Monday, May 1, 2023

The myth of crypto exceptionalism

"It is widely recognized – including by a sitting SEC Commissioner – that existing SEC registration and disclosure requirements are incompatible with digital assets, which differ fundamentally from the stocks, bonds, and investment contracts for which the securities laws were designed and that the SEC traditionally has regulated. The SEC at a minimum must set forth how those inapt and inapposite requirements are to be adapted to digital assets. But the SEC has refused to do even that."

- Coinbase, Inc.'s petition for writ of mandamaus to the United States Securities and Exchange Commission, April 2023 [pdf]

The claim that Coinbase makes in the above quote is a perfect example of what I call crypto exceptionalism.

The idea of crypto exceptionalism begins with the belief that crypto assets are fundamentally different from any sort of financial asset that preceded them. And so existing laws and regulations are inappropriate, (or "inapt," as described by Coinbase) and new rulemaking is required. This sounds right, at least on the surface, but it's a myth.

Stepping back in time, what happened in 2009 with the emergence of Bitcoin was not the creation of a fundamentally new type of asset. Rather, a new sort of financial database emerged: the blockchain. Others blockchains soon followed Bitcoin. But these new databases host the very same breadth of financial assets that pre-blockchain financial databases have always hosted: scams, stocks, deposits, ponzis, loyalty points, derivatives, gambling, bonds, loans, coupons, voting rights, and more.

There is nothing new under the sun. Go back to Ancient Rome or 1720s Paris and you can already find all of today's financial assets in use, including those currently traded on blockchains. That's why financial regulations created long ago (the ones Coinbase decries as "inapposite") remain relevant; they apply to financial typologies that are essentially eternal.

If you haven't recognized it, what I'm describing is just the digital substrate agnosticism approach that I sketched out in my previous post. We shouldn't treat the digital substrate on which financial assets are printed, or hosted, as being overriding. A blockchain is just another digital substrate, with one or two quirky characteristics. Rather, it's the assets themselves that should be the determining factor in the application of financial law.

If Coinbase's crypto exceptionalism is wrong, and changes in database technology do not create entirely new assets and thus don't merit new rules, what Coinbase does deserve is the following: rule makers need to show how the existing frameworks built to govern our eternal financial typologies map onto these new databases. While there's not much difference between an Azure or Oracle database, blockchains are quirky enough to merit a few extra pages of guidance, and perhaps a bit of tailoring where appropriate.


  1. I will expand on the reply I posted on Twitter so that readers on the blog may get a different perspective. I believe the claim made in this blog post is wrong.

    Up until the invention of public blockchains, all financial products had one thing in common: as a user / trader / consumer, it was impossible to inspect the inner workings of a given product. For example, bucket shop operators were only pretending to execute orders, and were falsely reporting on actual trades.

    This is fundamentally the reason for the emergence of certain regulations that require registration, disclosures, monthly reports, audits etc.

    But on public blockchains, there are constructs like permissionless automated market makers, exchanges which are not controlled nor upgradable by any single intermediary, whose inner workings are open for anyone to see and inspect, and where trade execution is guaranteed and verifiable atomically.

    What you are claiming in this blog post is equivalent to saying "it has always been possible for anyone to inspect the balance of a bank in real time", or "it has always been possible for anyone to verify correct execution of trades on the stock exchange in real time". It immediately becomes clear that this is false.

    Since such decentralized and transparent financial infrastructure has never existed, we can see that that blockchains do not host " the very same breadth of financial assets".

    Take for example the fictitious asset called DEX, associated with an on-chain decentralized exchange. Let's assume that holders of the DEX token are entitled to receive a portion of the fees generated by the exchange. By your reasoning, this is just the same as securities which have always existed. But one cannot ignore the fundamental difference that the decentralized exchange is transparent, and so is the fee accrual. There is no room for hiding information. It follows naturally that this is, in fact, not the same as existing stocks.

    This does not mean that existing regulations do not apply to *any* part of the current crypto ecosystem. In particular, exchanges like Coinbase are centralized financial intermediaries and even if one dislikes existing regulations, there is no reason why they should receive special treatment.

    What this means is simply that it is important to recognize that there are, in fact, new types of financial infrastructure and assets that exist on public blockchains, and that these merit new regulations.

    1. A DEX is an exchange where securities can be traded. It's a very old institution on a different substrate. (A DEX token is a claim to an exchange's earnings; also a new version of a very old financial asset.) Putting an exchange on a different substrate introduces some quirks, and regulators should look at DEXs to figure out how exactly to tailor, or apply, existing rules, but the same general principles that have applied for centuries to exchanges like the NYSE also apply to Uniswap.

    2. You're avoiding the core of the argument. Yes, exchanges where assets can be traded have always existed. Yes, a token which is a claim to an exchange's earnings is a slightly different version of something that has always existed. But the fundamental reason why the previous versions were subject to certain regulations was the *impossibility* of transparency. That is, it was required for a single operator to run the exchange and distribute stocks, and so we needed certain guardrails to prevent fraud through obscurity (i.e. unfair execution, lies about earnings etc). For an on-chain (truly) decentralized exchange though, there is something that has *never* existed before: real time transparency, execution guarantees, and decentralization. The problem that led to the creation of regulations in the first place does no longer exist, so there is no reason to think that existing rules should apply. You are portraying exceptionalism as a belief. It is not a belief. If you carefully consider how *some* of these systems work, you will realize that there is in fact something new that merits new rules. What you are arguing in this blog post is that no new rules are needed because there is *nothing* new. I have just demonstrated that this is false.

