Bitcoin is not the Future of Money
For all of recorded history gold has been money. Just as birds didn’t need a theory of flight before taking to the skies, humans didn’t need a theory of money to develop gold as a medium of exchange. Theories of money were developed thousands of years after the fact. In his 1912 book “Theory of Money and Credit”, Mises makes the case that:
“Whenever a direct exchange seemed out of the question, each of the parties to a transaction would naturally endeavor to exchange his superfluous commodities, not merely for more marketable commodities in general, but for the most marketable commodities; and among these again he would naturally prefer whichever particular commodity was the most marketable of all.”
In the natural world there are very few items that can serve as money. Beads and shells are not scarce, jewels are not fungible. Gold happens to be the most useful of many imperfect natural materials to function as money. In Mises’ words, gold is the most marketable commodity.
What about non natural items? Surely it’s conceivable that just as people engineered a purpose built flying machine, they could also engineer a better form of money. We may even get some new features that previously were not even known to be important, censorship resistance, anonymity, programmable scarcity, permissionless transfer, … Bitcoin has all of the features of traditional money and several other features that make it superior to gold as money. With each passing year it becomes harder to deny that bitcoin is taking over for gold as money. In March of 2019, “PlanB” published his now famous stock to flow model:
The price of BTC continues to neatly track the projections of the stock to flow model 2 years later.
If bitcoin is better than gold, then why is it not the future of the money? For that we must return to the words of Mises “among these again he would naturally prefer whichever particular commodity was the most marketable of all.” While the most marketable crypto is bitcoin today, in the future this is almost certainly going to be the native token of the proof of stake network that is running the majority of crypto applications. Bitcoin is unsuitable by design for building applications. Bitcoin script is limited. Very few serious applications can be built on a system boasting 10 minute latency and single digit transactions per second.
Which layer 1 proof of stake network is a good candidate to host the majority of crypto applications? We are still in the very early days of crypto adoption, comparable to the number of applications on the early internet in the days of dial up, so where we are today may not be a good indicator of where we will end up. The equivalent of broadband internet is coming quickly, and with it we will see an explosion of crypto applications. Layer 1 proof of stake networks are reaching the scale of transaction throughput and latency where interesting applications can be built on top of them. ETH 2.0 is being released slowly but surely. Avalanche is here today with 3,500 tps and sub second finality. Solana looks promising as well. Knowing that application popularity and consequent transaction rates will follow a power law distribution should inform us to prefer vertical scaling to horizontal scaling. Sharding and parachains are examples of horizontal scaling, while things like web assembly constitute vertical scaling.
We can even get more discriminating by fully understanding the value proposition of crypto networks. They offer higher efficiency by subjecting network operators to market forces. This requires supply elasticity in the number of network operators. Crypto prevents outsized rents from being collected by network operators, and it does this through supply elasticity. Anyone can become a miner in bitcoin, thus miner profitability is always existing at the margin. We can therefore eliminate from consideration any DPOS network such as EOS with its 21 block producers on this principle. We can also eliminate unscalable consensus algorithms such as tendermint which is O(n²) as found in DOT and ATOM. We can all recognize the abomination of Jeff Bezos as the world’s richest man while employees numbering in the 100’s of 1000’s toil for low wages in Amazon fulfillment centers. Would we be better off if we merely divided that fortune among 21 co-CEOs as in EOS or 1000 milliBezos as in DOT? No, there is no magic fixed number that equals decentralization, we need supply elasticity, no matter the number.
While I have a strong opinion on the supply of network operators, I’m not quite as certain on the supply of coin in the ideal crypto network. The stock to flow model implies that a fixed supply would be ideal money, but the question of whether a deflationary currency is better than a fixed supply is still open. Intuition suggests that a fixed supply is likely superior to deflationary as the time value of money is best expressed as an interest rate.
Armed with these ideas we can clearly see a huge opportunity. Some crypto network valued at less than $200 billion will eventually become money with a marketcap on the order of worldwide GDP at $100 trillion. The list of coins that can meet the requirements is very short: ETH, ADA, SOL, AVAX, ALGO. It’s quite likely one of those or something new that will increase in value 1000 times or more over the next 10 years as it displaces both gold and bitcoin to become money.
