Facebook Libra — a fundamentally well designed proposition
Libra is not based on proprietary technology or innovation
Perhaps the key understanding of the Libra initiative, though one that tends to get lost in the noise, is that Facebook et al. have not developed any new technology or solutions that would be unique or proprietary. The great leap is that Libra is proposing a fundamentally well designed, comprehensive system, which resolves most of the issues currently affecting decentralized applications, and Libra is of course is proposing deployment at an unprecedented scale.[i]
The Libra White Paper proposes a generally well thought out governance structure, use case, value proposition, business logic and financial engineering. Moreover, though it is not explicit in the Libra proposal, Facebook’s prowess in engineering useability and user engagement is unquestionable. Altogether this is a powerful combination.
There is generally nothing that would have prevented any major bank or other institution to have launched a similar initiative, apart from reasons that may be found in the mirror. In the same vein there is still nothing that prevents banks or other institutions from establishing similar competing initiatives, which does make it unlikely that Libra will remain the only private digital currency initiative for long.
A key observation from an investment perspective is that Facebook is voluntarily relinquishing the protocol layer to the Libra Association, while choosing to focus on commercializing the application layer. This is in direct contravention of the ‘fat protocols’ thesis which states that in decentralized systems economic value will predominantly be captured by the protocol, with applications taking a minor share.
Libra is a fundamentally well designed system
In our view, Libra has six core characteristics that are fundamental to the proposition, and which altogether have been missing of other decentralized B2C applications so far.
1. Transparent governance
Libra has a governance structure whereby it is clear to all, who is responsible and who makes decisions. How decisions are made, is less clear so far. The explicit governance of Libra is in particular contrast to the opaque, pseudo decentralized governance of permissionless networks, but it is an equal contrast to the highly centralized governance of platforms.[ii] These structural characteristics of governance should not be overlooked even if it is tempting to focus the discussion on the suitability of the governors.
2. Explicit use case
The initial proposed use case for Libra is the transfer of funds between nodes within a seamless transnational network. This itself is a valid, if somewhat limited use case. We assume that this will be followed by programmable transaction settlement (i.e. smart contracts) which will significantly expand the use-case spectrum.
3. Clear value proposition
The corollary of the use case is that there is also a value proposition to participants, which is a combination of convenience and cost. The value proposition is not diminished by the fact that (most) users may have existing alternatives to the use cases proposed.
4. A defined business logic
Libra Association and consortium members will benefit from financial income and fees from the distribution of Libra tokens, from lower transaction costs using Libra, from potentially increased transaction volumes, potentially from improved engagement with their customers and potentially from new sources of customer data. From a business perspective, this is a very strong proposition.
5. Underlying value basis to the currency
The financial basis of Libra is that the coins in circulation are fully backed by a fiat asset reserve. This creates an explicit value basis, which is transparent, understandable and workable. As the White Paper is explicit in that Libra coins are always redeemable against the underlying fiat assets, it is implicit (though not made explicit) that the value of the underlying assets should be regularly published.
6. A well designed UI/UX
We shall assume that Libra will offer a highly intuitive user experience. Facebook has a record of excelling in user engagement and Libra is likely to benefit from this.
The real-life analogy — Libra is an iteration of a money market fund
Conceptually Libra is equivalent to a money market fund, which is internationally diversified and is quoted in several currencies, and which has fund-units which are highly divisible, transferable in the secondary market without cost, and which are bearer securities (as opposed to registry entries). However, the investment return of the ‘fund’ is fully allocated as fees to the managers.
Competition in financial services is a good thing.
The current offering of financial services is not necessarily cheap, efficiently produced, easily accessible nor well governed. Introducing competition in this market should be socially, economically and regulatorily desirable. The reason incumbents decry new competition generally stems from defending their own profits, not from any general consideration for market good.
On the other hand, financial stability and market standards must be protected. The stability of the financial system is a key public good, and financial innovation should not be allowed to endanger it or lead to biased markets which can be abused.
To add perspective, though, much of the current ‘fintech’ industry is based on regulatory arbitrage where new service providers skirt regulatory oversight either by choice of jurisdictional domicile (e.g. Malta, Gibraltar, Latvia, Lithuania, etc. in Europe) or by leveraging narrow regulatory definitions that exclude their business. Libra, and others, should not be allowed to exploit the same. Implicit in the above view is that the regulatory framework of the financial industry as it stands is not perfect and leaves significant opportunities for abuse. An objective should be that there should not be multiple standards for companies providing essentially the same service.
