Crypto is not money, money is not utility, utility is not real

The cryptocurrency exchange FTX (alongside the separate-but-actually-the-same trading firm Alameda Research) was revealed to have fraudulently pissed away billions of customer funds in November ‘22. This is provoking a lot of soul searching in the crypto community, where FTX was purported to be one of the “good guys” who was supposed to shame the scammers, not join them. The CEO Sam Bankman-Fried (“SBF”) was also a poster child for the “Effective Altruism” (“EA”) movement that focuses on charitable giving by first optimizing wealth so you have more to spend. Many of EA’s causes were hugely funded by SBF; how culpable are they for what he did to get them? Does this reflect negatively on EA as a whole?

I’m not going to focus on who deserves how much blame for what happened; that’s a hard thing to do as an outsider to all of these scenes, especially during the fog of war we still have around the finer details. Instead, I want to ask: why were so many people dealing with FTX and SBF such easy marks?  That isn’t hard at all for me to write! (I once did it in my sleep.) My brother-in-arms Neil has also explored in what sense is utility actually, like, a thing? This moment where billions of dollars accumulated by an avowed utility-maxer has proven to actually be nothing at all is an excellent time to repeat ourselves: no, really, is utility actually, like, a thing?

Crypto is not money, money is not utility, utility is not real. The impact of the FTX crash gives us an extremely resonant example to draw from, so we’re going to use it. But this post isn’t really about crypto and money per se; it’s about a certain way of seeing that comes up a lot in discussions of crypto and money, and the cases when using that way of seeing makes you a sucker. As you read the first two sections, focus less on the literal treatment of crypto and money and more on seeing what the examples have in common.

We’ll use the article FTX's Balance Sheet Was Bad from Matt Levine to give us a bit of background. From the leaked balance sheet that was being shopped around to potential investors, we saw that FTX’s single biggest asset was 2.2 billion “worth” of Serum, a cryptocurrency used for the Solana blockchain that FTX heavily boosted and invested in. Only a small amount of Serum can actually be traded at any given time - 97% of it is held by “core believers” who are often waiting for the token to “unlock” before they’re liquid at all. And crucially, this means that FTX did not actually spend billions of dollars to get this Serum; they got the tokens as early backers of the Serum protocol, to show their commitment to grow its “long term value”. 1 Matt does a good job of expressing how indignant we should be seeing this balance sheet:

Like, $16 billion of dollar liabilities and $16 billion of liquid dollar-denominated assets? Sure, great. $16 billion of dollar liabilities and $16 billion worth of Bitcoin assets? Not ideal, incredibly risky, but in some broad sense understandable. $16 billion of dollar liabilities and assets consisting entirely of some magic beans that you bought in the market for $16 billion? Very bad. $16 billion of dollar liabilities and assets consisting mostly of some magic beans that you invented yourself and acquired for zero dollars? WHAT? Never mind the valuation of the beans; where did the money go? What happened to the $16 billion? Spending $5 billion of customer money on Serum would have been horrible, but FTX didn’t do that, and couldn’t have, because there wasn’t $5 billion of Serum available to buy. FTX shot its customer money into some still-unexplained reaches of the astral plane and was like “well we do have $5 billion of this Serum token we made up, that’s something?” No it isn’t!

One simple point here is that FTX’s Serum holdings — $2.2 billion last week, $5.4 billion before that — could not have been sold for anything like $2.2 billion. FTX’s Serum holdings were vastly larger than the entire circulating supply of Serum. If FTX had attempted to sell them into the market over the course of a week or month or year, it would have swamped the market and crashed the price. Perhaps it could have gotten a few hundred million dollars for them. But I think a realistic valuation of that huge stash of Serum would be closer to zero. That is not a comment on Serum; it’s a comment on the size of the stash.

