Libra Shrugged — Chapter 15: Central bank digital currencies

Excerpt from Libra Shrugged: How Facebook Tried to Take Over the Money by David Gerard

Update, November 2021: Here’s a book excerpt a year after its publication — because I keep referring people to this chapter. Buy the book for more!

This chapter has stood up quite well through 2021. Sand Dollar, D-Cash and DC/EP (now e-CNY) are still in ever-growing pilot programmes. Nigeria has just launched its eNaira CBDC as an alternate payment system.

Consumer use cases for CBDCs are still thin on the ground — basically when commercial banks aren’t serving the public sufficiently, and the central bank has to step in. Even then, they partner with the commercial banks as absolutely far as possible.

He had his cash money, but you couldn’t pay for food with that. It wasn’t actually illegal to have the stuff, it was just that nobody ever did anything legitimate with it.

— William Gibson, Count Zero

Central banks issue physical cash — but what if they could print official legal tender digitally? The real world result from Libra may turn out to be “central bank digital currency,” or “CBDC.”

A central bank’s job is to keep its economy on an even keel. Central bank researchers look at a new idea and think: “what would happen if this weird thing got popular?” The structure of money can change surprisingly fast.

So the researchers routinely come up with wild plans to rebuild the financial system, but upside-down and inside-out — just in case they suddenly need them.

Bankers noticed Bitcoin and other cryptocurrencies early. When people talk about a new form of “money,” and there are news headlines about the price of Bitcoin, central bank research papers will follow.

Then Libra threatened to put the idea of a blockchain-like currency into practice, at massive scale — but badly. Central bank researchers quickly dusted off their old papers, with an eye to heading off disaster. And if there was anything to the Libra idea, maybe it could be implemented with an eye to caution.

Central banks have issued electronic money before — but current discussion of “CBDC” involves a lot of misapplied blockchain hype. End users don’t need to worry too much about this particular back-end plumbing.

What is a central bank digital currency?

E-money (electronic money) is money for consumers that’s transmitted electronically outside the banking system itself. PayPal is a well known example. A central bank digital currency is e-money issued directly by the central bank — it’s regulated differently, but the end user shouldn’t see much difference.

Central banks had been thinking about what central-bank-issued e-money might mean since the 1990s256 — but the current round of CBDC discussion was largely inspired by Bitcoin and blockchains.

CBDC on a blockchain was first suggested in 2013 by economics blogger J. P. Koning as “Fedcoin.”257 Koning proposed that the US Federal Reserve run a widely-distributed blockchain-based database, and create and destroy dollar tokens on the chain as needed. This could also be a consumer payment system, running on your phone.258 259

Central banks wrote papers on what a popular retail CBDC might mean, and what the economic risks might be.

For a long time, all of this was just an academic exercise — with a fair bit of blockchain hype. But as CBDC proposals get closer to reality, they tend to have less and less to do with blockchains.

Some proposals are for wholesale CBDC — between the central banks and the commercial banks — though it’s not yet clear what this gets anyone over the banks just sending numbers between known and trusted computers, as we’ve done for decades. Banks have experimented with blockchains, but the results are, at best, slightly worse, slower and less sophisticated versions of existing systems. Wholesale CBDC is mostly blockchain hype, and mainly of interest to the sort of crypto news site that thinks CBDCs are good news for Bitcoin.

Responding to Libra with CBDC

Libra frightened the central banks — a popular private currency run by people who didn’t seem to know what they were doing could be disastrous.

Central banks started looking into CBDC with more urgency. There might be a gap in the market, mainly for low-cost international settlement — and Facebook said Libra could fill that need. But Libra would run at a large enough scale to risk financial stability.

So central banks would need to do a coin themselves — one that was responsible to the public, rather than to a private company or consortium. The People’s Bank of China finally kicked its CBDC programme into high gear in direct response to Libra. (Though they were also reacting to US dollar hegemony and the size of the local payment providers.)

It wasn’t clear that Libra could in fact do what it claimed — nor was it clear that a central bank version would do the job either.

Why do a CBDC?

If you live in Europe or China, you probably have good electronic payment systems — within your country, at least. So you know that good payment systems are possible without adding a blockchain.

Other countries are different — because local conditions are important.

  • The US is an advanced economy — but the payment system is old, creaky and complicated, and payment companies have to register federally and in fifty states.
  • Many countries still run on physical cash, cheques and so on. But better payment systems speed up the flow of money, and get the economy moving faster.
  • Even in Europe, the systems inside a country will be fast — but payments between countries may still be slow or expensive.
  • Sweden has the opposite problem. Cash is used less and less — but some sort of cash is considered to be socially necessary, and the Riksbank, Sweden’s central bank, worries about how to manage an economy with almost no central bank cash in it.