    3. @Experience - The thing that really jumped out to me in your original response above are the words "real time". Is that fact that public blockchain networks allow anyone to inspect the execution of trades of tokens on these networks in real-time sufficient to claim that the tokens themselves are a new type of financial asset? I'm not sure how that makes sense.

    4. @Gagan You're right, this is not sufficient. Some of the assets traded on such decentralized exchange actually neatly fit in the existing framework of securities regulations. My comment is about two related things
      1/ The infrastructure itself: any regulation that finds its origin in the opaque nature of legacy financial infrastructure should be revisited in the light of this newfound transparency.
      2/ Certain assets that are themselves transparent. For example, in the above example, some assets act like stocks, but the entitle you to a portion of earnings from the transparent decentralized exchange. The earning distribution is autonomous, transparent, and immutable. There is no board, and no possibility of further stock issuance. Assets of this type also exist for some on-chain lending markets and end up being traded on decentralized exchanges.
      So the argument is that some (but not all) of this infrastructure does not fit in the existing regulatory framework, and that neither do some (again, not all) of the crypto assets. The argument is *not* that merely by putting an asset on the blockchain, it does not anymore fit in the existing regulatory framework. For example, if one were to tokenize $AAPL on the Ethereum blockchain, it would still be a security, even if it could be traded on a decentralized exchange. Or another example, a new company can't just issue a token instead of a shares without registering, and say that just because their token is traded on the blockchain, it is not a security.

    5. "You're avoiding the core of the argument."

      I grant you that DEXs have certain novel features and regulators should tailor existing rules appropriately. That is, they might have to change the way a particular rule applies or exempt the exchange altogether from a certain rule when the exchange can comply with the intent of the existing rule in alternative ways.

      But the idea that the transparency of DEXs somehow solves all the problems for which existing regulation was designed is far too idealistic. For example, DEXs don't comply with rules surrounding appropriateness & suitability, they don't have listing standards, and do nothing to prevent insider trading and deceptive trading.

  2. So in your view, is the bitcoin asset a “chain letter” on a new sort of financial database? If so, shouldn’t the SEC make a case for this?

    1. Yep, in my view bitcoin is an old-fangled chain letter that's hosted on a new-fangled blockchain.

      You could therefore make the argument that the SEC should prohibit bitcoin since chain letters are illegal. The other argument I've made in the past is that bitcoin is a relatively 'fair' or innocuous chain letter; what gets paid-in always gets paid-out, and that could put it in the same category as a gambling game.

  3. gm,

    Thanks for the post.

    I agree that crypto assets are not by default unique or warrant special regulation. I can tokenize Apple Stock, that does not make Apple stock no longer a security. So as you rightly point out, crypto assets can represent all sorts of things and therefore fall under all sorts of regulatory regimes.

    However, a subset of crypto assets are "crypto native". These assets (like ETH and BTC) solely represents cryptographically-enabled rights to use a permissionless digital network. This is a new type of asset... you can certainly argue *how* unique it is but I don't think you can argue that it is so exactly the same as other assets that no new regulatory considerations are present. I also believe those regulatory considerations extend to the essence of what it is... not just that it is issued on a blockchain.

    I think Rodrigo did a good job of summarizing a few of the unique aspects here (

    So if a security is tokenized on a blockchain, of course, the SEC should still regulate it. But what about ETH itself? And other tokens like it? Maybe the SEC should regulate it but those regulations do need to be adopted... or the CFTC, or even the CFPB! I'm agnostic to the regulatory agency but would like understanding of the unique attributes of a crypto native asset vs. any random token put on a blockchain.

    1. What makes crypto assets complicated is that they often combine two or three functionalities for which there are already existing regulatory frameworks. So in the case of Apple stock, it's just a security. But tokenized Apple is both a security falling under SEC jurisdiction and also a payments object, which entails compliance with anti-money laundering frameworks (i.e. FinCEN).

      This layering of functions also applies to bitcoin and ether. I see both as a type of spontaneous gambling game, and thus subject to gambling laws, but because they are transferable they also fall under FinCEN jurisdiction, and finally they function as a sort of coupon in that they can be spent to get access to digital network. This multi-functionality makes it difficult to draw bitcoin and ether into existing frameworks, but doesn't mean we need new ones.

  4. "it's the assets themselves that should be the determining factor in the application of financial law."

    I don't think it's fruitful to try and analyse financial assets outside the context of its underlying database. Sure, the use-cases of today's blockchain tokens have all in some way been tried on some kind of existing database. But the universalizability of the blockchain (along with its immutability and in the case of Ethereum, programmability) solves important trust and coordination problems (governance) for the tokens issued on it. They're not "fundamentally" different, but they're different enough that applying existing modes of political regulation harms more than it helps.

    1. "...but they're different enough that applying existing modes of political regulation harms more than it helps."

      Imagine that Apple issued shares on Ethereum. Certainly you're not arguing that tokenized Apple stock is sufficiently different from regular Apple stock that existing securities laws should be inapplicable, since they harm more than they help.

  5. Thanks for sharing this valuable information. Crypto assets are fundamentally different from any sort of financial asset, have no collateral like currency with gold collateral