Three Things #43: November 13, 2022

Eth Isn't Money

For years I’ve found the “ETH is money” meme and, more recently, “Ultra sound money” obnoxious and distasteful but I could never quite put my finger on why. Ether never struck me as an especially good form of money. What’s more, most of the people I’ve seen use these terms clearly don’t know what they are actually supposed to mean, nor why ether would qualify, which I think is why they don’t always understand why these ideas sound silly to people who aren’t drinking the Kool-Aid.

In this issue I’ll talk a bit about the abstract concept of money, I’ll explore whether and to what extent bitcoin and ether are money (sound or otherwise), and I’ll share some of the lessons we’re taking from this analysis at Spacemesh.

The AI really hit the nail on the head with this one. Image created using Stable Diffusion.

Thing #1: Unsound Money

Let’s start with some basics. Lots of things have value. Apples have value. Art has value. Microsoft stock has value. But none of these things is money. Have you ever considered why?

In order for a thing to be good money, it has to have certain properties: it should be identifiable, portable, divisible, durable, transferable, verifiable, unforgeable, fungible, stable, and scarce. In the interest of time and space this is actually an abbreviated list, and I won’t go through each property in detail here—you can find that elsewhere—but it shouldn’t be necessary anyway because these properties are pretty self-evident. If you can’t divide something to make change, it’s not good money. If it decays or degrades over time, it’s not good money. If its value fluctuates wildly, it’s not good money. You get the idea.

It should be noted here that the property of money-ness is not binary. It’s not the case that a thing simply is or is not money. Rather, some things are better money than others. Throughout human history, people have used some pretty bad things as money—for instance, consider rai stones (not portable or divisible) and salt (not very durable, not easily portable)—but the overall trend has been from bad to good and the things that are commonly and widely used as money today have most or all of these desirable properties.

Okay, so what’s sound money? We tend to throw around the term without really understanding its origins or what it means. The “sound” in “sound money” comes from the sound a pure gold, silver, or copper coin makes when struck against a hard surface, a useful trick given that, since time immemorial, people with the authority to issue money gave into the temptation to debase it (a process that continues today).

Frustratingly, the term appears to have two definitions today, one accepted by the Austrian school of economics, one accepted by Keynesians. When Austrians talk about sound money, they mean a money whose supply is predictable and fixed, i.e., one whose issuance is not politicized and one whose value is therefore set by the market (rather than manipulated by political actors). By contrast, when Keynesians talk about sound money, they generally refer to money whose purchasing power is stable. That is to say, if one cuck buck bought a hot dog fifty years ago, it should be able to buy the same thing (or more) today.

These two definitions cause a lot of confusion, especially since they’re at odds with one another. One money cannot be “sound” by both definitions. Keynesians believe that the way you keep the purchasing power of money stable is by intervening in monetary policy: e.g., increasing or decreasing issuance. To an Austrian, such an interventionist money is anything but sound. And, in practice, a money with a fixed, predictable supply—such as bitcoin—actually tends to have a very volatile purchasing power, at least in the short to medium term (gold, which also has these properties, has been pretty stable over the very long term). To a Keynesian, such a money is anything but sound.

Both definitions of sound money are in fact a tall order for any currency today. Even the world’s most stable fiat currency, the Swiss Franc, has lost about 90% of its purchasing power over the last 100 years and its supply has inflated by around 100x over the same time period. Obviously, these figures are set to get much worse given today’s high inflation and the massive increase in monetary supply in response to the Covid pandemic. While gold might qualify on grounds of having both a predictable supply and a relatively stable long-term purchasing power, it’s not a good form of money because it doesn’t have several of the properties described above: it’s not easily portable, divisible, or verifiable, and thus isn’t widely accepted. So, Keynesians and Austrians would both agree that fiat money and gold are out as good, practical currencies. What about bitcoin and ether?

While bitcoin does have most of the above properties, it’s not stable. Its price and purchasing power have been extraordinarily volatile over the years. This makes it a poor store of value and a poor unit of account. To a Keynesian, it’s definitely not sound money. Austrians, by contrast, consider bitcoin one of the best forms of sound money today due to its predictable, fixed supply.

By “sound money,” Ethereum folks are referring to the disinflationary-to-deflationary nature of ether—but this is a contrived definition with no grounding in real economics. It’s not what Keynesians mean by sound money, nor is it what Austrians mean when they use this term. Ether is not sound money by the Keynesian standard since it’s not stable. A Keynesian wouldn’t even consider it good money in the first place for the same reason (money should be stable). And it’s not sound money by the Austrian standard since its supply is not predictable, not fixed, and rather politicized—more on this in a moment.

