Stabilising stablecoins: A pragmatic regulatory approach
This column is a lead commentary in the VoxEU Debate “The Future of Digital Money“
Policymakers are extremely concerned about the risks that stablecoins pose. They want to prevent stablecoins from influencing open market operations, affecting monetary bases, weakening capital controls, draining deposits and increasing systemic risk (Cœuré 2019). Public attention on the topic spiked after Facebook announced its intention to launch its own stablecoin. So far, regulators have been extremely cautious, and preliminary talks have been complicated and lengthy.
We approach the issue from a different angle, to try to accommodate business needs while addressing the concerns of regulators (Somoza and Terracciano 2019). Specifically, we propose a regulatory framework that would bind stablecoins in an ETF-like structure, benefitting from its stabilising properties (Cecchetti and Schoenholtz 2016).
Stablecoin structures and the case of Libra
So far, stablecoin providers have used many different structures. There is no industry standard and a lack of transparency, so it is often hard to understand risks for investors.
The experience of Tether Limited, issuer of a cryptocurrency of the same name, shows the problems that stablecoins can cause and emphasises the importance of a clear regulatory framework. Tether claimed that it stored safe financial assets for each token it issued. In the first quarter of 2019, it updated its website, stating that loans to Tether affiliates and third parties were included in its reserve.1 Clearly, this was problematic. It changed the risk profile of Tether tokens, transforming them into a hybrid instrument of debt. This created a conflict of interest, and implied that Tether might need a banking license to operate. Various lawsuits, including a class action, have been filed against Tether.
In contrast, Libra, the cryptocurrency planned by Facebook, has a structure similar to an Exchange-traded fund that should help it remain stable. Libra’s white paper (2019) explains the economic mechanism behind the stablecoin (Figure 1).
Figure 1 Comparison between a standard ETF structure and Libra
Source: Somoza and Terracciano (2019), based on Libra Association Members (2019).
The Libra Association defines a basket of assets that can be traded for Libra tokens by selected intermediaries, called authorised resellers (ARs). ARs purchase these assets on the market and give them to the Libra Association. Assets are transferred to third-party custodians and Libra issues new tokens, which are delivered to the ARs. Retailers can purchase Libra tokens only through ARs. On the contrary, when the demand for Libra decreases, ARs can redeem the tokens in exchange for the basket’s assets.
The ARs operate in an arbitrage corridor: when the price of the token is higher than the basket price, they create new tokens – and vice versa when the price of the token is lower than that of the basket. This arbitrage opportunity guarantees that the token’s price remains close to the basket’s price.
This mechanism feels like dejà vu for anyone familiar with ETFs. The creation-redemption mechanism is the same and ARs, as defined by Libra, operate like authorised participants in the ETF market.
The key innovation of Libra is its secondary market and concerns its trading infrastructure, which in turn allows for its tokenisation. Retailers can exchange Libra tokens through a special kind of permissioned blockchain, so they can to be used as private currency in commercial transactions and remittances.
The comparison has limits. First, ETFs are independent legal entities, separate from the ETF sponsor, while Libra’s legal nature is not yet defined. Second, Libra’s fee structure differs significantly from ETF industry standards. These aspects are certainly important, but they do not affect the economic substance of our argument.
Stablecoins as a new regulated asset class
We propose to regulate stablecoins under a new ETF-like asset class. If it looks like a duck, swims like a duck, and quacks like a duck, then it probably is a duck, and our argument that stablecoins function with an ETF-like structure is the key point of our proposal. ETFs and Libra share the same economic structure, and this should be the baseline for regulatory scrutiny because similar risks should be regulated in the same way (FINMA 2019).
This regulation would bind stablecoins to an ETF-like structure, with full backing, token convertibility, a primary market with authorised participants, third-party custodians, and full disclosure of the underlying basket. Stablecoin sponsors should also be required to collaborate with the regulator to with the regulator to develop orderly liquidation plans. Regulation would turn stablecoin sponsors into a specific kind of ETF sponsor, reducing uncertainty around their business model.
Further regulation is needed to allow tokens to be traded freely and easily by retailers. In particular, it would be reasonable to constrain stablecoins to include only risk-free securities and deposits in their basket, avoiding excessive financial risks for the financially unsophisticated retailers who might use them as a means of payment.
For most digital currencies, users possess their own keys. This allows them to store their tokens physically (for example, on a USB stick) and to move them anonymously. This clearly poses huge capital flight risks, or the danger they can be used to support criminal activity. Therefore, we propose that users who want to dispose of their tokens always pass through a certified institution (maybe using a mobile application)
Finally, when the regulator expects the launch of a stablecoin to have a large impact (Libra is a good example here), it could impose a step-by-step introduction. The sponsor and the regulator would preventively agree on temporarily constraints on the basket, such as picking shorter maturities or imposing a higher share of deposits. Something similar already happens in the ETF industry.
