Libra: economics of Facebook’s cryptocurrency

Facebook last week announced plans for Libra, a new global cryptocurrency. The name seems to be a marriage of the words “livre”, the French currency throughout the Middle Ages based on a pound of silver, and “liber,” which is Latin for “free.” Facebook claims that Libra will give the freedom to easily transmit funds across borders to the 1.7 billion adults in the world without access to traditional banks.

Money is defined by three attributes. It is a unit of account (prices of most things you buy are quoted in dollars), a medium of exchange (you can deliver payment for those items by transferring your dollars to the seller), and a store of value (you can hold your wealth in the form of cash dollars until you want to spend it).

Why do the pieces of paper we call dollars have value? When paper currencies are introduced, the sovereign often gives them initial value by promising to redeem a one-pound note for a pound of silver. But eventually the pieces of paper come to be regarded as having intrinsic value even in the absence of such promises. This value stems from the fact that having a common unit of account, medium of exchange, and store of value greatly facilitates transactions. We can do everything much more efficiently– and therefore, we can all be richer– if we live in a world where we can pay for oranges using dollars instead of delivering to the seller the physical item that the orange-grower happens to want. A social agreement to use paper dollars as money allows us to enjoy that benefit, so it’s in everybody’s interest to keep it going.

As a result of money’s value as a facilitater of transactions, governments obtain an extra source of revenue just by printing seemingly worthless pieces of paper, essentially appropriating for the government some of the social value that the existence of money helps generate. Governments use this power to get things of real value, such as pay the soldiers or acquire interest-earning assets. This revenue source is known as “seigniorage.” Seigniorage in the form of interest earned on investments by the Federal Reserve and turned over to the Treasury contributed $62 billion in revenue for the U.S. government in 2018.

No surprise that private companies would like a piece of that action, if they could get it. Facebook earned $7 billion in profit last year, more than the entire GDP of some small poor countries like Malawi and Sierra Leone. So why not act like a sovereign government and issue its own currency? Moreover, Facebook isn’t doing this alone, but has partnered with dozens of other companies, including Mastercard, Visa, Ebay, Paypal, Spotify, Uber, and Lyft. Unlike Paypal, which is a functioning medium of exchange, Libra is also proposed as a unit of account: prices would be quoted and transacted directly in Libra. And unlike cryptocurrencies like Bitcoin, whose value fluctuates wildly, the idea is that the value of the Libra would be fixed based on a basket of stable currencies like the dollar, euro, and yen, supported by an endowment of reserves (interest-earning assets in those currencies) that are added to or subtracted from with each transfer into or out of a Libra account:

For new coins to be minted, there must be a commensurate payment of fiat by resellers into the reserve. Through interaction with authorized resellers, the association automatically mints new coins when demand increases and destroys them when the demand contracts. Because the reserve will not be actively managed, any appreciation or depreciation in the value of the Libra will come solely as a result of FX market movements.

This sounds essentially like a 100%-reserve currency board. When you convert your dollars into Libra, the association invests your dollars in low-risk assets of the currency basket. If the value of the euro subsequently depreciates against the dollar, your Libra will be worth fewer dollars than you put in. But it’s still worth the same amount in terms of Libra. Though who’s to stop some other institution from offering a Libra-denominated account, just as banks in Europe can offer a dollar-denominated account, and operate (as private banks do) with a fractional-reserve balance sheet? That would introduce a new dimension in the rapidly mutating world of shadow banking. And it’s unclear who would or could regulate it.

There are additional risks to your potential Libra account, including security or logistical problems with the blockchain technology, development of superior payments technologies, or government regulation. And it’s not clear how much your holdings would be worth if, for example, Facebook were to declare bankruptcy. If the Libra becomes successful, and then for some reason experiences a collapse, it could generate its own financial crisis.