The Spacemesh Take: We can’t make our coins stable, but we can at least keep bitcoiners and Austrians happy by making sure that our supply is known, fixed, and unchanging. Spacemesh coins, like bitcoin, will make great collateral for stable coins built on top of the protocol.

Thing #2: Equity

With this understanding, we can return to the question: why aren’t apples, art, or Microsoft stock money? The first two are obvious: apples aren’t durable or scarce, and art isn’t fungible, portable, or divisible. But what about stock? This one is less obvious.

In fact I’d argue that Microsoft stock also satisfies most of the properties of good money. Some of them aren’t perfectly clear: e.g., whether and to what extent stocks in their modern form are portable and transferable could be debated, and it only recently became widely possible to trade fractional shares, but the properties are still basically intact. To understand what’s going on here, we have to go deeper and talk about trust.

At the end of the day, money is built on trust. The above properties are all manifestations of trust—trust that my money will still be there and will still be valuable tomorrow (identifiability, durability, stability, unforgeability, scarcity), trust that I can easily transact with it (portability, divisibility), etc.. We use dollars because we trust they’ll still have value tomorrow, and that the value won’t change too much day to day and month to month. We use them because we trust that others will, too, and that others will accept our dollars in exchange for goods and services.

An important component of trust is credible neutrality: the idea that no individual party can unilaterally break those guarantees, and that the rules are clear and everyone is subject to them. As Vitalik described, “a mechanism is credibly neutral if just by looking at the mechanism’s design, it is easy to see that the mechanism does not discriminate for or against any specific people.”

Money that’s issued in a credibly neutral fashion—say, when the Federal Reserve purchases Treasury securities on the open market to increase the money supply—will tend to lead to a money that everyone, or nearly everyone, trusts. (Of course, the dollar supply can be changed in many other ways that aren’t as straightforward or as neutral. I’m oversimplifying.) A money that isn’t credibly neutral won’t be trusted by everyone, since some people will see that it discriminates against them. This is the property that Microsoft stock fails to satisfy. Why?

First of all, because a small group of people—namely, the Microsoft board—can unilaterally decide to authorize and issue more stock at any time, and award it to anyone they like. Secondly, because the existing allocation is anything but credibly neutral. A tiny number of insiders and investors own the lion’s share. This is just how equity works: the founders hold 100% of the equity in the beginning and unilaterally sell it or give it out to the people they believe deserve it. To be clear, there’s nothing wrong with this. Equity is designed to solve a different problem than money! It’s not meant to be democratic, nor inclusive, nor credibly neutral. (Actually, MSFT isn’t the best example, because it’s a mature company that’s been around for a long time; a better example would be a younger company, but you get the idea.)

This is also how Ethereum works. The Ethereum Foundation premined 70 million ether and sold most of it prior to genesis. That amount today represents 60% of the total ether in circulation. To be clear, the Ethereum crowdsale was better than what most other chains have done. For one thing, anyone with bitcoin could participate. For another, many chains have had much larger premines, and sold fewer coins. But it doesn’t change the fact that 60% of the amount of ether outstanding today was sold to a few hundred lucky people over 42 days in 2014.

My main issue with the premine and crowdsale is the size and speed relative to the mining process. Proof of work mining is an excellent example of a credibly neutral distribution mechanism. Anyone, anywhere can permissionlessly participate using the right hardware, an Internet connection, and some basic technical know-how. It’s a gradual process that plays out over years. It’s naturally decentralized, since the required resources—cheap energy, hardware, expertise—are widely distributed, there are diseconomies of scale beyond a certain point, and new miners are always appearing to compete with incumbents.

By contrast, the Ethereum crowdsale was like a company’s founders issuing stock for themselves and selling stock to investors.

Bringing it back to money and trust, nearly all recent projects have done only private sales, or only sales to accredited investors, which are worse than what Ethereum did. In order for these projects to justify their absurd valuations—all before finding protocol-market fit, and most even before launching—they have to make the case that their native coin becomes adopted as money. Surely this is what they’re telling their investors.

But the thought that A16Z or some other big investor is going to own, say, 10% of money is absurd to the point of being laughable. If these projects and these investors don’t see why this is absurd then they have a serious problem—and they have another thing coming. There’s zero chance that any of these projects are going to achieve a “monetary premium.”