Gone fishing and red herrings
The Libra proposal has already attracted a significant amount of critique from various parties. Without any attempt to be exhaustive we address some of the most often raised points below.
1. The network (essentially the governance) is not decentralized enough
At very large scale the topology of a decentralized network begins to resemble that of a distributed one. However, the governance of the decentralized one remains transparent, while the that of the distributed network becomes opaque.
There are currently some 30 institutional members of the Libra Association and the stated aim is to broaden this to some 100 members. There are few operational organizations or associations in the world today that handle the interests of this many parties and retain the ability to function. For all practical purposes, Libra is likely to become the largest decentralized network ever implemented.
2. The network is not permissionless
Generally all networks retain the right to exclude participants that do not fulfil certain standards (either implicitly or explicitly). This is generally done to secure compliance with the aims and norms of the network, as defined by its governance. It is a libertarian fantasy that permissionless networks exist without governance or that they exist in a social vacuum with respect to the broader social environment (e.g. national laws).
3. Facebook cannot be trusted. And Libra is a commercial enterprise
The Libra Association will be the issuer of the Libra tokens, the primary focus of regulation, and Facebook has a minor, non-controlling stake in the Association. Facebook trustworthiness should be a moot point in this instance. Facebook intends to build applications (e.g. the Calibra wallet) on top of the Libra protocol and Libra tokens; trustworthiness in this context is likely to be measured by customer acceptance and adoption of the applications.
The current financial system is run by banks and other institutions which are private companies. Alternative networks such as bitcoin or Ethereum are also dominated by commercial actors with explicitly commercial interests[iii].
4. There is no one reference currency
US based users may be astonished that Libra is not dollar referenced; European users may miss the euro; for emerging market users the key (positive) is that Libra is a hard currency. Again, the analogy to a fund with units that are traded in the secondary market is useful — in practice Libra distributors will quote daily fiat exchange rates which should offer sufficient clarity on worth. We have no doubt there will soon be competing initiatives that are referenced against single currencies.
5. Libra is unnecessary, there is an existing solution for everything Libra proposes
If Libra does not address a market use-case with a sustainable value proposition then it is likely to suffer poor adoption and die off. Ex-ante, we continue to maintain that competition in markets is generally positive.
6. Banking for the unbanked is not a relevant use case
Facebook has unparalleled global reach with 2.4bn registered users across its properties. There is no other network or service that can match Facebook’s penetration in any number of national markets. This alone provides backing to the claims of universality. Nevertheless, we agree that claiming to launch Libra primarily to serve the unbanked is somewhat disingenuous and we would be surprised if the bulk of the application/feature development would not be driven by the (first-world) sources of value. Nevertheless, this is an example where technology may allow the leap-frogged introduction of first-world services to third-world contexts, with potentially powerful impact.
7. Libra may cause monetary policy disturbances due to being beyond central-bank control
There already are multiple money market funds in operation with assets in excess of EUR100bn. Libra may never reach this magnitude, given that the explicit absence of investment returns is likely to preclude its general use as an investment product.
8. Libra may disrupt the currency reserves of weaker economies
In many emerging economies remittances form a major part of national income, and specifically of foreign currency income[iv]. In these economies, if Libra is able to increase the value of received remittances by up to 10% (due to lower costs)[v], it may lead to a significant boost to macro-economic development. The people relying on remittances are generally those with the least savings, and the highest proportion of consumption to income. It is likely that they will be just as driven to convert Libra remittances to local currency, as they have been to convert current dollar remittances.
9. Libra will resolve inflation protection in high-inflation economies
There is a popular myth, that crypto currencies are the answer to shield individuals from the effects of high-inflation. This is a myth, because in these economies access to crypto or digital currencies is no simpler than is access to dollars or other hard currency. The assertion is only applicable in cases where an individual has access to hard-currency but may need to hide the origin (and destination) of the funds and in cases where a party has access to energy at scale, and may be able to locally convert this through POW mining into digital currency.
10. The data structure is not a blockchain
The single most highlighted problem of permissionless networks continues to be scalability. The Libra design appears to resolve the scalability problem, while maintaining other core ‘blockchain’ characteristics.
The Libra data structure embodies key elements such as a shared ledger with state, programmable resources that are owned by independent accounts, accounts that are owned by key-pairs, a decentralized BFT consensus protocol, an immutable transaction history[vi], a Merkle tree authenticator structure and native tokens and native execution economics. The underlying data structure, though, is not a sequence of blocks but a single data base with a sequence of states. For most network participants this distinction is likely to be simply irrelevant.
