Digital Euro Association Blog

It’s time to abandon the “Token vs. Account” discussion

Perusing any whitepaper or report on central bank digital currencies (CBDCs), the reader is almost certain to find a section devoted to the distinction between “token-based” and “account-based” CBDC, where ownership of the former is demonstrated by having knowledge of a private key (or digital signature), while that of the latter is tied to identity.

But, as Charles Kahn of the Federal Reserve Bank of St. Louis noted, this “distinction makes no sense”.

To be fair, Kahn made that statement as it relates to “the deepest levels of computer architecture” and went on to conclude that “at higher levels of design [the distinction] remains important”. Indeed, Khan helpfully highlights that adoption of new products often require useful analogies to existing options, reminding us that automobiles used to be referred to as “horseless carriages”.

The same can be said for CBDCs, where we often use the example of bank deposits to help explain the idea for an account-based CBDC and banknotes that exist in one’s physical wallet as akin to that of a token-based CBDC, which may one day exist in one’s digital wallet.

These distinctions were undoubtedly useful when CBDCs were merely a theoretical idea, but now that more and more central banks begin to experiment with architectural designs for CBDCs – including recent news of the Federal Reserve’s potential “intermediated model” –  insisting on maintaining such distinctions not only causes confusion about what is (currently) legally possible, but it risks obfuscating the real impact on user anonymity.

As such, I believe it’s time to abandon the “token vs. account” discussion.

“I am, and I know, therefore I own”
To put it simply, pure token-based CBDCs do not currently exist. 

This idea may be obvious if one believes that a central bank would opt to use a legacy system for its CBDC design, as such systems, by default, would require the use of accounts that are maintained on a traditional ledger. 

But if a central bank instead opted to use blockchain technology currently being deployed to support cryptocurrencies and stablecoins, the situation becomes less obvious. 

To help illustrate the potential confusion that exists, it’s instructive to look at how official institutions like the International Monetary Fund (IMF) explain the issue. In a report examining various legal aspects of CBDC, the IMF differentiated account-based CBDC from digital tokens thusly:

  • For book money in general, including account-based CBDC, the identity of the account holder allows the holder to access the funds: “I am therefore I own”
  • For digital tokens, the knowledge of a password (often referred to as “private key”) allows the holder to transfer the funds: “I know therefore I own”

On the surface, this might seem fairly straightforward, but if one digs a bit deeper, the flaws in this argument begin to appear rather quickly. For example, how does an account holder typically prove her identity when she wants to access her online bank account to pay bills? Quite often she would enter a password that she knows.

As for the digital token holder, he certainly needs knowledge of a private key (or password), but that key is just one of a pair – the other being a public key address. And what is a public key address if not an account? 

Indeed, researchers at the Federal Reserve Bank of New York used Bitcoin as an example to show that digital currencies fit the very definition of an account-based system, noting that “[t]he account is the Bitcoin address, and the private key is the proof of identity needed to transact from the account.”

Since digital tokens need an account (e.g. public key) to send and receive transactions, it’s clear that pure token-based CBDC do not exist. Additionally, it’s also clear that for both account-based CBDC and digital tokens, the IMF’s defining statements should be revised to read: “I am and I know, therefore I own”.

Questioning the Legality of Retail CBDC
Importantly, if all CBDC is thus account-based, there are significant ramifications on the potential legality of retail CBDC being issued by a central bank to the general public.

In the same legal report mentioned above, the IMF argues that since “account-based CBDC is ‘book money’, it can only be offered to entities for whom the central bank is authorized to offer cash current accounts.” But many central banks are not currently authorized to offer accounts to the general public. 

The Federal Reserve’s recent paper on CBDC acknowledges this point, noting explicitly that “[t]he Federal Reserve Act does not authorize direct Federal Reserve accounts for individuals”.  However, they attempt to sidestep this issue by suggesting that “[u]nder an intermediated model, the private sector would offer accounts or digital wallets to facilitate the management of CBDC holdings and payments”.

Unfortunately, it’s unclear what sort of account an intermediary could offer relative to CBDC. It’s certainly true that an intermediary could offer a digital wallet, facilitating the management of CBDC holdings and payments. But this wallet merely facilitates transactions, it’s not an account in and of itself.

