovery scenarios.⁵⁷ The similarity between EMTs and MMFs ceases when considering the store of value function MMFs can have, as they may also produce yield for their investors, contrary to EMTs which will not. MMFs in their turn, are not used as a means of payment but rather as a liquidity instrument by institutional investors and are not expected to be used as a means of payment either.

Based on the similarities and key differences identified above, the dual designation under MiCAR of EMTs as electronic money and their own type of crypto-asset becomes comprehensible. Whilst EMD2 principally should provide for a technologically neutral definition of electronic money and thereby cover tokenized forms thereof, strict adherence to such approach would not adequately cover the fungible and potentially systemic nature of EMTs.⁵⁸ Therefore, EMTs’ function are broadly similar to that of electronic money yet EMTs’ form resembles that of certain MMFs. As will be discussed below, particularly in respect of the run risk EMTs are faced with, MMFs prove to be a more useful comparison than electronic money. Given the absolute nature of the redemption right for EMT holders, a strong incentive to redeem exists in stressed circumstances whereby first movers will be rewarded, an effect that cannot—or is not rather—slowed down by MiCAR.⁵⁹ Nevertheless, both EMTs and electronic money are characterized by a relatively limited maturity transformation which should, at least in comparison with other stablecoins, provide for comparative stability of EMTs, causing less financial stability risks.⁶⁰ Thus, MiCAR has chosen a compromise between the EMD2 and MMFR when it comes to the prudential regulation of (EMIs issuing) EMTs. In doing so, MiCAR provides for a top-up of EMD2, as it were, with certain requirements similar to those of the MMFR, covering the additional risks which an EMT presents.

4. Financial stability and crypto-market interconnectedness

However, what exact risk does this top-up target? Predominantly, the official regulatory concerns of MiCAR regarding EMTs seem to revolve around consumer protection and financial stability and less so on innovation and legal certainty, though as mentioned above, MiCAR goes a long way for the latter.⁶¹ Above, the financial stability risks posed by EMTs were alluded to and to a lesser extent those risks posed by electronic money and MMFs. A lot of recent research and international policy making have centred around questions of financial stability in light of stablecoins, driving the IMF,⁶² FSB,⁶³ and BIS⁶⁴ to publish a string of reports on the interplay of (global) stablecoin arrangements and financial stability concerns, with an increasing recognition of the prominence of financial stability concerns.⁶⁵ We believe that it first and foremost has been this policy concern that has—rightly—driven MiCAR’s framework for (significant) EMTs.

Financial stability and stablecoins

Since the inception of the first stablecoins, extensive debate has been had as to the potential financial stability impact of such stablecoins on the traditional financial markets.⁶⁶ Typically, the financial stability concerns derive from a potential large uptake of stablecoins as a means of payment, as discussed above, and the instability of the peg of stablecoins and the consequent effects thereof on markets and investors. To a lesser extent, research has also dealt with the increasingly unlikely scenario of stablecoins as a store of value.⁶⁷ When a stablecoin loses its peg this will, depending on the magnitude of the deviation, trigger a redemption run for the stablecoin holders that will dash for cash. Such a run causes the stablecoin issuer to have a massive need of liquidity, quite possibly leading to a fire-sale of the reserve of assets, impacting the markets wherein those assets are traded.⁶⁸ Such fire-sales are not dissimilar to fire-sales by regular financial sector entities, for example, MMFs and banks, that are faced with a sudden liquidity demand and, on a basic level, the spillover effects into the real economy are equal.⁶⁹ To this end, MMFs have been subjected to strict qualitative and quantitative investment requirements, as discussed below.

Financial stability and EMTs

Equally, EMTs are strictly regulated by MiCAR so as to credibly retain a stable value and to provide holders of the tokens with an actual redemption right at par value. This regulatory regime meets many of the appeals of the BIS and FSB, by attempting to ensure that stablecoins are actually stable setting reserve of asset requirements, own funds requirements, liquidity controls, asset segregation rules, prudential requirements in general, and a plethora of other rules. However, given EMTs designation as electronic money and the technological neutrality of the definition thereof included in EMD2, a regulatory framework containing prudential measures of its own seeking to safeguard financial stability, we must ask why EMTs are subject to the additional requirements of MiCAR.

Here we recall the difference in fungibility of electronic money and EMTs. Not only can the uptake of EMTs be significantly larger, creating so-called Global Stablecoins (hereinafter: GSCs), holders of EMTs do not have a strong incentive to redeem in a business as usual phase.⁷⁰ Such users opted for the use of EMTs for one reason or another over regular money and all else being equal such motivation does not necessarily change.⁷¹ As such, EMT holders will only have a strong incentive to redeem at the moment that an EMT loses its peg, or, is expected to lose its peg.⁷² We distinguish these two conditions as the loss of the peg of a stablecoin, and therefore EMTs, may occur due to fundamental reasons such as a negative value between the collateral and outstanding tokens or due to market sentiment, which may be informed by (the lack of) perceptions of the collateral value.⁷³ As a result, the stablecoin issuer is required to liquidate the financial assets held as collateral in order to meet the liquidity demand for the redemption requests, which may quickly spiral into a fire-sale with all of its subsequent effects for the markets wherein such fire-sale occurs.

Technically, electronic money may suffer a similar run risk, where holders of electronic money are rationally expected to run if they have the perception that the reserve of assets no longer covers the outstanding balance. However, due to the lack of a secondary market for electronic money and related lack of an available market price, such perceptions are expected to arise less frequently and consequently the volatility of the value of electronic money is lower. It is only reasonable to expect that an electronic money run will ensue when there are serious concerns about the solvency of the issuing EMI. Economic shocks having an effect on the assets held as countervalue for electronic money could be construed to have less of an impact on the run risk as the average user of electronic money is not a highly informed investor that will anticipate on a redemption run of others. Therefore, the financial stability risks and associated spillover effects of EMTs are more similar to those posed by MMFs.⁷⁴ Due to MMF participations being traded on secondary markets, MMFs are sensitive to adverse information or market sentiment in general and as such susceptible to redemption runs. As a result, the MMFR has as one of, if not the, regulatory objective the prevention of contagion effects as a result of the failure of an MMF, whereby the MMFR primarily targets the interconnectedness of MMFs with the financial sector and real economy.⁷⁵