On the other hand, there is no doubt that having a unified and stable system for international transactions would be a big benefit for many people in the world, and thus a not inconsequential source of seigniorage revenue to Facebook and its partners from the interest earned by the invested assets. Certainly having a unified stable currency benefited members of the European monetary union, making it easier to buy and sell across borders. But we saw in 2011 that it can also be a big problem. It would have been very much in the interests of some of the periphery countries of Europe to see a sharp depreciation of the exchange rate at that time. But they couldn’t, because they were tied to the euro. If most Congolese transactions come to be transacted in Libra, it could make it harder for the country to deal with a shock like a collapse in real commodity prices.

If the Libra becomes successful, it could undermine the ability of governments to force negative interest rates– the Libra could start to look like a more attractive money than euros. And it also poses potential competition for governments’ seigniorage revenue, which poorer countries often rely on because the lack of infrastructure makes it hard to raise revenue using other forms of taxation. Though if it does help improve the efficiency of the payment system, perhaps the wealth created would generate other new targets for tax revenue.

It is an interesting experiment. I don’t know how it will turn out.

47 thoughts on “Libra: economics of Facebook’s cryptocurrency

  1. pgl

    Best discussion of Libra I have seen so far. On the seigniorage the FED turns over to the Treasury:

    “Seigniorage in the form of interest earned on investments by the Federal Reserve and turned over to the Treasury contributed $62 billion in revenue for the U.S. government in 2018.”

    It is interesting that this is going down even as interest rates have turned up a bit. Of course the FED has scaled back its balance sheet on the last few year. With GDP now in excess of $20 trillion, this $62 per year is a mere 0.3% of GDP. We have never relied heavily on seigniorage to finance the Federal government. Of course we have not relied on tariffs that heavily in a last few generations – at least until now.

  2. Moses Herzog

    Some very naive consumers and web users trusted Zuckerberg with their privacy. Now they want to trust him with their savings?? I wish them all good luck and I’ll do my best to hide my utter disdain for the level of their intelligence quotient. Does the phrase “high tech Amway” register with anyone here??

    1. pgl

      Trust is an essential aspect for establishing a currency. I would not trust Zuck to take out my trash.

    2. pgl

      Multilevel Marketing! Herbalife is a publicly traded MLM peddling overpriced and overhyped junk science. I would never buy their products and have certainly never bought any of their shares.

    3. Bruce Hall

      I find myself in the peculiar situation of completely agreeing with you, pgl, and 2slug. pgl hit it squarely on the head: trust is the essential element for any currency.

  3. 2slugbaits

    JDH One small correction. The Latin word “liber” does indeed mean “free”‘; however, the Latin “Liber” (with a capital “L”) refers to the Roman god of wine and sexual license (i.e., another association with Bacchus or Dionysius). Since Facebook uses “Liber” rather than “liber” I’m assuming they were being ironically sly. Or maybe they just didn’t take Latin back in college.

    1. Moses Herzog

      If it is a proper noun designating a product or “medium of exchange”, wouldn’t it be capitalized either way??—aside from underlying meaning??

      Other than not attaining his university degree, everything I have read about Zuckerberg indicates he is hyper-educated (again, aside from not getting his degree). I kinda doubt that is any kind of an “error”.

      1. 2slugbaits

        Moses Herzog If it is a proper noun designating a product or “medium of exchange”, wouldn’t it be capitalized either way??

        Well, we don’t capitalize “dollar” or “euro” or “pound” or “franc” or “yen” so who knows whether he had an ulterior motive in naming it “Libra” rather than “libra”.

        1. Moses Herzog

          Wouldn’t the proper comparison be to other cryptocurrencies, not government issued paper currency?? Again, it’s a proper noun. You’re not referencing Barkley Junior’s favorite site “Quora” again, are you??

          1. 2slugbaits

            Moses Herzog Wouldn’t the proper comparison be to other cryptocurrencies

            Fair point. Another motive for capitalizing might be trademark reasons.

            You’re not referencing Barkley Junior’s favorite site “Quora”

            No. In college I took three years of Latin.