Maybe CBDCs will fix all of these problems! Somehow. Libra also promised to fix all of these problems — somehow.

Watching every move you make

Physical cash doesn’t have Know Your Customer (KYC) — your transactions are private (unless they’re large). But one thing that the anti-money-laundering (AML) agencies have required of digital cash so far is KYC. So they would want a CBDC to have KYC.

The idea of all transactions being on a blockchain also appeals to the anti-money-laundering agencies — imagine having a complete ledger of your entire economy. No criminal transaction could escape.

Bureaucrats also keep being seduced by the prospect of a full Eye of Sauron panopticon of every transaction in the economy. This appeals to those who think their problem is not having enough knowledge to exert control as finely as they’d like to.

This is otherwise known as the Big Data Fallacy — where you think that just getting enough data will surely solve all your problems. It’s a fallacy because your problems are usually political — you know perfectly well what you need to do, and you’re hoping the big data will help convince others to let you do it.

Nevertheless, bureaucrats and anti-money-laundering agencies find the prospect of a CBDC with a full record of everything hard to resist.

Libra would record less data centrally — a lot of transactions would happen internally to Facebook’s Calibra/Novi, and not even make it outside of that company to be recorded on the main blockchain.

Can CBDCs give us better payment systems?

A common fallacy of CBDC proposals is thinking that changing the back-end is all you need to make a better payment system.

CBDC proposals tend to have a disconcerting lack of detail about what magic a CBDC will bring that the commercial banks couldn’t do better — and that commercial banks haven’t done better in other countries. A lot of CBDC plans have been vague, and don’t offer much that’s new.

Facebook claims Libra can give us better payment and settlement systems. Facebook’s method appears to be to ignore regulations. Central banks can’t do that.

Agustín Carstens of BIS acknowledged in June 2019: “There needs to be evidence for demand for central bank digital currencies and it is not clear that the demand is there yet. Perhaps people can do what they want by using electronic wallets provided by banks or fintech companies. It depends on the development of payment systems.”260

Retail CBDCs are only useful if they let you do something that wouldn’t be possible without them. But current payment systems are pretty good, and mostly just need to be put into place in infrastructure-poor economies.

The more solid CBDC plans aim to get the commercial banks to build better inter-bank payment systems. A central bank may even build a CBDC system to kickstart better retail payment systems.

Real-world retail CBDCs

Central banks have issued electronic money for broad public use twice before: Avant in Finland, and Sistema de Dinero Electrónico in Ecuador. Both failed — Avant wasn’t as convenient as debit cards, and Ecuador’s central bank just wasn’t trusted by the populace.

Uruguay ran a very limited CBDC proof-of-concept over 2017 and 2018, which involved a small number of ordinary end users.

China’s DC/EP proposal — which isn’t live yet — will be accepted by end users to the degree it works as well as Tenpay/WeChat Pay or Alipay.

New systems in pilot tests with real end users include the Bahamas Sand Dollar and the Eastern Caribbean DCash.

These are very different from each other, and show the many success and failure modes of projects labeled “CBDC.”

Finland: Avant (1993)

The very first CBDC was Avant-Kortti (“Avant Card”), which was launched by the Bank of Finland in 1993.261

Avant was a stored value smart card — you bought a card with money on it that you could then spend. From 1994, you could top up an Avant card with more money. The encrypted smart card was far more secure than the magnetic stripe debit and credit cards of the time. So this was e-money, but backed by the central bank.

Cards were anonymous and individual transactions weren’t tracked, so money on Avant would work like cash. The card was labeled an “electronic purse” — the Bank of Finland hoped Avant would replace small change. Though users couldn’t exchange Avant money between themselves directly — this was possible in the design, but that function was not enabled.

Avant fit most modern definitions of CBDC — it was in the national currency, it was entirely electronic, the money was a claim against the central bank, it was a widely accepted payment instrument, and it didn’t need to be continuously online to a central authority. The main difference to modern schemes was that Avant didn’t start from blockchain ideas.

The money in Avant was not legal tender — despite being issued by the central bank — so that merchants would not be legally required to accept it in payment of debts, which would have required them to buy expensive terminals.

In 1995, the Bank of Finland sold Avant’s operator, Toimiraha, to a consortium of commercial banks. This made no difference to the users — nothing changed about the system.

Avant was one of the first examples of regulated e-money, and informed the European Union’s E-money Directive of 2000.

Avant was moderately popular — at its peak there were 900,000 cards in active use, in a country of five million people.