Interconnections between significant EMTs and the financial markets

Contrary to MMFs, however, stablecoins are not already located in the regulated financial sector nor do they provide for an essential link between the banking and corporate finance markets. Crypto-asset intermediaries largely move outside of the financial sector and have a tendency to attempt to avoid any direct engagements with the regulated financial sector.⁷⁶ Though an EMT is to be issued by an EMI, a type of entity typically located in the traditional financial sector, the activities of such EMI could entirely restrict themselves to the issuance of EMTs and no other function. Therefore, if an issuing EMI would prefer to do so, it may move completely outside of the traditional financial markets. This holds to be true for issuers of stablecoins in the broadest sense, which could even opt to hold their collateral solely in the form of crypto-assets.⁷⁷ However, in light of the investment rules imposed by MiCAR, and given that more often than not stablecoins are backed by traditional financial assets, several transmission channels as facilitated by stablecoins from the crypto-asset sector into the traditional financial sector have been identified.⁷⁸

Such transmission channels could take the form of (i) direct exposures of financial sector entities to stablecoins and their reserve assets, (ii) wealth effects, that is, the loss of capital due to sharp corrections in the value of stablecoins leaving investors bereft of their money, (iii) confidence effects, that is, the erosion of investor confidence in crypto-asset markets and potentially regular financial markets possibly triggering runs in other markets than that of the stablecoin and (iv) the (future potential) use of stablecoins for payments and settlements, possibly providing for disruptions of such critical functions in the event of failure.⁷⁹ Particularly these transmission channels and their (possible) contagion effects of GSCs have incentivized regulators and standard setters all over the world to unanimously agreed that the financial stability risks urgently needed to be addressed.⁸⁰

Global stablecoins and significant EMTs

GSCs are stablecoins that have potential reach and adoption across multiple jurisdictions, providing for an international means of payment.⁸¹ Such GSCs are considered to be considerable threats to financial stability and accordingly the discourse surrounding them largely limits itself to financial stability concerns. Therefore, a strict regulatory regime is prescribed by the international standard setting bodies to GSCs, though no true GSC has manifested itself yet.⁸² The possibility that a GSC might develop rapidly, and unexpectedly,⁸³ at any moment was perhaps best exemplified in the (in)famous Libra project.⁸⁴ Libra was a ‘stablecoin’ that was intended to be launched by Facebook (Meta), said to provide for safe efficient cross-border payments for all users of the Facebook (Meta) platform. If such a stablecoin would have been developed, a possible two billion users could have resorted to Libra, creating a huge outflow of fiat money into a private stablecoin, weakening the banking sector and creating funding risks.⁸⁵ In the event that a GSC would have been subjected to a redemption run, the contagion risks would have been substantial, severely impacting the markets in which the reserve assets would be liquidated.⁸⁶ Ultimately though, Libra failed and never got to be the GSC it was prophesized to be. Nevertheless, the manifestation of the mere idea of Libra fuelled regulators’ discussions on potential financial stability implications of GSCs. As a result, the European legislator almost certainly had Libra in its mind when it drafted the significance regime for EMTs (and ARTs).⁸⁷

The significance requirements of MiCAR, in a nutshell, see to aspects of (i) the amount of EMT users, (ii) total value of EMTs issued, (iii) the volume and value of payments per day, (iv) significance of the services rendered by the EMTs on an international scale, (v) bundling of crypto-asset services with the issuance of EMTs, so-called multifunction crypto-asset intermediaries, and (vi) the interconnectedness of the EMT (issuer) with the traditional financial system.⁸⁸ These criteria have been awarded concrete threshold values in the level 1 MiCAR text and are further refined in level 2 regulation as drafted by the Commission, with several sub-indicators.⁸⁹ Ultimately, it is the EBA that shall have to decide on the basis of a bound discretion as to whether an EMT is significant. That is, if the EBA considers the thresholds met, it must designate an EMT as significant.⁹⁰ In such assessment, the EBA shall have to base its decision on the objective thresholds and core indicators for the more qualitative criteria and only when these do not give a conclusive answer, on the basis of the ancillary indicators. If three of the significance criteria are met, the EMT shall in principle be designated as significant. The EBA considers it appropriate that a methodology, including a hierarchy and weighting of the assessment of the significance criteria, is developed, which is to be published after the delegated act on significance criteria is published.⁹¹ As part of the thesis of this article, it is expected that the requirements seeing to the interconnectedness with the financial system will be one of the, if not the, most important criteria.⁹² The interconnectedness is assessed by three core indicators. The first, is assessed by two sub-indicators assessing (i) the amount of non-deposit assets in the reserve of assets issued by financial institutions and (ii) share of non-deposit assets that are covered bonds issued by credit institutions. As will be discussed below, these ratios may very quickly indicate high proportions due to the investment restrictions and limited availability of eligible assets. The second core indicator expresses the direct and indirect interconnectedness, by assessing the share of EMT issuer’s asset holdings relative to total supply of specific financial instruments (eg units of a UCITS, sovereign bonds). The core indicator will, once the required data becomes available, see to the share of EMTs issued that are held by financial institutions. The ancillary indicators see to (i) the ownership structure of the issuing EMIs, particularly the presence of financial institutions therein, (ii) the concentration of reserve assets or deposits in financial sector (custodians) and (iii) to the portfolio overlap of reserve assets with other ART/EMT issuers.⁹³

These indicators clearly seek to define any excessive exposure to the financial sector, be it direct or indirect, as a sign of significance of EMTs. Such an approach is unheard of for MMFs and regular EMIs and if nothing else, openly demonstrates the EBA’s perception that the crypto-asset markets are a separate item from the traditional financial sector and that any comingling of these sector is a risk factor in and of itself.⁹⁴ As a counter argument, the remark by EBA that: ‘the expression ‘financial system’ also includes issuers of ARTs or EMTs themselves, and hence this type of interconnectedness should also be covered in the significance assessment’ would read as an outright contradiction of the previous statement. However, no such considerations exist for EMIs and MMFs, and it seems possible that even this latter ancillary indicator is meant to identify concentrations of EMT related holdings across EMT and ART issuers. In that sense, it is more the mirror of the penultimate ancillary indicator, seeing to concentrations of assets in financial sector entities. Moreover, due to the restricted investment rules of the reserve of assets, a significant overlap is to be expected by design and would not necessarily provide any useful insight as to the interconnection of EMT issuers with other EMT or ART issuers.