            Instead of calling it the “Libra” they could have called it the “Thaler.” The thaler was widely used in the Middle Ages east of the Rhine, and via the Hapsburg dynasty evolved into the historical basis for the currency used by most Western Hemisphere nations. Derivations of the thaler are still used in some countries today.. It’s where we got our “dollar” via the Spanish dollar.

    2. Barkley Rosser


      They selected “Libra” because it means “pound” in Latin. Trying to make it look like a real currency. Obviously there are some other evocations, but that is the official reason.

      1. 2slugbaits

        Yes, “pound” (in twelve troy ounces) is a secondary meaning of the Latin “libra”. It’s primary meaning is a scale or balance. In those statues of the blindfolded goddess of justice she is holding a libra.

        1. Barkley Rosser


          There never was a Roman currency named “Libra” (or “libra”). The first currency derived from that name was established by Charlemagne due to a decline in the availability of gold in Western Europe after the rise of Islam cutting off flows from North Africa. It was based on a pound of silver and bore the Old French name derived from the Latin of “livre.” From this ultimately would come currencies called “lira,” such as the former Italian one and the current Turkish one. William the Conqueror took Charlemagne’s currency to England in 1066, where the name eventually was Anglicized to “pound sterling,” although it has been a long time since a British pound was worth a pound of silver.

          Regarding “Quora,” I have never been to the site. I once here quoted somebody else’s citation of it as a source for something, that was correct, although a certain person decided to make a big fuss about it and use the fact that it originally came from Quora to question the fact. Bringing this up now with exaggerated extra claims is almost as silly as some of the stuff CoRev has been coming up with.

          1. 2slugbaits

            Barkley Rosser I don’t think I said there was a Roman currency named libra. I thought I said that “pound” was a secondary meaning and that its primary meaning was a scale.

            established by Charlemagne due to a decline in the availability of gold in Western Europe after the rise of Islam

            FWIW, I have read Henri Pirenne, although the Pirenne thesis has fallen out of favor.

            Sorry, but I’m not following your comment about Quora. Moses asked if I got the definition from Quora.

          2. Barkley Rosser


            I never said you said there was a Roman currency of that name. I was just trying to clarify the historical situation. I am not sure which part of Pirenne you thinki has “fallen out of favor,” (much he argued for is much debated and has been), but the part about Charlemagne establishing the livre as a pound of silver unit of account because of a gold shortage due to Muslin-related drying up of trade routes through North Africa is historically correct and easily checked. Just google “French livre” and you will find multiple sources such as encyclopedias presenting basically exactly what I said, with the added detail that it was Pepin, Charlemagne’s father who decided to go off gold due to its shortage, although it was Charlemagne who established the new currency (whose related coin form was the denier).

            As it is, while balance may be the primary meaning of “libra” in Latin, that the inventors of Libra have named it for the pound meaning from Latin comes from them, not me.

            This Quora business is a weird obsession of Moses’s, basically not worth bothering with. Go see his latest posting on this at the end of one of the recent Soybeans threads if you are curious, the one that keeps getting more comments. I have just commented on it there and will not do so here further, but this is just Moses getting all bent out of shape again over nothing other than again refusing to accept that his knowledge of probability theory is poor. As it is, I linked to Quora once and know nothing about the site, although Moses insists that it is awful and I am awful because I quoted something from it once.

    3. Willie

      I would go with the latter. But then, Facebook started out as a snide fratboy thing so far as I know, so maybe the former is true.

      What could possibly go wrong here? If you have a run on the LIbra, who bails out Facebook? Would the next great depression be the end of Facebook and the Libra because no taxpayers are on the hook when Libra based savings evaporate when a bubble pops? The mind boggles at the possibilities. There are worse ideas, but private currency is right up there with private standing armies.

  4. pgl

    Art Laffer on MSNBC right now. He just claimed that the low interest rates when Obama was President was the cause of the Great Recession. But he followed that by saying Trump is correct that the FED should lower interest rates today. This dude is an economist??????