The biggest problem was that users were unhappy at being charged a fee for loading and unloading money, especially as ATM withdrawals were free by this time. Users hated fees more than they liked anonymity. Avant was discontinued at the end of March 2006.

Ecuador: Sistema de Dinero Electrónico (2014)

The Sistema de Dinero Electrónico (SDE; “electronic money system”), run by the Banco Central del Ecuador (BCE), was a bit odd. For one thing, the currency was the US dollar. For another, the BCE didn’t quite have the powers you’d expect of something called a “central bank.”

SDE wasn’t quite a CBDC — in practical terms, it was really a failed payment system.

Ecuador went through a hyperinflation in the late 1990s. The national currency, the sucre, went down in value rapidly, and the economy was already substantially running on cash in US dollars — so the government chose dollarisation, and formally declared the US dollar the national currency of Ecuador in January 2000.

This left the central bank unable to issue its national currency, act as the lender of last resort, or manage monetary policy in any meaningful way. So the BCE’s focus shifted to guidelines and policy. It was also the regulator for payments.

In 2014, Ecuador’s Congress approved a fully-backed electronic dollar — so this was technically the first CBDC US dollar, sort of! The system was launched as SDE.

Each dollar in SDE was backed by a dollar stored at the central bank. BCE was the only legal issuer of e-money — cryptocurrency was also disallowed — and the phone company, CNT, was the only legal distribution channel.

You could sign up with an Ecuadorian identity card and a mobile phone. This gave you an account at the BCE. You could go to an authorised outlet and deposit cash into the account.

The stated reason for SDE was to improve access to the banking system for poorer residents, and reduce the reliance of the economy on physical cash, to get the economy moving.

Users could exchange dollars on SDE between themselves, just like cash. The poor ran on a cash economy, a lot of which was an informal economy, and the central bank hoped that SDE would bring them into the financial system.

Both local commentators and foreign debtors worried that Ecuador would create unbacked dollars in SDE to fund government spending and pay off the country’s debts — particularly given that the government had defaulted on bond issues just a few years before.262 But Ecuador denied any such plans, and didn’t end up doing anything along these lines with the dollars in SDE.263

SDE had a limited launch in late 2014 and a full public launch in 2015. The BCE expected 500,000 users by the end of 2015.

SDE failed to take off — the system only had 5,000 users by the end of 2015. By the end of 2017, SDE had moved $62 million via 5.1 million transactions in its entire existence264 — with only $11 million on deposit, as compared to the $24.5 billion in cash in the Ecuadorian economy. There were 402,515 e-money accounts — but less than 30% of these were ever used, at all.265

Ecuadorians with money trusted private banks more than they trusted the central bank — because they remembered the ways the government had abused its financial powers in recent years.266

Nor was SDE taken up by the unbanked or underbanked.267 Even in the rural areas — “Aquí pagamos con la palabra y el dinero contante y sonante” (“here, we pay with our word, and hard cash”) — people trusted their local co-op bank, but they didn’t trust the central bank.

The central bank just wasn’t good at serving retail customers, either — even if it had had the full trust of the users.

The government eventually acknowledged SDE wasn’t going to do the job. The commercial and cooperative banks put together a new mobile payment system, Bimo, with BCE as the regulator. SDE was shut down in March 2018, and accounts could be migrated to Bimo.268

The main lesson of SDE is: don’t try to back your system with a bank that your users don’t trust with their money.

Uruguay: e-Peso (2017)

Banco Central del Uruguay (BCU) issued an e-Peso in a public pilot programme that ran from November 2017 to April 2018. The pilot was part of an official push for financial inclusion.269

BCU issued 20 million e-pesos. Commercial banks were not involved in the pilot — though once it started, some banks approached BCU asking if they could join in. Digital wallets were operated centrally by Antel, the national phone company.

The 10,000 end users could hold a maximum of 30,000 e-pesos in their wallet, and registered participating businesses could hold 200,000 per wallet. Users could transfer money to each other — if they had phone coverage. Transaction data was encrypted, but could be decrypted if legally required.

To encourage participation, the first thousand users got a 500 e-peso credit, and there were monthly lotteries for the most active users and businesses.

The pilot finished in April 2018. The e-pesos were cashed out and destroyed.

BCU considered the pilot a success — the system basically worked on a small scale, and commercial banks wanting to join suggested that such a system could be popular. BCU is still considering whether to do a full-scale public e-Peso — the technical bit works, and now they’re going through the economic risk assessments.

Bahamas: Sand Dollar (2020)

The Central Bank of the Bahamas (CBB) wanted to encourage an electronic payment system that reached the outer islands of the Bahamas — the cash infrastructure was shrinking as commercial banks closed branches, but phone coverage was surprisingly good, if patchy.