Some authors have criticized the rigid functioning and comparatively low thresholds of the significance regime, and its severe additional prudential requirements,⁹⁵ discussed below, as unsuitable to the regulatory goal of preventing systemic or other risk. They compare, and plead for the closer alignment with, the regime concerning the designation of global systemically important banks (hereinafter: G-SIBs). They note the relative discretion in designating G-SIBs and the scalability of the consequent additional prudential requirements applied to them, commensurate to the systemic risk posed by such institutions.⁹⁶ By comparison, the authors argue, no concrete evidence of systemic risk posed by EMTs has been proven yet and the requirements set under the significance regime would be prohibitive to the development of larger EMT programmes.⁹⁷ For example, they propose to set the minimum limit for designation as a significant EMT at an issued volume of EUR 100 billion or more.⁹⁸

Interim conclusion

Such a plea, however, is based on a wrong interpretation of the regulatory goal of the significance and the broader EMT regime. The additional regulatory requirements seek to paint a clear and contained picture of EMTs’ interconnection with the financial sector, and by doing so limiting financial stability risks.⁹⁹ The regime does not seek to express a systemic risk of the EMTs strict sensu. Instead, the EMT regime, particularly the significant EMT regime, seeks to contain the financial risks inside the crypto-asset sector and attempts to limit their size. A key indication resides in the fact that neither EMD2 nor the MMFR contains a significance regime, that is, all MMFs and EMIs are principally treated equal. EMIs issuing non-significant EMTs are principally also dealt with as such, largely remaining subject to the provisions of EMD2. Yet where an EMT becomes, or could become, a GSC, MiCAR provides for the significance determination and all the (mechanic) consequences thereof. The intensification of the prudential requirements under the significant EMT regime is no mere coincidence, but directly targets the shortcomings of EMD2 in not providing such regime.¹⁰⁰ Rather, the significance regime for EMTs seeks to prevent the creation of any systemic risks ex ante, setting the significance thresholds comparatively low, or where EMTs have become systemic, by regulating EMTs in such a way that they have a minimum spillover effect into the traditional financial sector and limited incentives to develop their business. The ECB even said so much in its Opinion on MiCAR: ‘Additional mechanisms to incentivise issuers to limit the scale of issuance, including stress-testing requirements with possible capital add-ons, should be included in [MiCAR].’¹⁰¹ The lessons learned after EMD1 surely must also still be in the minds of the legislator, as it concluded in its review that the (too) strict framework of EMD1 had effectively prevented the electronic money market from developing. EMD2 relaxed this framework, and the electronic money market did develop, which is possibly an additional reason why MiCAR does not want to rely on the (facilitating) EMD2.

This interpretation of the significant EMT regime will be further evidenced by analysing some of the prudential requirements applicable to significant EMTs and comparing these to the prudential requirements of MMFs and regular EMIs. Once the prudential requirements are approached through the lens of the aforementioned regulatory objective, they are understandable. The deviations from EMD2 and mixture with the contagion risk oriented MMFR can be explained. The requirement where such intention can be clearly seen, is in the provisions relating to the reserve of assets.

5. The reserve of assets: an exercise in limiting contagion risks

As mentioned before, EMIs are required to safeguard the funds they receive in exchange for EMTs, that is, equivalent to a situation where the EMI would be issuing electronic money.¹⁰² These funds can be subsequently invested by the EMI in secure, low risk assets denominated in the same currency as the EMT references. Such investment of the reserve assets generally constitutes the business model of stablecoin issuers, which provides for a form of maturity transformation.¹⁰³ The reserve of assets needs to be composed and managed in a manner that the liquidity risks associated to the permanent rights of redemption of the EMT holders are addressed.¹⁰⁴

Under EMD2, eligible assets for investment of countervalue for electronic money, are those assets that attract a 1.6 per cent risk charge pursuant to market risk regime of the Capital Requirements Regulation (CRR),¹⁰⁵ broadly speaking being government and public sector bonds, bonds issued by certain credit institutions, bonds issued by certain highly rated corporates, certain covered bonds and certain securitization exposures.¹⁰⁶ Moreover, units in undertakings for collective investment in transferable securities (UCITS) that invest in the aforementioned assets are also eligible. When an EMT is deemed significant the investment of the funds received in exchange for EMTs is subject to the specific MiCAR rules on the reserve of assets.¹⁰⁷

The reserve of assets of EMTs shall have to at least comprise an amount of assets that together have an equal value to that of the outstanding EMTs, ensuring the issuance at par value. The investment of the reserve assets is subject to a specific ratio of eligible assets, comparable to the diversification requirements of the MMFR but fully incomparable to EMD2.¹⁰⁸ At least 30 per cent of the reserve assets held by the EMI issuing EMTs must consist of deposits with a credit institution, 60 per cent in the case of significant EMTs. The remaining 70 per cent or 40 per cent of reserve assets may be invested in secure, low-risk assets that qualify as highly liquid financial instruments, that is, those assets with minimal market risk, credit risk and concentration risk (hereinafter: HLFI). As becomes clear from the name of HLFI, the European legislator stresses the importance of adequate liquidity to match the potential rapidity of redemption runs that EMTs are exposed to, as mentioned above.

Assets included in the reserve of assets have to be valued on the basis of market prices, determined by using a mark-to-market approach¹⁰⁹ and if not possible, a mark-to-model approach.¹¹⁰ MiCAR therefore does not allot for the amortized cost approach that is permitted for certain CNAV and LVNAV MMFs nor for the small (0.2 per cent) room of deviation of the net asset value from the market value of the reserve of assets.¹¹¹ As a result, the reserve of assets of EMTs shall be permanently subject to fluctuations in value, exposing the issuing EMI to constant market risk. Recently, the Commission reviewed the functioning of the MMFR and the amortized cost method included therein and concluded that it should be retained, though it may be somewhat distortive, as otherwise a large outflux of participants from such MMFs may be expected. Moreover, the Commission considered provided the safeguards included in the MMFR work appropriately, no systemic risks are created.¹¹² One therefore wonders why no equivalent regime was included for EMTs, especially given the fact that the reserve of asset requirements resembles those of LVNAVs eligible for the amortized cost method and they are principally exposed to the same systemic risks.¹¹³ Whilst no considerations on this can be found, it seems not unlikely that the legislator considered unpreferable to award such a stabilizing regime to EMTs as this might have increased the uptake of EMTs as a form of MMF.

