Letting the horses out of the stable[coin]

Build Canada has been really cranking out the policy memos, and the most recent one is going back to an old classic from the crypto bubble of 2022, stablecoins. If you're not familiar, stablecoins are cryptocurrencies that are intended to actually function as currency by being a stable store of value (what a concept!). Since the general concept of cryptocurrencies provides no mathematical magic to support this (they've tried this with 'algorithmic' stablecoins, it went great), this is typically done by pegging the coin to some fiat currency and backing the coin with actual fiat reserves to maintain that peg. Whether this is actually done, or is propped up by accounting fraud, is left as an exercise for the implementer.

Mr. Seif, who wrote the stablecoin memo for Build Canada, suggests that "Stablecoins backed by the Canadian dollar will make transactions faster, cheaper, and more efficient," which if it were true, would be fabulous! Who wouldn't want to make transactions faster AND cheaper AND more efficient. Let's examine the claims more closely.

The first benefit suggested is that "stablecoins allow Canadians to hold and transfer money directly, without relying on banks or payment processors." This is technically true, in that they do allow Canadians to hold money directly, much like you are allowed to have a bunch of cash in your wallet and pay for things that way. However, the ergonomics of directly transacting on-chain are pretty awful, and so the vast majority of people's interactions with cryptocurrencies happen through exchanges like Binance, Coinbase, and Kraken. In this way they are in fact very much like digital representations of fiat currency, and we'll revisit this in relation to some of the ostensible benefits proposed later.

Next, we're told that "smart contracts" allow you to make complex financial processes more transparent, cheaper, and faster. At this point we have descended into pure Disneyland thinking. Smart contracts have been around for a long time now, and these vaunted benefits have simply not appeared. What has appeared is ample evidence that smart contracts are not any smarter than other software, and as such are subject to the exact same issues around bugs, security exploits, and straight-up malware that other software has. Are these hurdles insurmountable? No. Are they a big problem if your first selling point was letting the general populace directly transact without a "middle-man" (aka subject matter expert who can vet the legitimacy of the contract)? I would say yes, though Mr. Seif doesn't seem to agree.

Next, we get into the 'real-world financial challenges' stablecoins will solve, starting with "high fees and slow transactions". Stablecoins, it is claimed, provide low-cost, instant global payments. They even cite a research paper from the Bank of Canada! Unfortunately, if you actually click through and read the abstract of the paper, you'll find that it explicitly states that their findings suggest stablecoins are "costlier relative to traditional payment arrangements". Whoops. In fact, their general finding is that stablecoins "provide less consumer protection for fraud, present higher risks to the payment system and to efforts to combat financial crime" and "do not currently serve as substitutes for the suite of traditional payment arrangements but instead address niche use cases". Double whoops.

As for the claim of "instant" global payments, this is approximately true if you're talking about transacting through an exchange like Coinbase, where Coinbase has a custodial relationship to the actual coins and can reconcile the transaction asynchronously, but it is patently false when talking about transferring money directly without a middle-man. The very nature of blockchains requires that consensus is formed across many distributed machines to confirm an on-chain transaction. This takes time, and costs money (in the case of Ethereum, this cost is priced in 'gas'). When the network is congested (i.e. there's a lot of transactions happening) both the time and the gas cost of confirming an individual transaction increase. In practice, a genuine on-chain transaction on the Ethereum chain (which by definition any transaction not involving a middle-man would have to be) takes minutes at best and up to hours during times of high congestion. Now admittedly, hours is still pretty good turn-around time when we're talking about cross-border transactions, but it's a long long way from instant.

The next claim is that stablecoins eliminate foreign exchange fees. No actual evidence or even proposed model is presented for why this would be the case, so it's hard to assess the merits of the claim. I will say that yes, if we are comparing the cost of exchanging currency to how much your average retail bank charges (as opposed to say, a service like Wise), it would be pretty hard for currency exchange via stablecoins (i.e. buy CAD-backed stablecoin, swap for USDT, cash out in USD) to not be cheaper. It's already much cheaper to go through Wise, and cheaper still if you do the proposed stablecoin transaction with existing securities.

The claim about 24/7 access is just asinine. Yes, a web service is more readily accessible than a brick and mortar retail bank. Weirdly enough, I don't actually go to my bank right now for most transactions, I can do them from my smartphone with, you'll never believe it, 24/7 access! I haven't decided whether this claim is evidence that the memo was substantially written by ChatGPT, or evidence that Mr. Seif has a very dim view of his readers' intelligence. Both are insulting.

The final claim is the one I actually think has merit. I worked for a payment processing company and while I didn't work in the treasury function, I picked up enough to know that managing multi-currency treasuries effectively is a gnarly task that does not benefit from the pace of traditional banking's inter-day clearing processes. This is I think a perfect example of what the Bank of Canada described in their erroneously-cited paper as the utility of stablecoins: a niche use case for a user segment "that values their benefits and can accept their risks or costs".

The rest of the memo is a bunch of policy details about how the government ought to go about enshrining stablecoins in the financial regulatory framework. From my limited knowledge I don't find any of the concrete proposals objectionable. Mostly I find it objectionable that this is being presented as the critical policy objective regarding banking and payments that is relevant to the average Canadian, when it's really a niche pet policy that benefits a real but narrow set of businesses.