Friday, July 5, 2019

Classifying cryptocurrencies



Whenever biologists stumble on a strange specimen, they first try to see if it fits into the existing taxonomy. If it doesn't fall within any of the pre-existing categories, they sketch out a new one for it.

For people like myself who are interested in monetary phenomena and finance, Bitcoin and other cryptocurrencies like Dogecoin and Litecoin have presented us with the same challenge. How can we classify these strange new instruments?

Because they have the word 'currency' in them, the knee-jerk reaction has been to put cryptocurrencies in the same bucket as so-called fiat money, i.e. instruments like bank deposits and banknotes. But this is wrong. Bitcoin, Dogecoin, and other cryptocurrencies are fundamentally different from $100 bills or Citibank deposits. 

To see why, here is a chart I published last year at Sound Money Project:


I've located cryptocurrencies in the zero-sum outcome family. Banknotes and deposits are in a different family, win-win opportunities. The property that binds all zero-sum games together is that the amount of resources contributed to the pot is precisely equal to the amount that is paid out of the pot. Jack's ability to profit from his cryptocurrency is entirely dependent on the next player, Jill, stepping forward and taking them off him at a higher price. Likewise, the amount Jack wins from the lottery is a function of how much Jill and other players have contributed to the pot.

Compare this to a stock or a bond. As long as the firm’s managers deploy the money in the pot wisely, the firm can throw off more resources than the amount that shareholders and bondholders originally contributed.

People have been asking me to extend this classification to other assets. Below I've made a more extensive chart:


Similar to the first chart, I've put Bitcoin, Dogecoin, and other cryptocurrencies in the bets & hedges category along with insurance, futures & options, and various gambles such as lotteries. I describe the members of this family as sterile uses of wealth. Unlike more productive uses of wealth, which increase society's resources, bets and hedges transfer existing resources from one person to another.

I disagree with you, JP

No doubt others will disagree with my classification scheme. For instance, why not put cryptocurrencies in the consumer goods section? After all, aren't cryptocurrencies sort of like collectibles? Don't people primarily collect sports cards, old coins, and crocheted doilies because they expect these objects to rise in value, just like the people who buy cryptocurrencies?

Collectibles and other knick-knacks have sentimental, symbolic, ornamental, and ceremonial value. Even if they can't be sold (most knick-knacks can't), they are still valuable for the above reasons. Not so Bitcoin, Dogecoin, and Litecoin. Cryptocurrencies are pure bets on subsequent people accepting or buying them. If no one steps up, the tokens don't have any other redeeming features that can salvage their value.

Are cryptocurrencies like art? Leonardo da Vinci's Salvator Mundi sold for $450 million to a Saudi prince in 2017. Surely Salvator Mundi's consumption value isn't that high. It would seem that its value is entirely predicated on what the next aesthete will pay, in the same way that bitcoin's value hinges on whether another bitcoiner arrives.

Perhaps. But most art pieces aren't Leonardo's Salvator Mundi. The great mass of paintings that have been created over time are relatively cheap. Secondly, prices in high-end art markets may seem to be disconnected from the consumption value they provide, but that's only because these prices are being drive by the preferences and tastes of consumers who are far richer than most of us. It is this ability to consume the beauty and meaning of art that separates it from cryptocurrency.

What about categorizing cryptocurrencies as commodities? For instance, Bitcoin is often described as digital gold. Or consider George Selgin's reference to bitcoin as a synthetic commodity. Selgin's argument is that cryptocurrencies are commodity-like because they are scarce. And they are synthetic because, unlike commodities, they have no value apart from what other people will pay for them (i.e. they have no nonmonetary value).

I agree with Selgin's analysis. But because cryptocurrencies are synthetic—i.e. their purchasing power is entirely predicated on another person entering the game—I've put them in the bets & hedges category along with other zero-sum games, not the commodity family. Sure, the supply of cryptocurrencies is fixed, say like copper. But that only makes it a very special type of bet, not a commodity.

Blurred lines

The categories in my classification scheme do sometimes blur. At times the stock market becomes incredibly speculative. People start buying shares not because they expect the underlying business to produce higher cash flows, but because they expect others to buy those shares at a higher price, these buyers in turn expecting others to purchase it at a higher price. Thus buying stocks becomes for like betting on a zero-sum game than an effort to appraise the earnings potential of an underlying business.