    Oh wait – he also just said that real GDP grew by 8% per year from the first of 1981 to the end of 1984. Seriously?

  5. Bernard Leikind

    Prof. Hamilton,
    The Fed earns it’s seigniorage mostly on Uncle Sam’s bonds. The US Treasury pays this interest to the Fed. Then the Fed gives it back to the Treasury as profit. It seems as if the Fed and the Treasury just move money back and forth.

    The Fed buys its federal bonds from the private sector, which got them by loaning the Treasury some money. For the Fed to buy such a bond means that the private sector investor has been repaid for the money he or she gave to the Treasury.

    Why doesn’t the Fed stamp the bonds it buys Paid in Full and return them to the Treasury? After all the bond is no longer part of Uncle Sam’s debt held by the public. Why have the Treasury pay the Fed the bond’s interest, and the Fed give it back with a different name?

    1. James_Hamilton Post author

      Bernard Leikind: Suppose that the Treasury borrows a million dollars in order to buy a tank and later the Fed purchases that million dollar Treasury bill by creating new reserves (electronic credits for cash). You could say that, from the perspective of the unified federal budget, the government paid for the tank by printing cash. But it is very important to have these two steps separated in terms of institutional decision-making. If a country does not have an independent central bank, and the fiscal authority thinks of newly printed cash as a way to pay all its bills, you have a recipe for disaster. The experiment has been tried many times in history, and the evidence is very clear. We want and need an independent central bank, and we want and need a complete institutional separation between the Treasury’s operation of issuing a new Tbill and the Federal Reserve’s decision of whether to purchase that Tbill.

      Once the Fed makes a decision to purchase a security with new reserves, it does not matter in terms of the accounting whether the security they buy is a Treasury bill or some other interest-earning asset. In either case, the Fed will earn some interest and return that interest to the Treasury. If the security is a Tbill, the Treasury says it paid interest on the Tbill to the Fed and received that interest back from the Fed, so it’s as if the Treasury bought the tank interest free with newly created cash. If it’s a private security, the Treasury says it paid interest on the Tbill to some private party and had an offsetting receipt in interest that the Fed contributed to the Treasury. So it’s still the same as if the Treasury bought the tank interest free with newly created cash.

      1. 2slugbaits

        Nice intuitive summary of the relationship between the Fed and Treasury. Pretty sure you can’t by an Abrams tank for $1M, but otherwise I liked it.

        In your main discussion you pointed out that money has three attributes. One of those attributes is transactional convenience. The Libra could be a global plus in that regard. Another attribute is as a store of value. Since I doubt that users would view the dollar and the Libra as having equal integrity as stores of value, shouldn’t we expect some portfolio adjustments that would affect the demand for dollar denominated securities? Any thoughts on how those portfolio adjustments might affect interest rates on Treasury securities?

        I can’t imagine that the State Dept. would be thrilled with the Libra because it could undercut the US’s ability to impose trade restrictions. How effective would US restrictions on trade with Iran be if the Libra were a widely accepted cryptocurrency?

        1. Rick Stryker


          Libra will reportedly follow all the money laundering laws and requirements that international banks do, including national restrictions.

      2. Matt Young

        How exactly do we separate the central bank and Congress without triggering the monopoly cycle.
        The typical method is jaw,jaw,jaw, but there is no jaw philosophy that separates Congress from Treasury. We need a method, any ideas?

  6. Joe

    Government fiat also pays off the local sheriff when he comes for your stuff. This is a very valuable service.
    Seigniorage is a license fee, paid to a monopoly, it is a monopoly wedge and all monopolies cycle, right?. No? What is the volatility theory of monopolies, anyone have a sound theory? They cycle, folks, they cycle.

  7. Moses Herzog

    @ Professor Hamilton
    I would be curious your thoughts on if the Libra can actually compare (at least at some point in the future) to something like the IMF’s SDRs (Special Drawing Rights) rather than “just another currency”?? It seems to have at least some similar characteristics there. Or is that way out of left field on my part??

