Tokens vs. Accounts: Why the Distinction Still Matters

October 05, 2020
IT technician in data center 

The conceptual distinction between payments methods that are based on tokens (e.g., coins) and payments systems that are based on accounts (e.g., debit cards) is pretty well entrenched in discussions of payments economics.See, for example, Edward Green’s chapter “Some Challenges for Research in Payments” in the 2008 book The Future of Payment Systems (edited by Andrew Haldane and others), as well as my 2009 paper “Why Pay? An Introduction to Payments Economics” (co-written with William Roberds) and the 2018 Bank for International Settlements article “Central Bank Digital Currencies.” For example, here is an important difference between the two types of arrangements:

  • When you pay with an account, the crucial question for the recipient is your identity: “Are you really the account holder?”
  • When you pay with a token, your identity is irrelevant; instead, the crucial question for the recipient is: “Is this object I’m receiving real or counterfeit?”

There are other differences I could point to.My 2020 working paper “Eggs in One Basket: Security and Convenience of Digital Currencies” (co-written with Francisco Rivadeneyra and Tsz-Nga Wong) points out a second important distinction: Accounts aggregate funds in a single location, making the entire amount vulnerable whenever the account is accessed. Tokens can be stored separately, potentially providing better protection for the unaccessed portion of the funds. But I regularly hear computer science professionals say that the distinction between tokens and accounts is meaningless—or more charitably, that it is irrelevant in the brave new world of cryptocurrencies and other revolutionary electronic payments methods.

A Meaningless Distinction?

Actually, I have sympathy with this reaction, because it usually comes when some ill-informed company executive has asked the computer specialist, “Should we make our proposed electronic money a token-based system or an account-based system?” If the computer expert’s specialty is the guts of the system, then as far as she or he is concerned the distinction is meaningless; one could say that at the base, all the systems are accounts (or one could say all systems are tokens or all systems are both). Or the expert could ask, “What do you think you mean by account versus token?” And then the computer expert would explain that the distinction has nothing to do with the fundamental architecture.

That is fine, and repeated attempts to ask computer specialists questions like that explains their irritation.

The mistake comes, however, when the specialist replies that the distinction is meaningless or that there is no real distinction between tokens and accounts. The distinction is only meaningless at the level of the architecture on which the specialist is focusing.

Explaining New Products

Go up a few levels in the design, until you are closer to the front end, and look at what the customer experiences. The customer has some categories in his head. He knows what a coin is. He knows what a bank account is. The startup company is offering him something new, and you, the company representative, are trying to explain it to him. Maybe what you are offering on the front end is a way to simulate putting cash into an electronic birthday card for a kid—that is something that “feels” like a token. Or maybe what you are offering is something that allows the customer regularly to pull funds from his paycheck electronically and ensure that he doesn’t spend the money until it’s time to do holiday shopping. That “feels” like a savings account.

Your company has to get people to adopt the new technology, whatever it is, and to do so, you need to explain the technology in terms of existing categories. And your marketers will be working with the software people who specialize in the front end to enhance the design in ways that allow the customer to relate your new payments system to the existing categories.

A Handy Shortcut

Now your company has some flexibility—the advertising can say “Our new product is just like a coin, except that …” and add in a couple of features that are different. Or it can say, “Our new product is just like a bank account, except …”

What the company can’t hope to do is to list one hundred different features that a payment system might possess.

In most respects, the product—no matter how revolutionary—has to start out as analogous to something that already exists. An automobile is not a “horseless carriage,” but that description was necessary as people began to adopt it.

For this reason, the token versus account distinction is useful; it is a handy shortcut to describe a set of features that coins, banknotes and stored value cards have in common, as opposed to credit cards and bank accounts. Yes, the distinction is fuzzy; old-fashioned numbered Swiss bank accounts straddle the line. But the bundle of features is still a useful categorizing device, as long as we are careful, when necessary, to specify the particular feature on which we are focusing or the feature we are treating as exceptional.

The Need for Distinction

And as long as people continue to think along the lines of traditional payments methods and the traditional distinctions between them, the legal and institutional structures will continue to do so. So if your company advertises its application as “Birthday Coins” and people rely on that description, a judge might not let you get away with saying “No, your honor, they are really accounts.”

That’s why I think it’s a mistake to claim, as many do, that we mustn’t describe any new electronic payments arrangement (e.g., Bitcoin) as a “token” because deep down they are all “really accounts.”For a thoughtful analysis of why bitcoin is not a token, see Alistair Milne’s paper “Argument by False Analogy: The Mistaken Classification of Bitcoin as Token Money.” For a thoughtful analysis of why bitcoin is both a token and an account, see the blog post “Token- or Account-Based? A Digital Currency Can Be Both” written by Rod Garratt and others. With complex systems sometimes we want to think in reductionist terms (for the purpose of determining her weight, my grandmother really is a bunch of molecules) and sometimes we want to think in holistic terms (for the purpose of thinking about the health care system, my grandmother is really a person, an important person with rights and needs).

And so rather than saying that “token versus account” is meaningless, or irrelevant in the digital age, what we should say is that at the deepest levels of computer architecture, the distinction makes no sense, but at higher levels of design it remains important—at least until the day comes when most people don’t remember paper money.

Notes and References

  1. See, for example, Edward Green’s chapter “Some Challenges for Research in Payments” in the 2008 book The Future of Payment Systems (edited by Andrew Haldane and others), as well as my 2009 paper “Why Pay? An Introduction to Payments Economics” (co-written with William Roberds) and the 2018 Bank for International Settlements article “Central Bank Digital Currencies.”
  2. My 2020 working paper “Eggs in One Basket: Security and Convenience of Digital Currencies” (co-written with Francisco Rivadeneyra and Tsz-Nga Wong) points out a second important distinction: Accounts aggregate funds in a single location, making the entire amount vulnerable whenever the account is accessed. Tokens can be stored separately, potentially providing better protection for the unaccessed portion of the funds.
  3. For a thoughtful analysis of why bitcoin is not a token, see Alistair Milne’s paper “Argument by False Analogy: The Mistaken Classification of Bitcoin as Token Money.” For a thoughtful analysis of why bitcoin is both a token and an account, see the blog post “Token- or Account-Based? A Digital Currency Can Be Both” written by Rod Garratt and others.

Additional Resources

About the Author
Charles Kahn
Charles M. Kahn

Charles Kahn is a research fellow with the Federal Reserve Bank of St. Louis. Read more about the author and his research.

Charles Kahn
Charles M. Kahn

Charles Kahn is a research fellow with the Federal Reserve Bank of St. Louis. Read more about the author and his research.

Related Topics

This blog offers commentary, analysis and data from our economists and experts. Views expressed are not necessarily those of the St. Louis Fed or Federal Reserve System.


Email Us

Media questions

All other blog-related questions

Back to Top