This archived content originally appeared at, the predecessor site to, and does not carry the sponsorship of the Cato Institute.

Warren Coats on Bitcoin

Worth your while to read. Warren, as some readers will know, has a wide range of experience with monetary policy and successful monetary reform over time and around the world.

  • michaelsuede

    I can tell this guy is a Keynesian policy wonk before I even bothered to look at his credentials. His arguments have nothing to do with real economics.

    He never talks about unique properties bitcoin has that make it valuable as a currency. It's completely fungible, extremely divisible, scarce, impossible to counterfeit, highly recognizable, incredibly easy to transport, incredibly easy to secure, in fact there is no difference between bitcoin's properties and commodity, other than the fact it's not physical.

    So we know the commodity gold has acted as a money since the dawn of civilization, and when it is used as a money, it sits as blank ingots or rounds in bank vaults. When gold is used as a money, it simply acts as an accounting mechanism. The "physicalness" of gold doesn't mean anything. It's just sits in vaults for accounting purposes. In fact, people used to bit pieces of gold coin off, so what good is a hunk of gold with teeth marks in it? It can ONLY be good for accounting purposes, nothing else!

    Bitcoin is simply a gold standard accounting system without the gold. There's no difference. Clearly bitcoin is going to keep growing in acceptance because even a ten year old can see the benefits it brings to the table.

    Coats never talks about gold's volatility in value when it was a money. Why not? Gold has so many more reasons why it should be volatile than bitcoin. There are no supply externalities with bitcoin, which means it MUST be more stable in value than a commodity money. Perhaps not now, but there's no reason why once it becomes widely accepted that it shouldn't be incredibly stable in value.

    Why cite Warren Coats on this blog? He's not an Austrian.
    [Post lightly edited for civility. — Kurt Schuler]

    • None of the Bitcoin properties you list are unique to Bitcoin. Litecoin has all of them. Peercoin has them and also has a more elastic supply. Zerocoin will have all of them and also supports untraceable transactions. Most people have never heard of these alternatives, but this fact is irrelevant.

      Most people had never heard of Zip compression before it replaced Arc compression practically overnight (following intellectual property claims involving Arc) in the early nineties. This rapid transition occurred despite considerable network externalities favoring Arc. Millions of people had data archives compressed with Arc on their computers and commonly exchanged these archives with other people using Arc compression/depression software, and Arc had these advantages when networks were much less mature, before the widespread adoption of internet protocols. People rapidly abandoned Arc regardless.

      Gold has not been money since the dawn of civilization and has never been very commonly used as money. Silver coins were more common in the American colonies before the revolution, both because the Crown monopolized gold more effectively and because silver was a more common metal but still scarce enough. Scarcity is a necessary property of money, but the most scarce commodity is not therefore most useful as money.

      Bitcoin's price instability is a direct consequence of its inelastic supply. Money is not immune to the laws of supply and demand. If the supply of money does not respond to changes in demand, prices are not stable.

  • Warren

    (Note: I'm not Warren Coats)

    It was a good article, I really have nothing to say on Bitcoin, however the following section stood out to me.

    —The difference between cash (physical banknotes and coins) and other means of payment such as checks, wire transfers, Visa, and Paypal, is that the payment of cash is made directly between the parties to the payment, i.e. peer-to-peer. All other means of payment involve trusted third parties, such as banks, credit companies, and exchanges,—

    This is something I've been wondering about. Given the recent stories of both government and freelance criminals rendering the electronic system insecure will there or is there a growing tendency for folks to prefer holding currency rather than trusting third parties?

    • An interesting question. Most or all of the hacking of electronic means of payment, to my knowledge, has been of credit/debit accounts, not bank depositories. One of the advantages of banks is that they are liable for the money we deposit their and thus responsible for making up for any losses if hacked or otherwise stolen (as long as they remain solvent, otherwise deposit insurance covers any loss up to the insurance limit). If you loss bitcoin you have no recourse. However, at least one of the bitcoin on-line wallet services (a trusted third party) did make good the bitcoin stolen from it in order to protect its reputation.

  • Bitcoin also has some technical issues related to the anonymity/traceability of transactions (not anonymous enough), vulnerability to attack by a sufficiently powerful (in terms of IT resources) attacker (like a state), the difficulty of creating a fully autonomous node as the size of the blockchain grows, for example; however, I agree with Coats that Bitcoin's principal weakness involves economics rather than information technology. The Bitcoin supply is not elastic enough to achieve price stability. Ultimately, it's not elastic at all. The cost of transactions is not the issue as much as the fitness of Bitcoin as a unit of account and a standard of value for extending credit.

    I don't agree that Bitcoin's head start gives it much of an advantage, via network externalities, at this point. Valuable externalities involve the decentralized nature of a cryptocurrency like Bitcoin, and the necessary proliferation of IT infrastructure supporting it, more than the advantages of using a common currency. I expect the value of a common currency to decline in the future. already accepts many currencies and could accept a practically unlimited variety at little additional cost (as long as buyers bear currency exchange costs), yet when I use Amazon, I see every price in dollars and can easily imagine everyone else on Earth using the dollar.