Deposits in the reserve of assets

The significant EMT regime doubles the amount of deposits that need to be held in the reserve of assets. On the one hand, one could argue that such increase in deposits is due to the expectation that runs of significant EMTs will transpire even faster than those of non-significant EMTs. On the other hand, it seems likely that the doubling of deposits also seeks to limit the amount of HLFI held by the EMI. Where an EMT becomes a GSC, that is, significant EMT, it might hold very significant amounts of short-term money market debt.¹¹⁴ If this had to be sold in a fire-sale, a large impact on those markets can reasonably be expected. Recalling the importance of the HLFI and not deposits for the significance criterion related to interconnectedness, the limited 40 per cent reserve of assets that may be invested in HLFI seems like an effort to reduce the spillover from GSCs into the financial sector. By comparison, it could be argued that a doubling of the required proportion of deposits is not directly explained by any increased run risk. Is a significant EMT truly subject to a faster redemption run than a non-significant one? And if so, would it be twice as fast requiring a double amount of deposits?

To answer these questions, we must briefly consider the detail surrounding the allocation of the deposits at banks by the EMI. EMIs issuing significant EMTs have to observe a concentration limit of 25 per cent of reserve asset deposits per G-SII or O-SII,¹¹⁵ 15 per cent per large credit institution,¹¹⁶ and 5 per cent per other institution. Originally, the EBA proposed a stricter 10 per cent deposit limit per significant institution and a 5 per cent for other institutions, thereby imposing a stricter regime than the MMFR which permits 10 per cent, or exceptionally up to 15 per cent, of deposits at a single bank.¹¹⁷ The regime was relaxed slightly in light of market complaints about the ability to find enough willing bank counterparties to accept the deposits. The EBA, recognizing this response altered the concentration limits, stating that it would: ‘[…] like to avoid any possibility that ultimately compliance with the concentration limits might cause unintended effects in the crypto-asset market and more generally in the whole financial system. The EBA would also like to avoid creating potential barriers of entry into a market that is promoted by the legislator.’¹¹⁸

Furthermore, EMT-related deposits may only be 1.5 per cent of the total assets of the deposit bank.¹¹⁹ Originally, EBA proposed a concentration limit of 2.5 per cent but lowered this after the consultation in response to the heightening of the concentration limit mentioned above, citing the reasons that it wishes to incentivize EMT issuers to deposit funds with larger banks. However, a small calculation example shows the limited significance of the relaxation of the concentration limit per institution if seen against the balance sheet limit of 1.5 per cent. If an issuer of an EMT has EUR 10 bln. outstanding, it can deposit up to a maximum of EUR 2.5 bln. with G-SII banks. In order for a G-SII bank being able to absorb such deposit, it needs to have a minimum balance sheet of EUR 166.67 bln, that is, (2.5 bln/1.5%)* 100%. Under the old framework, the same EMT issuer could deposit a maximum of EUR 1 bln. with the same G-SII. In that instance, the G-SII only had to have a total balance sheet of EUR 40 bln., that is, (1 bln/2.5%)*100%. The amount of eligible institutions has therefore decreased under the revised framework, though most G-SIIs will meet the threshold balance sheet size. The prudential rules for banks and their exposures to EMT issuers are not discussed in this article; however, it is not unthinkable that large banks themselves will only accept a certain amount of deposits from EMT issuers due to the prudential rules applicable to them, further limiting the amount of institutions available to each EMT issuer.

On the whole, including the other HLFI assets, the exposure of the reserve of assets to a single credit institution, or to entities with which the given credit institution has close links, shall not exceed 30 per cent of the reserve assets of a given EMT, 25 per cent under the first EBA proposal.¹²⁰ Moreover, EMIs depositing funds at a bank must perform a creditworthiness assessment in respect of that bank, contemplating whether the credit institution is expected to default on its obligation. In this assessment, an EMI must take account of its risk appetite and the final volume of deposits held at that banks.¹²¹Prima facie these requirements see to protect the EMI against the possible unavailability of a part of its reserve assets, which would align with the qualitative liquidity requirements for banks on which this measure is modelled.¹²² Moreover, the additional safeguards can be interpreted as off-setting part of the EMI’s liquidity and credit risks in the relatively large deposit holdings.

However, in its motivation for the concentration limits, the EBA considers: ‘A high concentration of deposits with a limited number of banks shall be avoided. This is to mitigate the risk arising from material interconnectedness between the financial system and the crypto ecosystem.’¹²³ In the impact assessment weighing the policy choices for the creditworthiness assessment, EBA states: ‘The interconnectedness between the banking system and crypto activities requires to implement prudent approaches of this kind to avoid any expansion of any risk to the financial system.’¹²⁴ It seems, therefore, that the requirement to hold deposits at banks and the doubling of that requirement for significant EMTs is surrounded by safeguards to protect the deposit holding banks from potential contagion risks. Moreover, it can be argued that weaker banks are protected from potentially fleeting EMT-related deposits by means of the creditworthiness assessment whilst the concentration limits put the largest potential liquidity burden on significant banks, that is, those banks with better access to liquidity on the capital markets and better risk management practices.

In doing so, MiCAR limits the potential impact of a redemption run on EMTs by creating a form of institutionalized burden sharing.¹²⁵ As a result, the ‘well’ that is the money market is not (directly) poisoned as the deposit holding banks will cushion the liquidity shock through their own liquidity buffers and management. If the proportion of deposits in the reserve of assets had been increased further, new contagion risks might have been created as too large of a proportion of the reserve assets would have been held directly with banks.¹²⁶ As such, it is recalled that the deposit part of the reserve of assets is principally excluded from the significance assessment, where the criterion related thereto restricts itself to the HLFI assets and the deposits are only reflected in ancillary indicators.

A more cynical motivation could also reside in the inability of many crypto-asset parties to establish an account with banks due to the aversion of banks of the anti-money laundering and counter terrorism financing risks of the crypto-asset market. This argument was somewhat defused by the amendment of the final RTS, as the amount of deposit banks for a significant EMT issuers went down from six to three, though for the reasons calculated in the example above, the concentration limits remain challenging.¹²⁷ Particularly, the banks that traditionally are inclined to serve crypto-asset parties will be filtered out, as these are generally the smaller less well managed banks, such as those that failed in the US in March 2023. Of these banks, a EMI issuing a significant EMT would most probably still need 12, which if achievable at all, would be high operational burden to manage for the EMT issuer. As such, this can be seen as another blocking or limiting factor on significant EMT emergence. Adding to this is the interconnectedness criterion in the significance assessment, where weaker banks shall lead to higher scores in the core indicators, for example due to a higher level of concentration of EMT and ART issuers in the same institutions thus leading to faster significance designation. Moreover, banks themselves face hefty liquidity charges for servicing EMT issuers, as the deposits held by the EMT issuer at the bank are most probably classified as wholesale funding, effectively treating the EMT-related deposits as ineligible for the funding of the bank. Whether small banks are willing, or able, to take this liquidity hit, is something that remains to be seen.¹²⁸