FTX’s stash of Serum could be evaluated at $2.2 billion sitting on the balance sheet, but would have been closer to zero in a sell-off. The vast asymmetry between the stashed token and the trading supply puts a huge amount of pressure on the “value” of the stash. Put a pin in that inconsistency for now and let’s change topics to wash trading.

At its heart, wash trading is moving your money around and lying about why you’re doing it. For example, you might give your buddy $105, then ask him to give $100 of it back to you publicly in “exchange” for an old boot. Anyone watching the exchange might think that old boots go for around $100, which will benefit you if you have a lot of old boots: you can sell them for $50 to people who think they’re getting an absolute steal, only to learn that no one actually wants to buy an old boot unless they’re doing it with your money. And there’s nothing limiting you to only one co-conspirator. You can imagine an entire ecosystem of people trading money and old boots back and forth until a single poor sap exchanges lots of real money for old boots, at which point everyone can stop pretending they want to buy an old boot and exit the scheme. This idea of an entire fake market designed to trick one person into one transaction isn’t even hypothetical: confidence tricksters call this a “big store” scam. This has been illegal in the United States since 1936, but the anonymity of crypto wallets makes it extremely difficult to figure out just how much wash trading is going on. 

What ties these together? You have a token — Serum or whatever — being exchanged for some amount of dollars. If you witness a few of these trades, it’s easy to conclude that Serum “has value”, and is worth the dollars that you see it being traded for. But asymmetries between stash and trading supply and the existence of wash trading are two very good reasons not to do that. They demand that your evaluation of worth is more contextual, instead of ascribing anything inherently to the tokens.

But hold on -- this header is “crypto is not money”. It’s not exactly a crypto-exclusive thing to have a stash way bigger than the amount of the thing being actively traded, and while wash trading is formally illegal in the US, that doesn’t mean it’s non-existent. What does it mean to say that these deficiencies with the concept of “value” make crypto not money? Is it just a difference of degree, or is there something more profound happening here? It seems like a question that can’t be answered without getting into a fussy philosophical discussion of what money is.

Hey, what the hell is money, anyway?

Let’s shake off any existing baggage and imagine a new currency from scratch. There’s a king who claims domain over some rich, fertile plains interspersed with unfriendly, difficult-to-traverse crags. Neighboring kingdoms want to steal the bounty of the plains. The king controls a professional standing army who train at soldiery year-round instead of farming, and takes some of the farmers grain as a tax.The neighboring kingdoms are the sort who would take everything, so the farmers acquiesce to having some grain taken if it means safety. 

But the king’s castle is in the center, while the forts are all along the border. It’s awfully inefficient to levy the taxed grain, catalog it in the castle, then send it back out all over the kingdom. The king would much rather have the soldiers get the grain from the farmers directly. So he mints some brand new coins with his face, gives them to his soldiers, then his tax collectors spread the news: this year, instead of asking for grain directly, we’ll be asking for coins. Now the citizens have to give the soldiers something they want in exchange for their coins. The coins symbolize the claims on the farmers' labor the soldiers demand in exchange for their work in preserving stability. 

Things are different up on the crags, though. They have patchy soil and they’re a pain to reach, so the tax collectors only went up there out of obligation, and didn’t expect much tax when they did. The crag people barely think of themselves as belonging to the kingdom, limiting themselves to paying the odd scrap of grain as a perfunctory gesture of submission. Now they hear that the tax collector wants coins instead. But soldiers don’t bother climbing the crags, and they truthfully claim that they have no coins to give and continue handing over a pittance of grain instead. The king doesn’t like this, exactly, but doesn’t want to start a civil war over it.

In the great markets of the plains cities, coins are exchanged for grain every day. A diligent observer of the markets could catalog the average amount of grain per day and record it. And if we were to ask “how valuable are these coins?”, one might think you can just point to that record and say, these coins are worth so much grain. But that’s the value of the currency in exchanges where it’s used. The value as a currency depends strongly on how many people live in the plains and how many live up in the crags. It depends on the continued stability of the king's regime and the particulars of the tax collectors. It depends on the timing of when soldiers get the coins and when the harvest is available. 