Uncertainties, unknowns and unresolved issues
Libra may be the best thought out large scale digital currency project to date, but it also involves a significant number of uncertainties and unresolved issues — which need to be resolved before it reaches market.
Regulation of Libra?
The first of the unresolved questions is that, from a legal perspective, what is it? Is Libra a security or not, and if it is not a security then what is it and how will it be regulated and taxed?
Under the Swiss FINMA framework that has been chosen as the preferred jurisdiction, the structure seems explicitly designed to have the Libra investment token capture the financial properties that make it an “Asset token” while seeking to guise the retail Libra coin as a “Payment token”. Asset tokens will be regarded as securities by FINMA and regulated under swiss securities law, whereas the Payment tokens “will require compliance with anti-money laundering regulations. FINMA will not, however, treat such tokens as securities”[vii].
Not being treated as a security essentially frees Libra from all restrictions around the marketing and distribution of the coins to retail customers.
If the Libra succeeds in obtaining Payment token status for the broadly circulating coins, then the substantive question is that is the regulatory regime sufficient — should there be a (mostly) unregulated, globally circulating digital currency even if it meets a narrow Swiss definition of a Payment token?
Again, conceptually the Libra coin looks very much like a highly tradeable, fungible, fund unit that has been converted into bearer-owner form. Perhaps the regulatory question should be that how would and how should such a unit be regulated?
Regulation of Libra in Switzerland will undoubtedly raise questions around the interpretation of financial sector equivalence agreements (currently in any case suspended between Switzerland and the EU). The question is not only academic for national regulators; the very potential ubiquity of the user base is likely to provoke significant debate in all jurisdictions where Libra is marketed or made available to retail customers. We would not be surprised to see national or EU regulators reserving the right to rule locally on Libra’s security (or not) status. Needless to say this will have a huge impact on how it may be marketed and used in differing jurisdictions., not to mention taxation.
It will be far more difficult for national regulators to altogether ban Libra use — simply due to Facebook’s ubiquity and the likely deep integration of Libra into the Facebook ecosystem.
Regulation of the Libra distributors?
According to the Libra Association White Paper, the members of the Association appear to have a protected status as sole distributors of Libra coins. The Association mints new coins in exchange for fiat deposited in reserve by the members, which may then further sell these to retail customers. As currently proposed, distribution would not appear to be a regulated activity.
Regulation of Libra trading?
FINMA does not require regulation of the secondary trading of Payment tokens so it would appear that the trading of Libra coins would not come under any oversight. However, if Libra were to be deemed a security then secondary trading of the coins would be a regulated activity. And, the issuer would likely be required to maintain an up-to-date register of token holders.
Why would Libra be a security?
Under the FINMA guidelines ”Payment tokens (synonymous with cryptocurrencies) are tokens which are intended to be used, now or in the future, as a means of payment for acquiring goods or services or as a means of money or value transfer. Cryptocurrencies give rise to no claims on their issuer.” In contrast “Asset tokens represent assets such as participations in real physical underlyings, companies, or earnings streams, or an entitlement to dividends or interest payments. In terms of their economic function, the tokens are analogous to equities, bonds or derivatives.”
The narrow Libra view is that the coin only has value as means of payment but has no other financial characteristics. The fact that investment returns are stripped out of the Libra coin would support the notion that it is only a Payment token. On the other hand, the Libra coin is explicitly tied to a physical underlying, the basket of fiat investments which underpins the value. Furthermore, certain Libra holders would have an explicit right to redeem Libra coins against the underlying assets on an on-demand basis.
Separation of investment returns from capital?
Conceptually we understand that Libra coins are not entitled to dividends, interest payments or other investment returns (these accrue to the Association members through another instrument). Practically we do not understand how this distinction can be made. Short term debt instruments do not generally carry separate coupons; it is generally impossible to exclude or separate capital returns from interest income; and, in a zero-interest rate environment it will likely be irresistible for Libra to engage in some maturity transformation.
So, if the daily value of Libra is calculated from the value of the underlying asset-values, the asset values are a function of accrued interest and capital appreciation/depreciation; there is no single reference currency (and all the reference currencies fluctuate against each other), how will Libra separate out the ‘asset value’ that belongs to Libra payment coin holders from the ‘investment return’ that belongs to the Libra investment token holders?
AML, KYC and identity?