Investment of reserve assets

The part of the reserve assets not allocated to deposits can be invested in HLFI denominated in the same currency as the EMT references.¹²⁹ The level 2 regulation detailing what is to be understood as HLFI primarily copies the framework of high-quality liquid assets (hereinafter: HQLA) for purposes of the Liquidity Coverage Ratio (hereinafter: LCR) of Article 412 CRR.¹³⁰ The perimeter of assets eligible as HLFI is set at level 1 liquid assets with a 0 per cent haircut, generally consisting of government bonds and extremely high-quality covered bonds and UCITS.¹³¹ Each type of asset that has been designated as eligible is subsequently bound to concentration limits vis-à-vis the issuer of the asset and the proportion of the reserve assets. Certain adaptations to the HQLA framework have been made for that of HLFI. For instance, the requirement in respect of government bonds that are included as HQLA for a bank to be valued on the basis of easily available market prices or for such bond to be traded on recognized exchanges or repo markets, does not apply for HLFI.¹³² A simplification for issuers of EMTs exists in the fact that extremely high-quality covered bonds and UCITS included in HLFI shall not be subject to any haircut.¹³³ Nevertheless, the relatively simplistic, mainly credit risk driven, investment rules of EMD2 are substituted by a regime more in the lines of the rules applicable to CNAV or particularly those applicable to LVNAV MMFs.

Other than assets eligible as HQLA for the LCR, the maturity of HLFI is not 30 days, but a mere five (5). This very short time horizon is supported by the expectation of lightning fast redemption runs of EMTs,¹³⁴ ie much faster than the redemption runs for banks which the LCR seeks to cover.¹³⁵ It can be disputed to what extent the LCR is still up-to-date, however, a redemption run that transpires during a five (5) day period is not only a regulators’ perception, but hard reality, as was seen in the US banking turmoil during spring 2023. The rules on HLFI provide for an unwinding mechanism regarding secured funding, lending and collateral swap transactions which considers the time horizon of five (5) days, pursuant to which the reserve of assets is to be increased or decreased with the amount of funds or assets to be paid out or received in connection to those transactions during the five (5) day liquidity horizon.¹³⁶ Concretely, the availability of assets included in the reserve of assets, that is, deposits and HLFI, to answer to redemption requests, needs to be assessed along a 1-day and a 5-day maturity.¹³⁷ At least 40 per cent of the reserve of assets needs to mature or be callable within 1 working day for significant EMTs.¹³⁸ At least 60 per cent of the reserve assets including those with a 1-day maturity needs to mature or be callable within 5 working days.¹³⁹ These timeframes are deduced from the recent failure of Signature bank,¹⁴⁰ which experienced a 20 per cent run off from deposits related to crypto activities, and the MMFR which requires LVNAVs to hold 10 per cent of their assets with a maturity of 1 day and 30 per cent of their assets in 1 week. The remaining 40 per cent of reserve assets (deposits and investments) are not bound to any maturity requirements, as the EBA considered such limitations unnecessary (for now).¹⁴¹ As a result, the average maturity of the reserve of assets may be higher, as issuers will likely compensate with HLFI with longer maturities for the remainder of the reserve of assets.¹⁴² Nevertheless, issuers must manage other risks in the reserve of assets adequately in accordance with their liquidity management policy.

Akin to the deposits in the reserve of assets, EMIs must also consider concentration limits with respect to their custodians of HLFI.¹⁴³ This limitation in exposure to a single counterparty is largely achieved by means of qualitative liquidity management requirements. Again, the concentration limit must largely be seen as a measure to limit spillover effects as a rapid liquidation of HLFI might also draw heavily on the liquidity position of the custodian. An important note needs to be made in respect of EMTs denominated in official currencies of the Union other than EUR. The short-term money markets in such currencies may be highly limited, with a large part of eligible HLFI generally only issued by one issuer, being the respective central government. Even the money market denominated in EUR, the largest money market in the Union, is already not very liquid with MMFs already holding around 50 per cent of the commercial paper issued thereon.¹⁴⁴ Additionally, as MMFs and EMT issuers will generally hold the same type of money market instruments, given the limited absorption capacity of the money markets, once one fire-sale has been initiated there is a limited chance that a second party can successfully liquidate its portfolio.¹⁴⁵ It is therefore not wholly unthinkable that a consideration of the European legislator in drafting the significance regime of MiCAR was to limit the involvement of EMT issuers in such markets, preventing the money markets to be even further concentrated and further exposed to similar, unpredictable, run risks. If EMTs are kept small, then a fire-sale of their reserve assets may not ‘spoil’ the money markets for other traditional financial sector participants.

Partly in recognition of such risks and generally the market fluctuations in the money market, MiCAR requires EMT issuers to hold a safety margin of reserve assets protecting against market fluctuations, in the form of a mandatory form of overcollateralization (proposed by the EBA).¹⁴⁶ The mechanism drafted by EBA caters for price fluctuations during stress scenarios.¹⁴⁷ Whilst overcollateralization is a fully understandable concept, and indeed MiCAR requires adequacy of the coverage of the reserve assets at all times, no clear legal basis has been included for such a mandatory requirement.¹⁴⁸ Possibly, it would be more on the path of the issuer to ensure that the reserve of asset suffices to cover the outstanding EMTs at all times, in line with its liquidity management policy. The Liquidity Requirements RTS in its turn could then have set out rules how to ascertain the possible effects of market risk by means of general techniques for stress scenarios, instead of requiring an automatic add-on.¹⁴⁹ Basically, the required overcollateralization expresses the positive difference between the total amount of EMT outstanding (ie the assets referenced) and the market value of the reserve assets, during any overnight period over the previous five (5) years, coined as the historical look back approach.¹⁵⁰ The overcollateralization calculation is to be made on a daily basis, as to connect to the obligation of the issuer to verify daily that it complies with the reserve of assets requirement.¹⁵¹