It’s easier to see those limitations in this toy example with one meaningful exchange and one king, but money is always about the diversity of the exchanges available to do with it and the normalizing force governing it. Let’s bring in a long quote from Ian Ground’s The relentless honesty of Ludwig Wittgenstein:

Say that we become puzzled about money. Here is something that people deeply desire, spend and risk their lives acquiring. People are “worth” so much money and so on. But perhaps we are struck by the fact that coins and notes are, in themselves just worthless bits of metal or paper. How can they have value? (Note that we have already slipped, even at the moment we first become puzzled, into thinking of “value” as a kind of property something has.)

Imagine that someone replies like this: it is true that actual cash is arbitrary – just stuff. What matters is that cash is backed by something that really does have value. The “promise to pay the bearer on demand” on UK notes. The gold in the bank is what really has value. The money is just an outward sign of that true value.

But gold is also just a kind of metal. Why should it have value? The same question we asked about the cash can now be asked about the gold.

Someone else might interject: gold is rare and hard to acquire. That’s why it has value. But lots of things are rare without being valuable. And in any case, no one actually trades in their money for gold. Banks won’t even let you do that. Yet we go on treating the money as valuable.

Here of course we will want to say this: actual money (coins and notes) isn’t intrinsically valuable. What matters is only that it is in fact used in trade and exchanges. The value lies in the use of the money. It’s not that the exchanges use money because the money has value. Rather the money has value because the exchanges have value. Or rather what we mean by monetary value is made manifest in and through the activities of exchange and the myriad things we do with money. And once we see things that way round, it will now seem rather strange to say that money is just worthless stuff. It looks that way and we became puzzled in the first place only because we tricked ourselves into separating out the notes and coins from their use in exchange. Our problem was how to explain how certain stuff – notes and coins – had value. So we started looking for another kind of stuff to carry that value. That is, we already committed to a particular view of what an explanation would look like. The solution was to change our view of what would count as an explanation or indeed whether one was actually needed at all. We solve the problem when we dissolve the source of our puzzlement.

The money has value because the exchanges have value. Crypto is not money because crypto is exchanged less than money is. (Crypto communities have the meme phrase “HODL” for those who intend to buy-and-hold indefinitely.) Crypto has some value in exchanges that aren’t possible without it (the canonical example for crypto supporters is escaping draconian currency controls or inflation in your home country; the canonical examples in the news are ransomware and child pornography). But mostly the value of a cryptocurrency on any given day is socially denominated in USD; mostly the crypto titans aren’t using their crypto to perform more and more valuable exchanges but are just making some arbitrary numbers go up.

“But don’t normal billionaires do the same thing with money? Not performing more and more valuable exchanges and just making their numbers go up?” Yes! This is exactly the parallel I want to draw! Crypto isn’t money because of its relative lack of use in exchanges; but money itself often has these layers of abstraction between it and the actual exchanges. Crypto is often a way to shift make-believe numbers around before dumping them back into actual money, money is sometimes a way to shift make-believe numbers around before dumping them back into actual valuable exchange. Think of points along a continuum where one end is purely self-referential scams and the other end is purely valuable exchanges. Crypto is significantly closer to the scam side than money is, but money is still quite far from the pure exchange side. It’s these valuable exchanges that we’re trying to point to with the idea of “utility”. And yes, the exchanges are real — but that doesn’t mean we can get to the end of the continuum and still end up with something that acts like a currency. 

I have to be careful here not to drown in an endless sea of caveats. “Utility” is sometimes meant as nothing more than “a word I use to describe exchanges both parties found beneficial” and “effective altruism” is sometimes meant as nothing more than “I like to have evidence charities I donate to are doing a good job”. These are basically fine and inoffensive statements. The idea of utility we’re pushing against here is something like “the goodness of an exchange can be measured against the goodness of other exchanges, and the total utility of a series of exchanges can be maximized”, and the idea of effective altruism we’re wary of is something like “good charitable giving is that which maximizes utility.”