Electronic payment systems for retail were usually built by commercial banks. But in this case, the central bank figured it needed to push things forward.270

The CBB started with a list of what they needed in an electronic payment system, and worked out the system from there — something to handle spotty network coverage and lots of poor prospective users, but almost all of whom had phones.

Most users connected via their bank account — the CBB left retail to the retail banks. Users who didn’t have bank accounts could get a Sand Dollar wallet with their name, address, phone number and a photo, which was enough to open a very limited account at a retail money changer.

The Sand Dollar wallet app on your phone could transmit small amounts of money when you were offline.

The Sand Dollar pilot in 2020 was small — just $48,000 in circulation in the system271 — but it went well, and the system was set to go live across the Bahamas in late 2020.

Eastern Caribbean: DCash (2020)

DCash, from theEastern Caribbean Central Bank (ECCB), started its pilot programme in late 2020 — with real consumers and merchants using it for real transactions.272

DCash had similar local conditions to the Bahamas Sand Dollar initiative. However, the system had no offline capability at the pilot stage — you had to be online.

Most DCash end users worked through their banks, with limited accounts available for unbanked users.

DCash had a blockchain in the system, though it was just an instance of IBM’s Hyperledger running as a back-end data store, and not doing anything you couldn’t do just as well without a blockchain.

China: DC/EP (in testing)

Digital Cash/Electronic Payments, or DC/EP, is a payment system project by the People’s Bank of China (PBOC), carrying central-bank-backed digital renminbi (yuan).

PBOC started looking into blockchains in 2014, and officially established the Digital Currency Research Institute in 2017, which had looked into CBDCs. In the wake of Facebook’s June 2019 announcement of Libra, Wang Xin of the PBOC said on 8 July 2019 that they were stepping up their CBDC project,144 to turn it into something that would stand up to public use at scale.

DC/EP has the following design goals:

  • Two-tier design. PBOC issues the money, the commercial banks distribute it — the central bank is not going to try to do retail itself.
  • Interoperability. You can’t send payments between the big private money transmitters, Tenpay/WeChat Pay and Alipay — but all DC/EP providers will be interoperable.
  • Financial inclusion. PBOC gave no specific plan to achieve financial inclusion, though there were aspirations to reach remote and rural populations.
  • Traceability of all transactions. WeChat Pay and Alipay don’t give the central bank full logs of every transaction. Some in the PBOC are seduced by the prospect of a full record of the entire Chinese economy.
  • “Controllable” anonymity. You will be anonymous to the retailer — but the central bank still gets a full feed of who’s doing what.
  • Working offline. You should be able to exchange money without being online 24/7. Transactions will be sent in when you’re back online.

PBOC found that even relatively fast permissioned blockchains were too slow in internal testing. DC/EP would have to be able to handle at least 300,000 transactions per second across the country at peak times to do what cash does. So DC/EP won’t be a blockchain. (The system will apparently contain some sort of blockchain descendant.) Mu Changchun, leader of the DC/EP initiative, co-authored a paper that recommended not to use blockchains to transform traditional payment systems.273

PBOC would like DC/EP to replace physical cash to some degree — “M0,” the physical cash in the economy. The money in DC/EP will be legal tender, just as paper notes are.274

The financial press, as well as the crypto press, leapt upon every new morsel of information about DC/EP — the PBOC had to keep stating that there was no release date as yet. The trickiest part was not the technology — it was regulatory issues, particularly around anti-money-laundering and tax evasion.275

The press covered what looked very like public pilot programmes in several areas from late 2019 on — even as the PBOC insisted on calling these only “internal and closed pilot tests.” PBOC Governor Yi Gang also mentioned testing in “some scenarios of the coming Winter Olympics” in 2022. He stressed again that there was no release date as yet.276

Pilot tests continue through late 2020, at increasing scale.

Why would I want a CBDC?

Consumers want systems that make their lives easier. Convenience is king. Nobody is going to choose a new payment system that’s less convenient.

The users won’t care who issues the dollars, pounds or euros, if they can assume basic legal protections — they’ll only care how easily they can get, move and spend them.

If you’re in a country with good electronic payment systems, a retail CBDC probably won’t do anything new for you. The Reserve Bank of Australia decided against a CBDC in 2020, because, even with the decline in cash use in the COVID-19 crisis, existing systems were still working fine, and there was still good access to physical cash.277

If you’re a banker, of course, it matters a lot who the money is a liability against. But for anyone else, knowing that a payment system is a “CBDC” tells you nothing else about that system. If someone wants you to get excited because something is a CBDC, look closely at what the system does and doesn’t do.