To illustrate this point, let’s assume the Federal Reserve decided to issue a CBDC on it’s own private, permissioned blockchain and, in keeping with the intermediated model, the Fed distributes this CBDC to commercial banks which already have access to its private network. To do so, the Fed would effectively authorize CBDC to be transferred from its account (or public key with a hypothetical address of 0x..3c) to that of the commercial bank (where its public key address is hypothetically 0x…5k). Simultaneously, the Fed would debit the commercial bank’s reserves held at the central bank by the amount of CBDC that was transferred (see bullet 1 in the chart below).

Now, let’s further assume that Alice would like access to the CBDC and goes to her bank to help her facilitate the management of these CBDC holdings. After confirming all compliance checks (KYC, AML, etc), the bank would help Alice download a digital wallet and obtain a public key address (hypothetically 0x…7j), as well as a private key. At this stage, it’s important to note that although the bank offered Alice a digital wallet to facilitate the management of CBDC holdings and payments, Alice’s account (or public key) related to the CBDC resides on the Federal Reserve’s blockchain where its CBDC is issued. 

Furthermore, when Alice instructs her bank (via the digital wallet) to debit her checking account and send the corresponding amount of CBDC to her public key address, the bank will send the CBDC from it’s own public key address (0x…5k) to Alice’s public key (0x…7j) and the entire transaction will be recorded on the blockchain ledger operated by the Federal Reserve (see bullet 2 below).

No alt text provided for this image

Putting an End to Token-Based CBDC

Returning to the idea that banknotes which exist in one’s physical wallet could serve as a useful analogy for (token-based) CBDC that exist in one’s digital wallet, we can now see how confusing and misleading that comparison really is. Although both physical banknotes and digital CBDC are liabilities of the central bank, a separate staff paper published by the Federal Reserve Board of Governors reminds us that “tokens in the cryptocurrency space are not stored locally but rather on a blockchain”.

Thus, Alice’s CBDC is not a token stored locally in the digital wallet provided to her by the bank, but rather an account entry on the Federal Reserve’s blockchain ledger. Indeed, this idea of a “token” is problematic, because as the Board of Governors staff paper goes on to say, “[w]hat the cryptocurrency community calls tokens can be tracked in a form that central bankers might recognize as accounts”.

Therefore, if a Federal Reserve-issued CBDC is inherently account-based – as it would currently seem – it‘s important we recognize that fact sooner rather than later, as such a system would currently violate the Federal Reserve Act and require new legislation before being implemented. 

Instead of focusing on issuing a retail CBDC, perhaps now is the time for the Federal Reserve to seriously consider the idea of allowing private stablecoin providers access to Fed liabilities, such as a master account or a (potential) Fed-issued wholesale CBDC. More details about the benefits of such a system can be found in my report on “tokenizing trust”, as well as George Selgin’s recent paper examining a “narrow path to efficient digital currency”.

Finally, putting an end to the “token vs. account” discussion would also be beneficial for potential holders of retail CBDC. Users, like Alice, need to understand that the CBDC she holds in her digital wallet is not fully anonymous like the banknotes in her physical wallet. Although her identity may not appear on the blockchain alongside each transaction, an identifier (e.g. her public key address) will be present, making such exchanges, at best, pseudo-anonymous.

Board of Governors of the Federal Reserve System. “Money and Payments: The U.S. Dollar in the Age of Digital Transformation.” January, 22, 2022,

Bossu, Walter, Masaru Itatani, Catalina Margulis, Arthur D. P. Rossi, Hans Weenink, and Akihiro Yoshinaga. “Legal Aspects of Central Bank Digital Currency: Central Bank and Monetary Law Considerations,” International Monetary Fund, Working Paper No. 2020/254, November 20, 2020,

Garratt, Rob, Michael Lee, Brendan Malone, and Antoine Martin. “Token- or Account-Based? A Digital Currency Can Be Both,” Federal Reserve Bank of New York, Liberty Street Economics, August 12, 2020,

Khan, Charles, “Tokens vs. Accounts: Why the Distinction Still Matters,” Federal Reserve Bank of St. Louis, On the Economy, October 5, 2020,

Lee, Alexander, Brendan Malone, and Paul Wong (2020). "Tokens and accounts in the context of digital currencies," Board of Governors of the Federal Reserve System, FEDS Notes, December 23, 2020,

“This article was prepared by the author. The views expressed in this article are the author’s own and do not necessarily reflect the views of the Digital Euro Association.”

No Comments Yet

Let us know what you think