Moreover, EMT issuers are required to draft liquidity contingency plans wherein early warning signs of redemption runs are to be included.¹⁵² This qualitative approach is contrary to the quantitative regime initially proposed by the Basel Committee on Banking Supervision (the basis risk test), whereby the level 2 text does not set concrete thresholds itself. If the issuer notices the value of outstanding EMTs deviating from the reserve of assets, it shall have to be able to promptly and appropriately react to such shortfalls, pre-empting any risks of market panic. In course of the stress testing exercise EMIs need to reverse stress test the limits of the resilience of the reserve of assets,¹⁵³ whereby that limit could be used to inform future liquidity measures, or potential adjustments of the overcollateralization. One could compare the deviation regime applicable to LVNAV MMFs, that is, that where they have to switch from a stable NAV to a floating NAV when the mark-to-market value of the reserve of assets deviates with more than 20 bps than the amortized cost NAV, to the overcollateralization of EMTs. MMFs have a strong incentive to stay within this collar, so as not to lose market confidence, overcollateralizing where necessary. EMT issuers will want to keep the mark-to-market value of the reserve of assets above the par value at all times as they will immediately lose market confidence. ESMA has investigated that LVNAV MMFs’ resilience would be aided mostly by expanding the NAV deviation collar, and not so much by retaining more, or less, weekly liquid assets though noting that MMFs with high weekly liquid assets are less prone to liquidity outflows than those with low weekly liquid assets.¹⁵⁴ Based on these findings, it could be argued that EMT issuers will, once faced with large redemption requests, be incentivized to first sell their less than weekly or daily liquid assets and only thereafter use their daily/weekly maturing assets. The market losses, however, incurred by selling such relatively illiquid assets shall have to be cushioned entirely by the overcollateralization as a breach of par value, that is, loss of the peg, would mean the end of the EMT.¹⁵⁵ The overcollateralization requirement itself will simultaneously increase during times of market stress. As such, the composition of the reserve of assets does not necessarily seem to be catered towards investor protection, but more at heavily penalizing EMT issuers, as research has already shown that more efficient paths to MMF (EMT) resilience exist, such as permitting a switch to a floating NAV.¹⁵⁶ Nevertheless, such an architecture does connect more to the electronic money function of EMTs, as a floating NAV for an EMT is prohibitive for the uptake as a means of payment and thus having a stable value.

Redemption and recovery plans

It is at this point that a short reflection on the recovery and redemption plans is in order. Under MiCAR, EMT issuers are required to draft recovery plans that should enable issuers to return to going concern situations in times of stress and redemption plans support the orderly redemption of EMTs, when they are unable or likely to be unable to fulfil their obligations.¹⁵⁷ This latter requirement is deemed to have been met if the prudential requirements are no longer met by the issuer or if the reserve of assets requirements are no longer met. Moreover, competent authorities are to take a range of circumstances into account in order to assess whether such requirements will no longer be met in the near future.¹⁵⁸

The recovery plan is technically the first to come in to action in stress events for the EMT issuer as its recovery triggers/indicators are set at a level where they do not necessarily entail a situation where the issuer is unable or likely to be unable to fulfil their obligations. One of the recovery indicators must be based on the de-pegging risk, which permits a 1 per cent deviation interval of the par value of issued EMTs against the reserve of assets.¹⁵⁹ If recovery indicators are breached, the issuer shall escalate the situation to management board level within 24 hours of such breach and notify the competent authority within 24 hours of the activation of the escalation procedure.¹⁶⁰ It is in principle up to the EMT issuer to decide what action is appropriate, if any.¹⁶¹ Possible recovery actions that could be taken by the issuer are the imposition of liquidity fees on redemption request and the setting of limits on the amount of EMTs that may be redeemed in a certain time period. These limitations are the only permitted infringements of the absolute redemption right of Article 49(6) MiCAR that may be effectuated by the issuer. However, the calibration of these limitations is principally left to the issuer, which shall have to substantiate these with qualitative and quantitative evidence, which calibration shall be assessed by the competent authority. Moreover, in the event that recovery triggers are activated and the issuer fails to comply with the requirements on the reserve of assets, the competent authority itself may engage any of the recovery measures and suspend any redemptions of the EMTs.¹⁶²

Redemption plans must ensure the equitable treatment of EMT holders and safeguard that they do not incur undue economic harm. For instance, redemption plans must specify that the liquidation costs of selling the reserve of assets, for example, intermediary or advisory costs, may only be allocated to the proceeds thereof if and when the EMT holders have been fully satisfied in their redemption claims. EMT issuers must analyse certain redemption scenarios in their redemption plans and how they can maximize the liquidation gains in amount and time, considering the composition of the reserve of assets and market circumstances.¹⁶³ Interestingly, drafters of redemption plans need to account for the effect of the liquidation of the reserve of assets for the markets in which the assets are traded. Such an external requirement is one the one hand logical as it seeks to prevent the further spreading of financial unrest to the affected markets, whilst on the other hand it seems directly opposed to the requirement of maximizing the return for token holders in as short of a time frame as possible.

Apparent regulatory objective

The general design of the reserve of assets regime apparently serves a dual regulatory goal. First, MiCAR attempts to protect investors against egregious liquidity mismatches or risk taking by the EMT issuer. Second, the concentration limits, compare the more investor focussed wording of the MMFR: diversification rules, seek to protect the short-term money markets from the potential effects of a fire-sale by the EMT issuer in light of a redemption run. We consider there to be a focus by MiCAR on the latter, contagion risks, for several reasons.

In first instance, the prudential rules of EMD2 should theoretically have been sufficient in ensuring the protection of the investor. Admittedly, due to the MMF form of EMTs, such prudential rules would not have been sufficient and there indeed is a need to supplement the requirements from an investor protection angle. To that end, it is reasonable that EMTs would be partially regulated as MMFs, giving their fungible nature and liquidity profile.¹⁶⁴ What cannot be directly explained is why the reserve of assets requirements are so much stricter than those applicable to LVNAVs or why EMTs do not benefit of the stabilizing regimes of MMFs.

Arguably, the first mover risk that is the underlying reason for redemption gates and fees in the MMFR also exists at EMTs, prompting the inclusion of redemption gates and liquidity fees in both the MMFR and MiCAR. Whereas in the MMFR these have been concretely described in the level one text, MiCAR leaves the calibration of such measures up to the issuer, whereby the EBA has provided (draft) guidelines that set the manner in which redemption gates and liquidity fees have to be calibrated.¹⁶⁵ As such, the level set for such measures shall be unknown for market participants, potentially mitigating the mitigating effect on the first mover risk that redemption limiting measures are supposed to have. On the other hand, the MMFR review pointed out that the thresholds in the MMFR for the imposition of the redemption gates and fees led to an anticipatory run risk and thus recommended that such thresholds should be removed from the MMFR altogether. In that sense, the manner of implementation of such liquidity tools in MiCAR is more in line with the most recent thinking on the matter. On the other hand, ESMA concludes in the MMFR review that competent authorities should not have the ability to impose redemption gates and fees, as that may very well create its own contagion and (anticipatory) run risk.¹⁶⁶ MiCAR, however, does give the competent authority the possibility to effectuate recovery measures, which presumably include redemption gates and liquidity fees, going against the conclusions of ESMA in the MMFR review and thus reinstating a possible run risk. This seems to reflect a reluctance of the legislator to leave the recovery process fully to the EMT issuers.