The critical point here is the difference between noting the real value of a beneficial exchange being done and looking at catalogs of potential exchanges through the lens of an imagined currency that one can spend freely on any exchange in the catalog. As we’ve just seen, denoting things as currencies adds a layer of cruft and abstraction between you and the valuable trades; imagining yourself spending a hypothetical currency brings along much of that same baggage. Money and power generally compound, giving you more of them over time based on what you already have. Exchanges are obviously real and obviously valuable, so it’s easy to think utility as a currency pegged to the “exchange standard”, also obviously real and obviously valuable. Then all your metaphors at hand tell you to accumulate interest; that is, not exchange (the valuable part). And just like Serum, the bigger the imbalance between what you’re holding and what’s being used, the more likely what you’re holding actually rounds to zero.

Think of Leorio from Hunter x Hunter. On the one hand explicitly talks about earning to give when he describes the death of a childhood friend:

“It was a treatable disease…the problem was that the operation costs a fortune. I was naive! I thought I could become a doctor…I wanted to treat kids who have the same disease, and be able to tell them it was free of charge! Then I could have told his parents, too. That was my dream…what a joke. Turns out to become a doctor, you need even more money! Got it? The world runs on money, so I want money!

Leorio ends up becoming a pro Hunter (something like a professional mercenary) to fund his way to med school. Money was an intermediate step in the road towards his goals. But what if he decided to stop being governed by mere sentiment, and instead tried to maximize utility for people with that disease? Wouldn’t it be better not to focus on becoming a doctor individually, and instead found a successful business to get money to pay off several doctors? SBF himself was advocating that way of thinking:

Let’s say Leorio takes this advice and founds Leorio Co, a company focused on maximizing revenue so he’s able to give more long term. In his search for good ways to make money, he invested heavily in SBF’s FTX, one effective altruist working with another. Well, then he’d be bust and incapable of helping anyone. Or if Leorio had been involved in a scandal, or if the price had dropped due to fraud, or any number of things.2 The risk of going bust wasn't unknown to SBF, who mentions it explicitly in his Twitter thread: 

But trying to make “the play in EV” when it risks a significant chance of going bust is the exact same mistake as valuing Serum based on what’s being traded. You have infinite imaginary potential universes filled with vast reserves of “utility” and one actual linear time universe with the utility you can actually gain from the exchanges you perform. If you’re planning a modest improvement like “I’ll get the money to become a doctor and then I’ll be much more effective at trying to improve people's health”, you need to accumulate a little bit of money, time, and focus to achieve that goal. Thinking of that investment in terms of utility works fine, because becoming a doctor is a common, well-anchored thing to do, and you can trust that exchange to remain reasonably stable. But as your plans start ballooning into “I’ll gain an unfathomable amount of utility and spend it to change the world”, you start getting a bigger and bigger imbalance between your stash of imagined ability to do good and the actual exchanges you’re looking at when trying to evaluate how much your stash is worth. Imagine a disappointed Matt Levine writing: But I think a realistic valuation of that huge stash of utility would be closer to zero. That is not a comment on utility; it’s a comment on the size of the stash.

So let’s make sure to learn this lesson about imaginary tokens that are sometimes exchanged for value. Think of the idea of “maximizing utility” as instead “HODLing my stash of UtilityCoin”. With this frame, you’ll start asking the right questions: how do I know the exchange value of UtilityCoin to good outcomes will stay stable no matter how much I spend? Is the peg between UtilityCoin and good outcomes dependent on a single bottleneck? Are you evaluating your UtilityCoin by looking at a single good outcome that may have been manipulated in some way? And then maybe we can gently disentangle the EA proclivity for doing good from the EA proclivity of being an easy mark.