If you’re Facebook, you might want to run blockchain CBDC tokens over your network instead of Libra tokens — in Libra 2.0.


144. Frank Tang. “Facebook’s Libra forcing China to step up plans for its own cryptocurrency, says central bank official.” South China Morning Post, 8 July 2019.

256. “Implications for central banks of the development of electronic money.” Bank for International Settlements, 1 October 1996.

257. J. P. Koning. “Why the Fed is more likely to adopt bitcoin technology than kill it off.” Moneyness (blog), 14 April 2013.

258. J. P. Koning. “Fedcoin.” Moneyness (blog), 19 October 2014.

259. The phrase “central bank digital currency” seems to have been coined by Richard Gendal Brown of enterprise blockchain company R3, in his blog post “A Central Bank “cryptocurrency”? An interesting idea, but maybe not for the reason we think,” 5 March 2015. The initialism “CBDC” was first sighted in a speech by Ben Broadbent of the Bank of England, “Central banks and digital currencies,” 2 March 2016.

260. Claire Jones. “Central bank plans to create digital currencies receive backing.” Financial Times, 30 June 2019.

261. Aleksi Grym. “Lessons learned from the world’s first CBDC.” BoF Economics Review 8/2020, 15 September 2020.

262. Jennifer Sondag, Nathan Gill. “Ecuador Turning to Virtual Currency After Oil Loans.” Bloomberg, 11 August 2014.

263. Everett Rosenfeld. “Ecuador becomes the first country to roll out its own digital cash.” CNBC, 9 February 2015.

264. Evelyn Tapia. “Cinco cambios para impulsar el dinero móvil.” El Comercio, 18 December 2017.

265. “71% de cuentas de dinero electrónico, sin uso en Ecuador.” El Universo, 3 December 2017.

266. María Laura Patiño. “El dinero electrónico, una apuesta peligrosa.” El Universo, 1 June 2016.

267. “Juan Pablo Guerra: Dinero electrónico es un medio de pago, no moneda.” El Universo, 3 December 2017.

268.“Ecuador: Cuentas de dinero electrónico dejarán de funcionar el 31 de marzo.” El Universal, 26 March 2018.

269. Mario Bergara, Jorge Ponce. “Central Bank Digital Currency: The Uruguayan e-Peso case.” In Do We Need Central Bank Digital Currency? Economics, Technology and Institutions. ed. Ernest Gnan, Donato Masciandaro. SUERF, 2018.

270. “Project Sand Dollar: A Bahamas Payments System Modernisation Initiative.” Central Bank of the Bahamas, 24 December 2019.

271. Jim Wyss. “Bahamas Plans E-Currency to Connect Far-Flung Island Beaches.” Bloomberg, 15 September 2020.

272. “ECCB Digital EC Currency Pilot: Frequently Asked Questions.” Eastern Caribbean Central Bank.

273. Mu Changchun, Di Gang, Lu Yuan, Qian Youcai, Qing Su De’. “央行数字货币研究所谈区块链技术的发展与管理” (“The Central Bank Digital Currency Research Institute talks about the development and management of blockchain technology.”) Sina Technology, 21 February 2020.

274. Fan Yifei. “关于数字人民币M0定位的政策含义分析” (“Analysis on the Policy Implications of Digital RMB M0 Positioning.”) China Financial News Network, 14 September 2019.

275. Frank Tang. “China has ‘no timetable’ for launch of its digital currency, says central bank governor.” South China Morning Post, 24 September 2019.

276. “Interview with PBC Governor Yi Gang by Financial News and China Finance on Key Issues During ‘Two Sessions’.” People’s Bank of China, 30 May 2020.

277. Reserve Bank of Australia. “Payments System Board Update: August 2020 Meeting.” Press release, 21 August 2020.

Become a Patron!

Your subscriptions keep this site going. Sign up today!

3 Comments on “Libra Shrugged — Chapter 15: Central bank digital currencies”

    1. Wasn’t quite a CBDC (cash issued by the central bank) – though it’s pretty close. There’s a couple of posts on Consult Hyperion’s site about the similarities (they did a lot of the work on Mondex):

      “Of the 1990s electronic cash schemes, Mondex had a feature set which would most lend itself to a CBDC. I don’t say that a CDBC should be based on Mondex, but I believe it could be.”

  1. Heh. Yeah. I think I have used cash two or three times in Sweden, since landing beginning of June.

    I mean, I was not a massive cash user in the UK, mostly paying by card. But, here, there’s actually multiple places that only accept card or Swish payments, no cash handled (and, yes, clearly signposted).

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.