An approximation of the minimum required daily/weekly liquid assets may therefore also function as a red flag in EMTs as it does for MMFs. The overcollateralization requirements should ensure a certain liquidity buffer cushioning market fluctuations in the reserve of assets, however, this mandatory overcollateralization is not a releasable buffer, it appears to remain applicable during periods of stress. It seems therefore that the overcollateralization does little for investor protection during stress periods and solely provides for investor comfort as long as no one has started running. On the other hand, the mandatory overcollateralization will be a very expensive measure for EMT issuers, that will have limited returns on their investments to finance the buffer with. Therefore, EMT issuers might have to pony up significant amounts of capital from their own pockets to support the reserve of assets, be it in order to keep up the overcollateralization or in course of the recovery plan to restore the capital position.

The concentration limits in respect of banks with which the EMT issuer holds deposits will also provide for a severe operational burden that may be hard to overcome, especially in light of the (understandable) AML/CTF-driven aversion of banks in serving crypto-asset parties. Moreover, the increased percentage (60 per cent) of deposits required by significant EMT issuers does not let itself translate into investor protection concerns. Larger stablecoins are not per se susceptible to faster redemption runs nor should there from an investor protection point of view be a large preference for deposits over repo’s or other assets that could be liquidated within one or five days. No, these requirements can be explained when looking through the financial stability lens, and specifically one that seeks to protect financial stability by compartmentalizing EMTs away from the traditional financial sector. Given the already limited size of European money markets and the fragility thereof as evidenced during the COVID-19 pandemic, it appears to be apparent that the European legislator wanted to shield the money markets from unstable stablecoin (EMT) issuers. By limiting the proportion of HLFI in the reserve of assets, as well as the feasibility of a significant EMT arising, the impact for money markets has been kept to a minimum.

It can be stated that the absence of hard thresholds for the imposition of redemption gates and fees limit the first mover incentive and that the higher levels of daily/weekly liquid assets bolster the liquidity profile of the EMT. Arguably, however, the first mover incentive is amplified as an EMT losing its peg exposes investors to a credit risk, as the EMT issuer may not be able to redeem any tokens at all if the losses taken in the liquidation of the reserve of assets exceed the own funds and overcollateralization maintained. Such credit risk may push investors to redeem their EMTs before it even loses its peg, provided they are sufficiently informed to come to such an anticipatory decision, which is confused by the unknown calibration of the recovery triggers prompting redemption gates and fees, which may also be applied by competent authorities. The disclosure requirements incumbents on EMT issuers will surely aid investors into making such decisions, though they are left to make a guess as to when liquidity management tools may be triggered.¹⁶⁷

6. Own funds requirements for EMIs issuing EMTs

A final indication of the European legislator’s regulatory intent can be found in the new own funds requirements of MiCAR. Contrary to the general rule of MiCAR that issuers of non-significant tokens shall remain subject to the own funds requirements of EMD2,¹⁶⁸ MiCAR prescribes a regime that ultimately can lead to more than double the normal capital requirements of EMIs. Though such an increase perhaps appears steep, we have already established that GSCs present a unique financial stability risk which needs to be covered. Arguably no regulatory measure is better suited against the coverage of financial stability risks than capital requirements. Though, it can be reasonably argued that the reserve of assets, especially taking into account the overcollateralization, should already adequately provide for the risks to which an EMT issuer is exposed.¹⁶⁹ Nevertheless, this would be foregoing the preventative, maybe even prohibitive, intent of the capital requirements.

EMIs issuing (significant) EMTs are required to maintain 2 per cent, or 3 per cent for significant tokens, of the value of the reserve of assets as own funds, fully in the form of common equity tier 1 (CET1).¹⁷⁰ Due to the phrasing of the requirement, however, referencing the reserve of assets and not the outstanding EMT value as the EMD2 does, this requirement shall always be higher as it shall have to include the overcollateralization and any (unrealized) investment gains included in the reserve of assets. Moreover, this amount may be increased with a further 20 per cent due to indications of higher risks in a process that is not wholly dissimilar to the Supervisory Review and Evaluation Process (SREP) under the banking framework.¹⁷¹ The ultimate risk that is being assessed by the competent authority in the application of this capital add-on is the run risk and thus the EMTs impact on financial stability. The factors that are to be assessed in this EMT SREP are (i) whether the EMT issuer is likely, not expected per se, to breach the reserve of assets requirements, (ii) whether at all times the redemption of the EMTs at par value is ensured and not jeopardized in any circumstances; and (iii) where there is another increased risk of an essential failure or significant deterioration of the value of the reserve assets.

Furthermore, competent authorities shall be able to impose an additional 20–40 per cent own funds requirement as a result of the risk outlook and stress testing, and this stress testing is strictly motivated by financial stability (and consequent contagion) concerns,¹⁷² as undertaken quarterly by issuers of significant EMTs.¹⁷³ The EBA considers that these capital add-ons are without prejudice to one another and should be applied cumulatively, for a possible maximum add-on of 60 per cent of the own funds calculated as a result of the volume of the reserve of assets. This last add-on, however, is a bound discretionary ability of the competent authority, as Article 35(5) MiCAR uses the verb shall instead of may in Article 35(3) MiCAR, as a result of which the add-on shall have to be imposed if the stress testing indicates certain vulnerabilities. Nevertheless, increases in own funds should only be requested when there is a higher degree of risk which is not already covered by the issuer.¹⁷⁴ The ability for competent authorities to apply a 20 per cent lower capital requirement, as is possible for EMIs issuing electronic money, is barred by MiCAR. As a result, issuers of significant EMTs may be obliged to keep up to 4.8 per cent of the average of the reserve of assets as own funds in solely CET1. This is more than the double of the maximum own funds requirement under the EMD2 (2.4 per cent of issued e-money) and much higher than any capital requirement banks would have to keep for the largely 0 per cent risk weighted HQLA reserve assets. An interesting detail of the own funds add-on implementation plan is the requirement to ensure that the additional capital qualifies as CET1 after the deductions of Article 36 CRR have been applied in full, without the application of the threshold exemptions of Articles 46(4) and 48 CRR. Basically, this means that exposures held by the EMI to financial sector entities’ regulatory capital instruments, for example, in the HLFI insofar eligible, that would otherwise not be deducted from CET1 under CRR, now do have to be deducted by issuers of significant EMTs.