The same applies to gold:

There is another type of blurriness. Notice that neither chart has a category for money. That's because I prefer to think of money as an adjective, not a category. More specifically, moneyness is a characteristic that attaches itself by varying degrees to all of the instruments in the chart above. A more money-like instrument is relatively more tradeable, or marketable, than a less money-like instrument.

So we can have money-like commodities, bonds with high degrees of moneyness, and heck, money-like lottery tickets. Even some types of banknotes will be more money-like than others. For example, you'll have much better luck spending fifty C$20 bills than you will one C$1000 banknote. Or take the example of choice urban land, which is a lot more saleable than property in the middle of nowhere. Lastly, spending bitcoins is probably much easier to do than spending Dogecoins.

For the last few centuries, the most money-like instruments have tended to be in the debt category. There are many reasons for this. Debt instruments are stable, they are light and thus convenient for transporting, they can be digitized and used remotely, they are fungible, they are difficult to counterfeit, and they can be efficiently produced.

All of you folks with some spare funds who are mulling a big cryptocurrency purchase: be careful. There are plenty of people on the internet who are aggressively marketing crypto as some sort of new society-transforming elixir, or tomorrow's money. But much of their marketing is unfounded. It is unlikely that Bitcoin or Dogecoin will ever attract the same degree of moneyness as the most popular debt instruments. Their zero-sum game nature will always interfere with their ability to attract usage as a medium of exchange. But I could be wrong.

While there are elements of cryptocurrencies that are really neat, they aren't fundamentally new. Rather, they fit quite nicely in the traditional 'bets and hedges' category. If you wouldn't bet all your savings in a zero-sum game like poker, neither should you do the same with cryptocurrencies. A bond or equity ETF is naturally productive, as is an investment in human capital. Consider them first.

6 comments:

  1. There is one thing that I think you have overlooked: Proof of Stake cryptos that collect transaction fees.

    In a proof of stake cryptocurrency, you need coins to create blocks (the coins are not used up in the process), you can then collect transaction fees from blocks by including transactions. If people gain some real value for making these transactions (privacy, censorship resistance, etc) then that real value flows back to coin holders (even if they don't personally stake) making it a non-zero sum game.

    Of course, there is also a zero sum gambling component. Since most PoS cryptos are low market cap altcoins with low numbers of transactions, this massively overwhelms the non-zero sum component. In addition, all of the large cryptos are PoW, where transaction fees are not related to coin value.

    But if you imagine something like the ETH/DAI system where the ETH is PoS and people use DAI for a significant number of transactions that give them real value, I could see cryptos moving out of the pure zero sun game category.

    ReplyDelete
    Replies
    1. Interesting points, Ben. I must confess I don't know much about proof of stake cryptocurrencies. But I am familiar with the argument you are making. I once made the same sort of argument for XRP: https://jpkoning.blogspot.com/2013/03/birth-of-currency-welcome-to-world-xrp.html

      Delete
    2. Hi,But the value obtained from transaction fees ultimately depends on the number of users using the bitcoin network or any other block chain network.They can move out of zero sum game only when major transactions happen through the corresponding network.Also,we can see various banks like jpmc(jpm coin) and other central banks trying to start their own crypto currencies so, I guess this should make some difference in the position of cryptos as they are backed by these banks.

      Delete
  2. An old sailing friend sent me this photo of grandchildren on Yap with real hard money: https://scontent-sjc3-1.xx.fbcdn.net/v/t1.0-9/66467529_10216401367480937_6490175783074529280_n.jpg?_nc_cat=101&_nc_oc=AQmu2RMv-2ucoGAi6NV_SmbZnEDZWMh1yFHyzJ_ZyajJy5t1CS6inKDfDM6O_DB7ro0&_nc_ht=scontent-sjc3-1.xx&oh=872ce918677a5a02a734ca8f683aba3e&oe=5DBF9B26

    ReplyDelete
  3. I disagree with the statement "Sure, the supply of cryptocurrencies is fixed, say like copper." The supply of cryptocurrencies is infinite. They claim they can control the amount of the crypto created, but as we have seen software forks, and even clones of code while calling it by a different name all have the effect of making crypto very much not fixed.

    ReplyDelete
  4. Hi JP, really like your blog. Just discovered it....
    one comment.. i would challenge that derivatives is a zero sum outcome... In the academic literature as well in practice any practitioner can replicate the opposing "hedge" so as to replicate the expected/actual payoff...

    ReplyDelete