The Bank of Canada recently published a study on “cryptoassets” that contains a subtle but significant error. In this post I explain what went wrong and why the Bank, which has previously released high-quality scholarship on this topic, should prioritize a more nuanced understanding of the roles that digital money like bitcoin can play in the banking system.
The authors pose the question, “How much electricity would be needed to have all Canadian payments settled on a proof-of-work blockchain?” In the appendix, they calculate that settling every Canadian retail transaction on a bitcoin-like blockchain would entail astronomical energy costs, and conclude: “changes to the bitcoin technology and usage need to be substantial before it could be used as a retail payment method.”
This approach is based on a faulty premise about bitcoin and its blockchain that is emblematic of a broader misconception.
At its essence, the appendix conflates payment with settlement, and then determines that every retail transaction would be settled on the bitcoin blockchain, despite no such requirement at a technological or conceptual level.
Even putting aside the appendix’s use of widely–debunked energy estimates or its dismissal of bitcoin-integrated payment technologies, central banks should be conceptualizing of bitcoin as a money and the bitcoin blockchain as a high-security form of settlement. The bitcoin blockchain is not and was not designed to be a method of payment for every retail transaction in the world.
To analogize this distinction, imagine a payment between parties on an app like Venmo. If Alice keeps $100 in her Venmo account and sends $30 to Bob, the payment is merely numbers moved internally on Venmo servers. Only when either party decides to “cash out” — withdrawing funds from Venmo to their bank account, for dollars — might the transaction be considered settled:
Like Canadian Dollars or Euros or USD, bitcoin-based payments can be made available via Venmo/Cash App, credit cards, checks, ACH, Paypal, and other layered payment systems. In other words, retail payments could retain existing payment infrastructure and still be denominated in bitcoin, and not necessitate settlement to a bitcoin-like blockchain.
We do not even need to address major recent developments in bitcoin payments technology, such as the Lightning Network and other “Layer 2” development,” to understand that no changes to bitcoin’s technology need to take place before it can be used in Canada’s payment landscape. Bitcoin is not VISA or Venmo — it’s closer to money those companies move around.
To be charitable, perhaps the authors intended only to demonstrate what we discuss here: the futility of broadcasting all retail payments on bitcoin’s energy-intensive blockchain. If so, on this we agree. We do not mean to advocate that all Canadian retail payments should be conducted with or denominated in bitcoin. For many reasons, including familiarity, security, and price stability, Canadian customers and retailers may prefer to transact in CAD. Still, the conclusion that “changes to the bitcoin technology” are required is simply not true.
Moreover, the bitcoin blockchain’s proof-of-work electricity consumption is not an accidental byproduct: it is the means by which transactions are made all-but-irreversible a short time after they are added to the ledger and repeatedly confirmed by network consensus. In other words, the electricity and consensus confirmations offer a level of final settlement better-suited for high-value transfers (e.g., between banks) than for retail payments. If the intent here was to highlight the absurdity of using the bitcoin blockchain for retail payments, without acknowledging the utility of its blockchain for high-value settlement between institutions like central banks, then the authors should make that clear. As Hal Finney, the first person to ever receive a bitcoin transaction, predicted in 2010:
“I believe this will be the ultimate fate of Bitcoin, to be the ‘high-powered money’ that serves as a reserve currency for banks that issue their own digital cash. Most Bitcoin transactions will occur between banks, to settle net transfers.”
Bitcoin has a variety of pros and cons, uses and misuses. What it does not have is a centralized research department to create these sorts of reports, nor a communications department to publicize and distill its applications and policy implications for lawmakers and central bankers. Such are the consequences of decentralization.
Still, it would be a shame if policymakers do not realize that the questions here go beyond retail payments, and carry global, potentially historic implications. For example: when nations with their own currencies trade with each other, which money should they use for settlement? Former Federal Reserve Chairman Paul Volcker recently lamented the lack of a “common numeraire” in our international trade system. A numeraire, he describes, would act as the monetary unit in international settlement, and free central banks from reliance on the unpredictable supply of a third nation’s currency, most recently the U.S. dollar.
It is worth considering whether a digital money like bitcoin — with predictable and verifiable supply, as well as Internet-native design tailor-fit for digital banking — could fulfill the role of “numeraire” and establish a “metric unit” for international exchange. If that possibility seems outlandish, fair enough. So too at one point was the possibility of agreement on the mass of the kilogram or the length of the meter, but the measurements enabled trust and coordination, and rapidly became integral for global advancement in science and trade.
The Bank of Canada’s study shows that it is still all too easy to misunderstand what bitcoin is. This is especially true when communications from bankers and bitcoiners alike present strawman caricatures of each other. But this also means that if we are to get serious about the complex and overlapping realms of central banking, retail payments, and digital money, we all ought to be a bit more rigorous and critical. Especially when we read third party research, like this study here, that narrow and misconstrue the discussion of what bitcoin is, and could be.