Current regulatory concerns
A new asset class would significantly like this would help calm nervous policymakers. ETF-like stablecoins would not create new monetary base and would not have a significant impact on open market operations unless they became pervasive.
It also helps allay concerns about currency substitution, as any transaction in stablecoins would effectively be a transaction in its underlying currencies. For instance, if Swedes converted all their deposits into Libra, the ‘libralisation’ of Sweden would be a de facto dollarisation or euroisation, assuming that these were the main currencies of the basket. But note that Swedes already have access to these currencies, and neither dollarisation nor euroisation has yet taken place.
We might argue that this (potentially) superior technology might trigger this change. But it is worth noting that Revolut, N26, Transferwire, and many other companies already allow users to make instantaneous and frictionless cross-border transactions, in different currencies, at almost no cost. Apart from network externalities, it is hard to imagine that stablecoin platforms would be so superior as to induce a massive currency substitution.
Similarly, liquidity would not dry up if a large fraction of bank deposits were converted into stablecoins. Those stablecoins would have to hold deposits and highly liquid assets in their basket, and those assets would be under the custody of other financial institutions. There would be similar effects if depositors decided to move their accounts to another bank, or to buy risk-free securities. So there might be a different allocation of liquidity in the system, but liquidity would not dry up.2
The Swedish scenario is unlikely because it is reasonable to assume that people in developed countries would mostly use stablecoins for online transactions. They would not all suddenly redenominate their deposits and contracts because, if they did, they would become exposed to exchange-rate risk.
This implies that stablecoins would not crowd out traditional fiat currencies, though they would allow retailers to use them through different vehicles and, as with Libra, to use more than one at once. This might affect the shares of invoicing currencies in the global trade, even though the magnitude of such change would probably be limited, at least in the short term.
Framing stablecoins as a new kind of ETFs would also reduce the risk of disintermediation, because their legal structure would prevent them from crowding out traditional banks. The system may become more payment-centric (Brunnermeier et al. 2019), but banks would still have a direct interaction with retail customers because they would be able to supply services that stablecoin providers would not have the license to offer, such as mortgages.
Also, the risk of run and insolvency would be negligible. The ETF structure means that runs could hardly occur (if at all), and potential liquidations do not generate any significant problem because all shares would be fully backed with liquid assets and deposits.
Finally, users could access their wallets only by identifying themselves, which eliminates worries about anonymity. This would allow governments to prevent criminal activity and control capital flight, tax evasion, and money laundering. The current legal requirements for ETFs in this area would apply also to stablecoins.
Same risks, same rules
We argue that the best way to deal with the unknowns of stablecoins would be agnostically to evaluate the underlying mechanisms,and map them into the current regulatory framework. The similarities between the Libra and ETF structures suggest that a stablecoin could be interpreted as a special class of ETF, rather than an actual currency.
This approach does not take a stance on stablecoin business models per se. Rather, it follows the regulatory principle of ‘same risks, same rules’ to create a framework that allays the fears of regulators while accommodating their core features. This framework would reduce the uncertainty around stablecoins, limit systemic risk, and allow innovation to flourish.
Brunnermeier, M K, James, H and Landau, (2019), “The Digitalization of Money”, working paper.
Cecchetti, S and K L Schoenholtz (2016), “Reforming mutual funds: A proposal to improve financial market resilience,” VoxEU.org, 15 November.
Cœuré, B (2019), “Update from the Chair of the G7 working group on stablecoins”, 18 July.
FINMA (2019), “Supplement to the guidelines for enquiries regarding the regulatory framework for initial coin offerings (ICOs)”, Swiss Financial Markets Authority, 11 September.
Libra Association Members (2019), “An Introduction to Libra”, white paper.
Somoza, L and T Terracciano, (2019), “Stabilizing stablecoins: a pragmatic regulatory approach”, working paper.
 This is explained on Tether’s home page:
“100% Backed – Every tether is always 100% backed by our reserves, which include traditional currency and cash equivalents and, from time to time, may include other assets and receivables from loans made by Tether to third parties, which may include affiliated entities (collectively, “reserves”). Every tether is also 1-to-1 pegged to the dollar, so 1 USD₮ is always valued by Tether at 1 USD.”
 If the size of a stablecoin became worryingly large, regulators could force it to hold only deposits, alleviating liquidity concerns.