In part 1 and part 2 of this blog we saw that the members of the Libra Association all keep a copy of the global database of Libra accounts and transactions and execute all Libra transactions on this database. They manage a Reserve of stable and liquid assets to support the value of a libra. Users can exchange their libras for fiat currencies against the value stored in the Reserve.
The Association will govern the Reserve once the Libra is launched. Until that time, planned in 2020, Facebook develops and manages the Libra network through its subsidiary Calibra. In this blog I zoom in on the role of Calibra.
Calibra is a subsididary of Facebook that operates independently. The Association has asked Calibra to manage day-to-day development of Libra Core until launch (Libra: The Path Forward, page 2). Discussion about issues on Github is moderated by Calibra. The technical documents on the Libra that I referred to (The Libra Blockchain, The Libra Reserve, Moving Toward Permissionless Consensus) have been written by Calibra and are formally offered to the Libra Association under a Creative Commons license.
Given the expertise that Calibra will have by the time of launch, I would not be surprised if the Association will ask Calibra to continue its central role in development and maintenance after launch.
Calibra and Facebook are both Founding Members of the Association. So they each have a weighted vote in the Association Council. This raises the question how this is consistent with the rule that there should not be a concentration of voting power in the hands of one party. Specifically, the Association says that
- “The Libra Association Council will prevent related entities from presenting themselves as two distinct Founding Members” (The Libra Association, page 4).
So how do we count Calibra and Facebook? Are they one entity or two, even if Calibra is a subsidiary of Facebook?
As we saw in part 2, there are other relationships among Founding Members as well, such as investor and advisor relationships. To avoid the appearance of collusion, these relationships should be published and the above rule should be clarified.
Calibra is developing a custodial wallet, which is a piece of software that holds your private Libra key and by which you can transfer and receive libras, and can query the Libra database. Custodial wallets are more user-friendly than non-custodial wallets, because they keep your private keys. The downside is that you have given Calibra control over your money.
Custodial wallets allow off-chain transactions among Calibra wallets. (“Off-chain” is crypto-jargon for a transaction that takes place directly between wallets, without being registered in the Libra database.) This is an important precondition for scalability (The blockchain paper, page 22).
Calibra’s wallet will be available as an independent app, and will also be integrated with Facebook Messenger and Whatsapp. This creates a payment ecosystem of Facebook users.
According to Calibra’s customer commitment, it has registered as a Money Service Business with the U.S. Department of the Treasury and is obtaining licenses in U.S. states that treat cryptocurrencies as the equivalent of money. In the EU it will operate an anti-money laundering program under the money laundering directives. Calibra announced that it will implement best practices in identifying customers to comply with Know-Your-Customer (KYC) requirements.
This raises the question how the 1 billion people without proof of identity would be able to buy libras. A stated goal of the Libra is to empower the unbanked. Half of the unbanked have no identifying documents required for opening a bank account.
Perhaps other Libra wallet providers, operating from countries with less strict regulators, will serve the unidentified by waiving KYC requirements? A user of the Calibra wallet, say a shop in a provincial town in Mozambique, can start their own financial services business for local customers, outside the view of regulators.
Calibra’s customer commitment (page 1) states that Calibra will share account information with Facebook or third parties, without asking for customer consent, when this is needed
- to provide basic functionality,
- to keep people safe, or
- to comply with the law.
It will also share customer data with vendors and service providers that support Calibra — including Facebook. Other than that, it will only share data with Facebook or third parties if the customer has consented to this.
These statements are very fuzzy. What is “basic functionality”? What information sharing is needed to keep people safe? To comply with which laws of which country?
The other way around, Facebook will share its customer information data with Calibra
- to secure customers’ accounts,
- to mitigate risk,
- to prevent criminal activity, and
- to comply with the law.
If a Calibra product feature can be personalized or improved with data from Facebook, it will obtain customers consent before it does this (Calibra’s customer commitment, page 2)
This is even vaguer than the above list of conditions. Which risks must be mitigated by sharing information? What information sharing is needed to prevent criminal activity?
These conditions leave a lot of loopholes for sharing data between Calibra and Facebook. Facebook’s track record in breaking promises does not inspire a lot of trust in its privacy commitments.
In a nutshell
Calibra will develop a custodial wallet that creates a payment ecosystem for Facebook, Messenger and Whatsapp users. The administration of the transactions is done by Facebook, along with at most 99 other companies. All these companies will maintain a copy of the database of Libra accounts and transactions and the database will be accessible to the public.
Calibra and Facebook will count as two Association members even though there are close ties between the two.
Calibra will take custody of your money and will register as a financial service provider. This defeats the purpose of “empowering the unbanked”: Calibra will have to request proof of identity of its customers, but 1 billion of the unbanked have no proof of identity.
Privacy guarantees given by Calibra are vague and very weak.
Calibra and Facebook will make big bucks as Association members and as validators for the Libra network. If the Libra is successful, both will be able to add financial services to their customers. We need to take a closer look at the Libra business model. In part 4 of this blog we will see that a business model of the Libra ecosystem shows that the system costs money to users and is a gold mine for investors.