The Handwavy Technobabble Nothingburger

At this point pretty much every economist worth their weight in salt has given the public fair warning about the financial absurdity of crypto assets using the well-known basic economic arguments against the faux currencies. However economic crypto scepticism has to go hand in hand with a deeper understanding of why the technology doesn’t work as its advocates claim, in addition to the legal and regulatory arguments against its existence.

There’s a simple inescapable truth at the heart of technical crypto scepticism that almost all software engineers intuit at some level:

Any application that could be done on a blockchain could be better done on a centralized database. Except crime.

The essence of the financial arguments against crypto assets are quite easily summarized. As I previously described, crypto assets have no claim to be currencies because their deflationary properties and volatility don’t fulfill the theoretical or even practical function of money. They aren’t commodities because they have no non-circular economic use case. There is a somewhat coherent proposition that crypto assets are effectively unregistered securities contracts, basically like stock in an empty company that doesn’t do anything except promote the sale of its own stock. Historically these investments would have been called “Blue Sky Contracts” in the era before the Uniform Securities Act of 1956 outlawed such things. And then there’s the claim that crypto assets are a piece of performance art about libertarian politics, but this is an unfalsifiable proposition.

Despite their financial incoherence there are effectively three technology buckets you can put most of these crypto asset schemes into:

  1. Memecoins - Investment schemes that exist to perpetuate some narrative based on an internet or political meme and whose purpose is “number go up”. Examples: Doge, Shibu Inu, Bitcoin, Litecoin, Hex, etc
  2. Progcoins - Investment schemes which host other investment schemes and typically allow the execution of so-called smart contracts. Examples: Ethereum, Solana, etc
  3. Stablecoins - Investment schemes which maintain value against real world assets and used as a medium of exchange for extra-legal transactions. Examples: USDC, USDT, etc

Memecoins are pure greater fool investments, they’re basically a hot potato that people trade hoping to offload it on someone dumber than them who will pay more for it. And the implicit assumption behind the terminal value of these assets is that there’s an infinite chain of fools who will keep doing this forever. Nassim Taleb deconstructed this concept from a quantitative finance perspective in his whitepaper but nevertheless these assets persist because people behave economically irrationally and like lighting money on fire and dumping it into memes regardless of financial sanity. Meme coins like dogecoin exist simply for people to gamble on a fantasy about talking dogs, and bitcoin is a meme token for gambling on a fantasy about living in a cyberpunk dystopia. At the end of the day, memecoins are not that economically distinguishable from Ponzi schemes.

Progcoins are manifestations of what some of us programmers call decentralized woo woo, these projects claim to build all manner of programmatic applications. Yet when you dig into the details of such claims they’re very hand-wavy appeals to things that either don’t exist yet or are thinly veiled gambling schemes and outright scams. After twelve years of these technologies existing (roughly the same age as the iPhone) there is basically only one type of successful crypto business: exchanges which exist to trade more crypto. But the heart of this issue, and why there’s no other success stories, is because smart contracts tenuously look like a good idea until you actually try to build anything real that has to interact with the non-blockchain outside world. At which point they become too brittle, insecure, or strictly inferior to a centralized alternative.

In database terminology smart contracts are stored procedures that run one of the various incarnations of distributed databases these technologies are built on. In theory they act somewhat like self-automated vending machines but for more complex user interactions. In practice they act more like self-automated bug bounties which typically explode violently when certain exploits are issued against the coded logic, and at which point they spill all of the coins locked up in the contract. These disasters happen about two or three times a week now because coding at that level of correctness required in a Javascript-like language with loose and ill-defined semantics is near impossible. When a contract does finally meet its end, the only recourse is begging or threats to return the stolen tokens. However it’s unclear that “stolen” is the right word because the contract was simply behaving exactly as instructed and therein lies the core reason why “code is law” is an absolutely rubbish idea.

The second absurdity at the heart of smart contracts is their dependence on external data sources to function, the so-called “Oracle problem” is an intractable issue whereby these blockchain stored procedures must depend on data external to a blockchain in order to allegedly perform some business function. If a contract is modeling some sort of derivative contract then it depends on the price of the underlying asset, which it will have to pull from a price feed from Bloomberg. To check if the counterparty to the derivative has posted collateral it will have to pull out to query the balance of an account at a high street bank for one of the counterparties. To check if the counterparties are allowed to trade with each other they have to check whether either of them is on a sanctions list. So then by the time you’ve folded Bloomberg, Barclays and Uncle Sam into the trust boundary of your smart contract there’s very little point to saying this process is decentralized anymore, and begs the question why even construct this Rube Goldberg machine when it could be better done as a simple program running on a centralized server. It would be far more sensible and efficient to just build a web app that uses Stripe for payments. That is unless your business model fundamentally depends on selling unregistered securities or breaking the law.