  8. kav

    If the component of the currency board have negative interest rates (like 1/3 of developed country’s govy paper do atm), would it also be the case that libra has a negative interest rate? If not, how will the currency board be compensated for the melting away of the deposits in the reserve?

  9. KevinT

    I have yet to see anyone raise the possibility that Facebook sails into a sh–storm when problems arise due to money laundering and KYC regulations. How does that get addressed with a global “currency” and money transfer system?

    Personally, I also agree with the other comments regarding the idiocy of anyone placing that much trust in Facebook/Zuckerberg.

  10. Not Trampis

    Yes I echo PGL comment. This is the best article on Libra I have read. And I thought Libra was merely playing on astrological fantasies!!

  11. SecondLook

    Libra: astrological sign, symbol is balanced scales.
    Occam’s’ razor – particularly for a California based company.

  12. rafaminos

    best discussion on libra I have read
    you write
    “If the Libra becomes successful, it could undermine the ability of governments to force negative interest rates– the Libra could start to look like a more attractive money than euros”

    but it seems it cannot be true. if interest rates on libra are 0% and negative on the assets in the reserve. then the backing will progressively fall and it will undermine the libra’s credibility.
    It seems like a fixed exchange rate and open capital account implies a common monetary policy (impossible trinity).

  13. Willie

    If Facebook, or some other private entity, controls a reserve currency, what would prevent that entity from manipulating the currency for its own ends, to the detriment of the overall economy? It could use its currency to undermine competitors and to force its way into other markets. Unlike a government, which is at least theoretically accountable to the nation it governs, a private entity that controls a currency is accountable to no one. Never mind the central bank’s ability to moderate economic swings, the potential for vast corruption is staggering if a private currency becomes established.

  14. Barkley Rosser

    Jim’s post here has been linked to at Marginal Revolution along withone by Barrry Eichengreen and one by Ben Thompson. Jim’s was probably the most favorable to Libra. Eichengreen was harshly critical and pessimistic. Thompson was more in the middle and made a useful point I agree with. He noted that there is a tradeoff between trust/security and efficiency. He is clear that Libra is not a real cryptocurrency and does not use a real blockchain, with all the talk of that basically marketing. So it does not have the security that can provide trust that a real blockchain provides. But real blockchains are inefficient, gobbling up electricity with the mining and and taking a long time for somebody to make an actual payment for something as is the case with bitcoin. So presumably Libra would allow more rapid paying like Paypal or a credit card, but as basically just a distributed ledger,it could be more easily hacked than a real cryptocurrenccy using a real blockchain.

      1. Robert S Cartwright

        If you want to read Washington Post articles behind the pay wall, you can either (i) install multiple browsers on your machine and switch to a rarely used browser (since WaPo permits a small number of free articles per month) or (ii) periodically delete the WaPo cookies in your usual browser (you can search the internet for intructions on how to do this for your chosen browser). WaPo apparently keeps track of the number of articles that you have read using browser cookies.

        I am a computer scientist who stumbled across this web site and marveled at seeing a post from Barkley Rosser (whom I presume is J. Barkley Rosser Jr., professor of economics at George Mason University and son of the late J. Barkley Rosser [see, a pioneer in the mathematical theory of computation).

        1. Barkley Rosser


          I am at James Madison University, not George Mason. Otherwise, guilty as charged, with one person here regularly labeling me “Barkley Junior.”

    1. Rick Stryker


      The trust model and the security of the cryptocurrency are not the same. Bitcoin and its variants use a trustless model in which anyone can read and write to the database and everyone is untrusted. Satoshi, who designed Bitcoin, took off the shelf computer science concepts to implement the trustless model and he decided to use a proof of work protocol, which is responsible for the heavy use of electricity. There is no intrinsic connection between a trustless model and proof of work however. Etherium has for some time been planning to move from proof of work to proof of stake, which doesn’t require the power consumption.