Under the FINMA guidelines, even if Libra is not treated as a security, “FINMA will require compliance with anti-money laundering regulations”, the heart of which is the requirement to establish the identity of the beneficial owner. Furthermore, the financial intermediaries handling the sales (Libra Association Members?) may be likely to be required to be directly licensed and regulated by FINMA, though only for the KYC activity it would seem.
But, what is to happen beyond this? What would the requirements be for opening a Calibra wallet (would a Facebook account be sufficient??), what would the KYC requirements be for a wallet, and what if any restrictions would there be on acquiring Libra in the secondary market? If Libra secondary trading is unregulated, where does the AML responsibility of the Libra Association members end?
The white paper explicitly states that “Libra is pseudonymous and allows users to hold one or more addresses that are not linked to their real-world identity”. How the pseudonymousness can be made compatible with AML requirements would seem an unresolved contradiction.
An ultimate stated aim of Libra, according the White Paper, is to develop an open, de-centralized and portable digital identity standard. The details of this aspect of the plan remain totally opaque, but one cannot avoid surmising that a Libra based identity standard is somehow linked to resolving Libra pseudonyms and KYC requirements and satisfying (at least the letter) of AML requirements.
Regulating data collection?
Current payment processors (Visa, MC, et al.) already gather a huge amount of data, which the companies themselves use and also sell to third parties for commercial purposes. Much of this data may furthermore be linked with other data sets to create personally identifiable data records. It is a false assumption that these parties are inherently more benign than Facebook et al. However, it is true that FB is potentially in a unique position to make commercial use of the data it is able to collect, due to the extent of the other data it already collects on its users, due to its control of the Facebook ecosystem and due to its ability to persuade users to ‘voluntarily’ opt-in to personal data use.
This does not seem to us to be a reason to restrict Libra but to regulate platform company data collection and data use more generally. This is not a new debate, but the deep integration of financial services into platform functionality does introduce a potentially novel angle. In our view, the data use question is one that can hardly be avoided in any discussion of Libra.
Censorship resistance and judicial certainty
Ultimately the libertarian critique of the Libra design revolves around a single theme — is Libra censorship resistant or not? “Censorship resistance” in practice generally translates to immunity from regulatory oversight or intervention. Thus, the same question conversely is that will Libra be overseen and regulated, or not?
So far, commentary from the Librarians would seem to attempt to answer both questions concurrently in the affirmative. Unfortunately, the answers are mutually exclusive. This does give the appearance of a determined attempt to proceed with implementation undercover of wilful obfuscation. Perhaps a new iteration of the “move fast and break things” principle?
From the perspective of users coming to use Libra as a payment facility the same question is likely to translate into: “where are disputes about transactions settled, by who and on what basis?” Will Libra customer service be the adjudicator under Libra terms and conditions? Will the Association members form an arbitration panel, and if so will it be under which law?[viii]
Or, perhaps this issue, along with a host of others we have not yet identified, will simply be ignored as an externality until national regulators take action and compel a view, which could be many, many years from now.
[i] The Libra White Paper https://libra.org/en-US/wp-content/uploads/sites/23/2019/06/WhitePaperAndSupportingDocuments-2.zip .
[ii] Governance of permissionless decentralized networks generally suffer from severe contradictions between premise and reality; four typical authorizing strategies are conflating people with devices, appealing to technical expertise, assuming actors conform to notions of economic rationality and explaining contradictions as temporary bugs. These strategies are mobilized widely to legitimize a variety of applications of algorithmic regulation and peer production projects. Vidan, G. and Lehdonvirta, V. (2018) “Mine the gap: Bitcoin and the maintenance of trustlessness”, New Media and Society. 21 (1)
[iii] The commercial role and motivation of miners, exchanges, hackers, etc. is generally skimmed over in ‘crypto-folklore’, and it is arguable whether or not the motivation (and induction) of core developers is exclusively non-commercial.
[iv] The World Bank lists 30 economies where personal remittances form 10% or more of GDP. https://data.worldbank.org/indicator/BX.TRF.PWKR.DT.GD.ZS
[v] The global average commission for remitting money is 7%. Source: World Bank, Generation Investment Management.
[vi] ” As part of committing a transaction Ti at version i, the consensus protocol outputs a signature on the full state of the database at version i — including its entire history — to authenticate responses to queries from clients.”
[vii] https://www.finma.ch/en/news/2018/02/20180216-mm-ico-wegleitung/
[viii] Systems such as Libra are useful in demonstrating how popular tech catch phrases such as “code-is-law” have no applicability when it comes to real-world implementations.