    I don't much agree that Bitcoin's transaction costs are low either. Transferring Bitcoin from one Bitcoin wallet to another has a low cost to the parties at this point, but commercial transactions (other than donations) often don't occur this way. For example, does not accept Bitcoin this way. It rather uses a brokerage service to encourage Bitcoin holders to sell their Bitcoin for dollars before spending the dollars at, and this transaction is very costly to the buyer in fact. The buyer doesn't notice the cost, because he pays it in Bitcoin, and the cost in Bitcoin is small compared to the volatility of prices paid in Bitcoin.

    Even direct, peer-to-peer transfers eventually will be costly, because Bitcoin miners/double-spending-police must eventually charge fees. When there's nothing left to mine, the decentralized nature of the blockchain makes Bitcoin's solution of the double spending problem more costly than more centralized solution (like Dwolla's). The number of nodes policing the blockchain could be (and almost certainly would be) smaller at this point, but fewer nodes increases the network's vulnerability to attack.

    Bitcoin's transaction processing network has a head start, but a currency-agnostic network seems far more valuable, so I expect much of the existing Bitcoin infrastructure to support other crypto-currencies ultimately. If Amazon starts accepting Bitcoin tomorrow, online merchants accepting only Bitcoin, like, disappear overnight, and I don't expect Amazon or other merchants to have any zealous devotion to Bitcoin specifically.

    Bitcoin could morph into Bitcoin 2.0, a similar system with a more elastic money supply and fewer vulnerabilities; otherwise, I don't see it surviving the decade. How shocking is that really? Who uses MySpace these days? Who uses FidoNet? Who uses Arc compression? Who uses a Wright Flyer or an automobile with an external combustion engine? Why would anyone expect the first decentralized crypto-currency to be the last one?

  • Coats wrote, "The buyer of the toy pays dollars to a seller of the bitcoin equivalent of $100 via an exchange such as Mt. Gox in Japan and instantly transfers them to the merchant selling the toy. After a few minutes delay while “miners” verify that the bitcoin being transferred (first to the buyer of the toy and then to its seller) are genuine and not electronic copies (i.e. counterfeit), the merchant sells them for dollars."

    Services like BitPay and Coinbase don't operate this way. A merchant using these services need never hold Bitcoin at all. BitPay has a contract with the merchant guaranteeing a price in dollars for the merchant's goods. Essentially, the buyer of the toy first sells Bitcoin to BitPay for a specified price in dollars. The buyer then pays the merchant in dollars. BitPay (after a time) sells the Bitcoin for dollars.

    A merchant using BitPay may choose to hold Bitcoin, but BitPay must settle with the merchant in dollars at the end of each day, and the dollars owed are determined at the time of the transaction, the dollar price of the goods, not the dollar value of the "Bitcoin price" at the end of the day.

    In other words, BitPay bears the arbitrage risk associated with offering merchants a timely transaction, and its business model is sustainable only if it passes the cost of this risk as well as its infrastructure costs onto buyers. It must do so by charging a commission on the Bitcoin/dollar exchange, and this commission is necessarily proportional to the price of the good and to the volatility of Bitcoin's price in dollars. BitPay incorporates this commission into the "Bitcoin price" of the good seen by the buyer.

    By contrast, Dwolla can charge a flat rate per transaction, since it only transfers dollars from one account to another and bears no arbitrage risk. The infrastructure cost of all Dwolla transactions is practically fixed. The marginal cost per transaction is practically zero.

  • John S

    Kurt, Warren Coats, or other commenters: any thoughts on this cryptocurrency development in Singapore?

    "A new service launching in Singapore today uses the Ripple network to acquire, store and convert precious metals into any currency for customers anywhere in the world.

    Users maintain a balance in a Ripple wallet (similar to a bitcoin wallet), while the bullion itself is vaulted in Singapore. They can then use that balance to pay anyone instantly in any currency available on Ripple, or trade the bullion itself. Ripple Singapore’s model charges 0.2% per transaction, and nothing for vault storage.

    There are two ways to buy bullion with Ripple Singapore: Directly via its ‘Bullion Counter’ (for a minimum 10oz of gold/platinum or 500oz of silver), or indirectly from other users, with no minimum at all. Anyone who wants to take physical delivery of their bullion may do that too, with two working days’ notice."

    • John S

      From a Free Banking perspective, Ripple seems a lot more interesting (and potentially more influential) than Bitcoin. Also, any thoughts how XRP, Ripple's native chartal currency, might evolve?

      "To ‘establish a trustline’, or exchange value, on the Ripple Network, users must spend a small amount of its native currency, XRP. Ripple Singapore will even supply the XRP to its users at no cost to get things moving.

      Cox explained XRP simply: “Imagine each transaction on Ripple is a sealed envelope going from A to B, and XRP is the stamp on that envelope. The network is neutral regarding the the sender, receiver or the contents or value of the letter, it still requires an XRP stamp.”