      Libra is using a semi-trustless model in which only certain entities can read or write to the database. The algorithm to implement the trust model is a fault tolerant Byzantine algorithm (VMware’s hotstuff) in which up to 1/3 of the entities can be dishonest but the database entries could still be trusted. That doesn’t mean that Libra is less susceptible to hacking though. Security of the cryptocurrency depends on much more than the trust model employed. Like etherium, Libra has implemented a virtual machine and language called “Move” to implement transactions. No one really knows yet the security vulnerabilities implied by Move for example. Etherium’s programming language was hacked.

      1. Rick Stryker

        By the way, Libra plans to go to a fully trustless model in 5 years or so but there do not seem to be any definitive plans at this point on how they would do that.

      2. Barkley Rosser


        We have terminological problems here. You seem to be using “trust” to mean there are authorities involved whom you must trust. I think Ben Thompson’s meaning, and I see nothing in your remarks to offset this, is that there is high trust because there are no authorities you need to trust, but this is deeply linked to security as in not so easily hacked, which you deem to be a separate issue. If in fact a truly trustless system a la bitcoin is able to be that way because it has high security due to its proof of work with a full blockchain, then I think this is what Thompson means. Again, clearly as it currently appears, Libra will be somewhat intermediate, “semi-trust” if you like, a long with being “semi-efficient,” which means it is more easily hacked and thus less secure and less able to go to being a truly decentralized trustworthy trustless system.

        As it is, I am not at all ready to buy the marketing coming out of FB that Move is going to resolve lots of these problems, although perhaps I should not be so cynical and skeptical.

        BTW, Tyler Cowen’s latest line, which looks quite reasonable, is that the most likely use for Libra, if it gets off the ground, might be for international remittances. Otherwise it is not obvious that it will have all that much superioirty over just standard Paypal like ledger sysrtems for general transactions. But with all the fees and restrictions involved with remittances, there may be an opening there that bitcoin and etherium et al do not satisfy.

        1. Rick Stryker


          I believe Thompson is using the term “trust” in the same sense I am. Look at his chart. He has bitcoin in the low trust, low efficiency section and Libra in the high trust high efficiency section. He is saying that when you don’t need to trust the validators, as in bitcoin, efficiency is also low. However, if you are willing to trust the validators more (Libra), then efficiency goes up. By efficiency, Thompson seems to mean transactions processed per second.

          I don’t think he’s right about a trade off between trust and efficiency. It’s true the blockchain by itself has inherent limitations in processing transactions per second since the block size is only 1 MB and blocks are added about every 10 minutes. But there are ways around that, the most important being the Lightning network, in conjunction with segregated witness and Schnorr signatures. These innovations can increase transactions per second very dramatically while still keeping a fully trustless model.

          Move was added to enable smart contracts a la etherium, not to solve any particular problem. Bitcoin has a simple scripting language while etherium implements a Turing complete programming language running on a virtual machine. Move is supposed to be in between those extremes. They hope it’s more secure than Solidity, etheriums language, but we’ll have to see.

          I agree that Libra is targeting emerging markets and the foreign remittance market.

          1. Barkley Rosser

            Fair enough, Rick, that Thompson says bitcoin is low trust as it does not need ttust.

            In terms of being intermediate on both trust and efficiency, see the lower figure where ha compares bitcoin, Libra, and a FB coin in which FB would have full knowledge of what was going on. Libra is intermediate between the other two on both trust and efficiency.

  15. don

    Could Facebook be forced to divulge transactions and holdings so income tax or sales tax can be levied? And there would have to be an exchange rate to establish the tax liabilities… Milton Friedman would love the ‘freedom’ generated by such transactions, but it would seem to be a potential nightmare for taxing authorities.

  16. Turner

    Hey James, not to nit-pick, but Facebook earned $7B in net income in Q4 of last year. Profit was $22B for the entire year.


Comments are closed.