In this measure, the objective to keep the financial sector and crypto-asset market separated very clearly comes to the forefront. The entire set-up of the additional capital requirements for significant EMTs is to shield the traditional financial sector by providing for a robust cushion of capital held by the shareholders, often not traditional financial market participants or if so this serves as an indicator of significance,¹⁷⁵ of the EMI. This buffer could also be committed to the reserve of assets by investing it in HLFI to meet the overcollateralization requirement. The fact that the additional own funds requirements are partly triggered mechanically upon the designation of the EMT as significant further signals that the legislator did not necessarily meant for the capital requirement to directly provide against any systemic (contagion) risk, but to prevent it. The higher the regulatory capital buffer, the stronger the resilience perception of token holders of the issuer being the party that their redemption claim ultimately sees to, and furthermore the more losses will be internalized by the EMT issuer without spilling over to the traditional financial markets.¹⁷⁶ Furthermore, MiCAR does not contain a resolution regime for significant EMTs, although the severity of the going concern requirements might point towards the necessity of such a regime. The absence of such regime could be construed as a sign of the satisfaction of the legislator that it has sufficiently mitigated any spillovers due to EMT failure, by means of preventing the creation of spillover candidates ex ante.

Only the spillover rationale can truly explain the deviation from the capital requirements for MMFs and regular EMIs as a resilience of the peg argument could have logically been addressed through further reserve of assets requirements, as the ability of the token holder to discern between CET1 capital and reserve assets may be doubted. Similarly, consumer protection would have been adequately served with the requirements of the MMFR and EMD2 as their claims would be sufficiently safeguarded. Applying additional own funds requirements still serves the purpose of investor protection and peg stability, however, it cannot be denied that it also serves as a further compartmentalization of EMTs within the crypto-asset market and a general disincentivizing the issuance of significant EMTs.

7. Conclusion

On the whole, the additional prudential requirements for EMTs adhere to the expectations of the FSB and BIS in respect of GSCs, in the sense that the strictly regulate the backing of EMTs in a prudent manner. MiCAR therefore surely succeeds in bringing legal certainty through the concept of EMTs. One would not be mistaken in stating that the requirements appear rather severe, especially when compared against the otherwise relatively timid liquidity requirements applicable to EMIs, or even MMFs for that matter. Especially significant EMTs are faced with a range of requirements that seemingly make the operation of an economically viable propositions impossible. There should be no mistake, significant EMTs (GSCs) pose a not to be misappraized financial stability risk, one that seems more likely to possibly materialize than those of electronic money. If the European Union had not acted, it may very well have seen a significant uptake of a new private digital money, issued by a tech-giant such as Meta. The financial stability concerns of international standard setters, the EBA, ECB and the Commission must therefore also be appreciated as being founded, even though no real GSC has developed quite yet. Nevertheless, the purported stablecoins that have been developed until now clearly flagged the need for investor protection and financial stability.

To that end, MiCAR comes with comprehensive new prudential measures that are part based on EMD2, part based on the MMFR and slightly inspired by the CRR. The reserve of assets is to be allocated in a set ratio which ought to ensure liquidity and thereby the redemption right of the investor. This reserve of asset allocation is further regulated by granular concentration limits and qualitative requirements, topped-off by overcollateralization and sumptuous own funds requirements. Whilst bolstering investor protection the severity and inflexibility of the prudential requirements applicable to EMTs must be seen as an effort to regulate financial stability through containment. The absence of any of the mechanisms that where identified for MMFs as enhancing resilience and the unnecessary penalization of financial sector holdings through CET1 deductions cannot be easily explained from a financial stability perspective strictu sensu. The current design of the own funds and overcollateralization requirements appear more as an instrument to deter EMIs from issuing significant EMTs, which may be challenged by EMT issuers only at great cost.

Although MiCAR does seem to want to prevent the disorderly failure of EMTs, it makes no big effort to prevent failure of the crypto-asset market in general. Apparently, none of the prudential measures taken by MiCAR solely seek to safeguard crypto-asset market stability, if such a concept even exists. Contrarily, regulation catered towards the traditional financial markets permits individual financial intermediary failure but financial stability is never permitted to falter. MiCAR thus apparently pursues its own unique goal of limiting a new economic sector in the interest of the ‘old’ financial system. If you see crypto-asset markets as a truly global, decentralized phenomenon then perhaps the internal market mandate on which MiCAR is based should here be seen as protecting the EU financial markets against such global market. It must be said that it is commendable that the legislator produced a regulatory act that seeks to curb financial stability concerns without having an experience of a real shock scenario in the real economy. Instead, the failures of crypto-asset intermediaries, certain crypto-asset linked banks in the US and the suggestion of a GSC issued by Meta were sufficient to make the European legislator realize it wanted to prevent such undertakings to do damage to the EU financial markets. Especially where these are already somewhat frail at the current moment, for example, the short-term money markets for commercial paper, MiCAR is particularly protective.

Concluding, this article does not mean to argue that MiCAR does not seek to achieve investor protection at all or that financial stability is only achieved by blocking an EMT sector from developing, nor has it been comprehensive in dealing with the prudential requirements applicable to significant EMT issuers. It merely posits that once certain elements of the (significant) EMT regime are examined, a separate and unusual regulatory objective to prevent the development of a significant EMT market can be discerned. In the end, some EMTs will almost surely be issued, though we believe it to be unlikely that many, if any, significant EMTs will develop. Perhaps MiCAR is similar to EMD1 in that regard, killing a market before it could ever realize the doom scenario’s prophesized, but we understand the European legislator’s reluctance to empirically test the veracity of the omnipresent warnings in relation to GSCs and wholly subscribe to it. In the end, the premise of MiCAR included in Recital (6) does ring somewhat hollow when considering the significant EMT regime: ‘Union legislative acts should avoid imposing an unnecessary and disproportionate regulatory burden on the use of technology, since the Union and the Member States seek to maintain competitiveness on a global market.’ Though, arguably the regulatory burden is, in respect of the highly threatening GSC concept, not disproportionate or unnecessary.

Maarten Mol-Huging is a lawyer at RegCounsel Financial Services in Amsterdam. If not already declared, please include details of any funding in a footnote at the beginning of the article, including funder name and grant/award number. (Ideally the Funder Name should be spelled exactly as it appears on the Open Funder Registry, a copy of which can be found (in several formats) at the following location: https://www.crossref.org/services/funder-registry/.)

Footnotes