Would you trust a money if A16Z held 10% of it? Would you trust it if anyone held 10% of it? Think of the ways in which that whale could manipulate the markets, prices, and the money supply. To put things in perspective, 10% of the USD supply is 2.15 trillion dollars. We have enough trouble with out of touch billionaires; would you trust a trillionaire?

The Spacemesh Take: Spacemesh has no traditional premine as both team and investors are bound by four year vesting. After 10 years, the team and investors will collectively hold 25% of the coins, but issuance will continue for hundreds of years after this, and the amount will eventually fall to only 5%.

Thing #3: Proof of Stake

Ethereum successfully completed the Merge and transitioned fully from proof of work mining to proof of stake a couple of months ago. I’m on record as disliking proof of stake, something that by now I’ve spoken and written about at great length. I don’t want to beat a dead horse, so today I’m going to focus on one narrow aspect of proof of stake: the fact that it looks and feels a lot like equity.

As I described above, proof of work is truly permissionless and meritocratic and tends towards decentralization at scale. Miners that are new to the game actually have an advantage over incumbents because they’re using newer, more efficient hardware. As Lyn Alden describes:

Your ability to be a miner is based on your ability to put forth capital and find low-cost electricity. Rather than the entrenched miners having an advantage and increasing their advantage over time (as is inherently the case with proof-of-stake systems), newer miners actually have a technical advantage over existing miners in some ways because they buy the newer machines with more processing power per watt, thanks to Moore’s law, with no existing sunk cost. Mining businesses, old and new, are all constantly refreshing themselves with capital expenditures, making use of new cheap or stranded energy resources. Management quality and experience is critical, and economies of scale only get you so far.

By contrast, validators in proof of stake earn coins from their existing coins. There’s no need for ongoing resource expenditure, ongoing capital expenditures or innovation such as finding new, stranded energy resources. Lyn describes it in the following way:

It would be like a political system where you get a vote for every hundred dollars you have, and then also get paid a dollar by the government for casting each vote. Mary the high school science teacher with $20,000 in net worth gets 200 votes, and earns $200 from the government for voting. Jeff Bezos, with $200 billion in net worth, gets 2 billion votes, and earns $2 billion from the government for voting. He’s a more valuable citizen than Mary, by a factor of a million, and also gets paid more by the government for already being wealthy.

That’s not a system many folks would like to live in. Eventually it would likely consolidate into an oligopoly (if it wasn’t already), with a handful of multi-billionaires controlling most of the votes and ruling everything.

Indeed, if you have a system where the rich get rewarded for being rich, with little to no work required, over time it tends to centralize wealth. Those with more coins have more incentive and more ability to stake them; everyone else gets diluted.

There’s a place where oligopoly works well indeed: corporations. Shareholders get votes (and dividends) proportional to how many shares they own. Oligopoly doesn’t always produce optimal outcomes, but it is darn efficient, which is probably why corporations work this way: they need to make decisions quickly. There’s nothing inherently wrong with the oligopolistic corporate model, as long as one property holds: that it remains voluntary and competitive. Shareholders, employees, and customers should be free to do business elsewhere if they disagree with how the company is being run.

But that’s not how money works. Maybe we’ll end up in a world of many competing currencies that are all legal tender, but I doubt it: economics says otherwise. Good money is inherently centralizing and monopolistic. It has very strong network effects. And so you have no more ability to opt out of money than you have of opting out of breathing air. (You could move to another country, but that’s not easy, either, and countries also have strong network effects.)

This is why it’s so important that money be decentralized, democratic rather than oligopolistic, and credibly neutral. Equity is none of these things.

Equity and most crypto coins are costless to issue, like fiat money (and unlike gold). This leads to a problem: when money can be issued costlessly, those in control of the money supply will inevitably issue more of it to themselves, an especially pernicious political economic phenomenon known as the Cantillon Effect. As Adam Back put it, “[M]oney that doesn’t have a cost ultimately ends up being political in nature.” This is why it’s so important to bitcoiners that bitcoin be as unlike equity as possible, with a fixed, hard supply cap and issuance that occurs predictably and gradually via a credible neutral mechanism. No crypto coin without these properties stands a chance of being widely accepted as money for the same reason that Apple stock will never be money.

Bottom line: ether looks and feels even more like equity now after the transition to proof of stake. The Merge made ether issuance more political, and made it even less money-like.

The Spacemesh Take: Spacemesh doesn’t use proof of stake. We don’t use proof of work, either. We use a hybrid consensus mechanism with the best properties of both: proof of spacetime, which is permissionless like proof of work but as green as proof of stake.