And then that leads us into the third class of tokens: stablecoins. Stablecoins at face value might have some claim to have moneyness property. They are in essence a derivative of a national currency, usually a US dollar derivative that is issued on a blockchain and maintains a fixed stable value rather than being a speculative investment. Stablecoins are allegedly backed up one-to-one by reserves which should equal the total amount issued. You buy a stablecoin dollar effectively in the same way one buys chips at a casino, except stablecoins are used to gamble at offshore crypto exchanges who can’t get stable banking access because no regulated entity will touch these jurisdiction-hopping externational scofflaw casino boats. Stablecoins thus fulfill the customer “need” to arbitrage money transmitter regulation and move money to entities that exist outside the normal regulatory perimeter.

The casino chip analogy is accurate however unlike a casino, stablecoin issuers are not required to redeem tokens for real money and have no legal requirements to maintain reserves or even report on their contents. It’s a pretty good racket printing your own counterfeit dollar derivatives, and in practice many investigative financial journalists allege that some of these issuers are simply absconding with customer money and lying about their reserves. Stablecoin issuers are some of shadiest operators in an already rather shady ecosystem and many are widely believed to be outright scams that may meet the same fate as offshore Caribbean wildcat banks like Liberty Reserve.

Some people in technology think that stablecoins could be used to innovate in the banking sector and expedite retail payments. This almost makes sense, until you think about it for more than 10 seconds. Even if you had a completely legal and above-board stablecoin (which doesn’t exist today) you effectively have an institution which is for all intents and purposes basically acting as a bank, they take and custody customer funds and have enough liquid reserves to prevent a run on the coin and honor withdrawals whenever the customer needs. The Biden administration looked at this problem and came to the same conclusion, they should be regulated exactly like banks and be required to have FDIC protection on customer money, post collateral and be plugged into the Federal Reserve like any other bank would. At that point, yes, most of the consumer protection problems are mitigated for this kind of business but it begs the fundamental question: Why even bother?

A stablecoin bank would be subject to exactly the same FinCEN and OFAC money movement restrictions and compliance checks as banks; so know your customer gating, counter-terrorism financing, sanctions enforcement, and anti-money laundering enforcement. And these compliance requirements are the almost always the bottleneck consumers may encounter when doing cross-border transactions, and it’s not a technology issue. Nothing about stablecoins is either necessary nor desirable, and any alleged improvement these systems may offer at the moment are purely illusory and derived only from the unstable situation that they temporarily inhabit a yet-unregulated shadow banking system that is either non-compliant or entirely scofflawing. A regulated stablecoin bank is just a bank, but with a core ledger built on a terribly inefficient and bizarre piece of software not built for that purpose. All this while guzzling entire nation states worth of energy for no reason. Using inefficient blockchain as core banking software makes old legacy core banking solutions like Jack Henry look like a Ferrari by comparison. Our European allies all built extremely reliable real time payments like SEPA that work marvelously and they didn’t need any stablecoins.

Yet all of these technical arguments circle around a deeper truth: a technology which is purpose built to circumvent and arbitrage the regulatory perimeter cannot be brought within the perimeter without destroying its core claim to value or irreparably crippling it. Until proven otherwise it seems like the goal of the crypto ecosystem is to build an enormous unregulated casino with a crazy party scene. Along with a large lobbying arm to keep the musical chairs party going long enough with the hope of a government bailout through empty appeals to “American innovation” when the pyramid inevitably collapses.

I’m not alone in believing in the fundamental technical uselessness of blockchains. There are tens of thousands of other people in the largest tech companies in the world that thanklessly push their organizations away from crypto adoption every day. The crypto asset bubble is perhaps the most divisive topic in tech of our era and possibly ever to exist in our field. It’s a scary but essential truth to realise that normal software engineers like us are an integral part of society’s immune system against the enormous moral hazard of technology-hyped asset bubbles metastasizing into systemic risk.