How Often is Bitcoin Really Used as a Medium of Exchange?

Most Bitcoin Owners are Just Investors Not Customers

A friend asked me recently when I would admit bitcoin had won. What if the price of bitcoin went back to $60,000 or soared past its previous high to $100,000? Surely if bitcoin hit $1M (as many predict) I would have to sit down to my dish of crow and season it with all the opportunities I had missed on bitcoin’s meteoric rise.

If bitcoin hits a million dollars, I will undoubtedly wish I had bought a few bitcoin now and held on for dear life until the top just as — in hindsight — Laszlo Hanyecz is probably still kicking himself for parting with 10,000 bitcoin in exchange for two Papa John’s pizzas in 2010. Strange as it may sound, I do not believe that there is any particular price at which the case for bitcoin should be considered closed. A $1M bitcoin is not more conclusive than $100,000 bitcoin. Not in the least. Not unless and until other criteria have been met.

Instead of price, I think that the purchasing of pizzas — lots of pizzas and lots of other goods and services by lots and lots of people in exchange for bitcoin — is the only conclusive proof that bitcoin has secured its place as a going concern. At the point where I and many others routinely buy pizza (and other things) with bitcoin, I will concede. After all, this is the litmus test for the relative success of a currency: to what extent is it salable? To what extent is a hundred dollar bill in demand at all the various places selling valuables of various sorts and sizes?

Bitcoin may hit $1M in the future just as MicroStrategy hit $25B in March 2000. The price might even go higher, but unless bitcoin comes into its own as a preferred medium for pizza purchases, its position as a currency is not secured.

Easy Questions and Hard Questions

To some, the answer might seem like a foregone conclusion. Many seem to be under the impression that bitcoin is already used widely as currency. The easy answer is that bitcoin can and has been used as a medium of exchange. But so have cigarettes and ramen noodles — just not often outside of particular communities. The harder and more relevant question is whether the extent of this usage is proportionate to bitcoin’s place in the market as a top 30 world currency by market cap. Is, for instance, bitcoin used in a comparable number of market transactions to its neighbouring currencies by combined value?

Before we get started measuring bitcoin’s use as a medium of exchange, you might rightly point out that money actually has three roles and medium of exchange is only one of them — only one leg of the race. To rehearse them here:

1. Money is a Medium of Exchange — It lets two parties exchange across scale (sell a car, buy an apple), across space (sell house in Cleveland, buy house in Zurich), and across time (paid Friday, groceries Monday).
2. Money is a Unit of Account — a measuring rod for the value of goods and services. You know that $2 per apple is fine but $200 per apple is crazy. You know there’s something very wrong with the $80,000 house but there’s something very right with the $80,000 car. You know your nephew hasn’t earned $50 worth of leaf raking but that it is now worth $20 to you to send him home.
3. Money is a Store of Value — Ideally, the $1,000 you put in your bank on a sunny day in July will still buy the same amount of carpet and sheetrock on a rainy day next March when your backyard floods into your living room. The timeframe people need money to store value might be divided into short, medium, and long term.

Three roles. But money’s role as a medium of exchange is primary. Even Saifedean Ammous, writing in The Bitcoin Standard agrees: “Being a medium of exchange is the quintessential function that defines money” (emphasis mine). The reason this is the case is that being a medium in the middle between two other goods or services that might not otherwise be exchangeable (let’s say your services as a Veterinarian and that Jeep you want to buy) is money’s only real job. Its other jobs as a unit of account and as a store of value stem from and depend on this primary job. Money wouldn’t be a store of value if merchants no longer accepted it at the end of the storage period. Likewise, if money was rarely used in exchanges, it would not be used as a unit of account. Sometimes I barter using alfalfa with my neighbouring farmers but not enough to measure all the goods and services I buy in hay bales.

If a currency is a terrible store of value like the Turkish Lira, then it’s not useful as a medium of exchange across time because in the time between selling and buying, it may lose a good portion of its value. But the converse is not true. Midwestern farmland might be a fantastic store of value (people have to eat and they aren’t making any more land except in Dubai) but farmland would be terrible as a medium of exchange because you couldn’t, well, exchange it. This is the reason that lots of assets are used as stores of value that will likely never be used as currencies. Unless a currency works as a medium of exchange, it doesn’t work.

Fiat Acompli

As far as bitcoiners are concerned, bitcoin has been used as a medium of exchange for years. Quoting Ammous in 2017:

“After eight years, it is clear that this invention is no longer just an online game, but a technology that has passed the market test and is being used by many for real-world purposes, with its exchange rate being regularly featured on TV, in newspapers, and on websites along with the exchange rates of national currencies.”

The claim seems eminently justified. A lot has happened in the five years since Ammous wrote this and most of it has been positive for bitcoin’s market position. It is also clear that bitcoin is used as a medium of exchange by many in the bitcoin community, is actively traded on numerous exchanges, has been accepted as legal tender in El Salvador, and can now be spent more conveniently through integration with dozens of payment apps. You can even get Bitcoin-backed Visa or Mastercard debit cards and liquidate your bitcoin to pay merchants “everywhere you want to be”. The bitcoin network has not experienced any major security incidents even while processing $8 Trillion in transactions per year. The price of bitcoin has fluctuated but its market cap is now over $400 Billion. The bitcoin Core developer community has remained active and the broader bitcoin developer ecosystem has expanded. Setbacks like the FTX debacle have hurt, but not destroyed bitcoin’s reputation, bitcoin miner hash rate is up year-over-year, and rising interest rates have not stymied interest in bitcoin. There doesn’t seem to be a lot to complain about.

However, a skeptical person might examine Ammous’ claim and ask what those real-world purposes are and whether they are the reason for all the press attention given to bitcoin’s exchange rate. Is the press (or for that matter is the average bitcoiner) excited about bitcoin’s real-world use cases or is that not really the news? Are the real-world purposes causing bitcoin’s appreciating exchange rate? Are they even correlated?

The Canary in the Coal Mine

I do not think bitcoin’s utilization as a medium of exchange is as obvious or as extensive as it first appears. I will explain why, but first consider two peculiar facts which you can confirm from personal experience without leaving your couch:

1. Bitcoin Is Rarely Accepted as Payment. It is rare to find merchants that accept bitcoin natively (accepting bitcoin only after it has been converted to USD doesn’t count). This, 14 years after bitcoin’s creation.
2. Bitcoin’s Price Gets all the Press. Reading about bitcoin billionaires and crypto scams is much more interesting than reading about a new form of money. The extension of this is that if you own bitcoin, it’s likely you’re more interested (if you’re being honest with yourself) with the potential for your bitcoin to make you rich than with bitcoin’s usefulness in purchasing your next latte.

These two curious facts are just two sides of the same coin.

Two Sides to the Bitcoin: Product and Investment

Take the universe of all 100 million bitcoin owners. Now take out everyone who couldn’t give a flying fart whether bitcoin increases in value so long as it does its damn job as a zero-inflation peer-to-peer payment network — those who are using bitcoin as cold hard money for purely pragmatic reasons to buy and sell their goods and services.

Who’s left?

Almost everyone.

This is because every bitcoin has two sides. In addition to being a new kind of money — which to most people sounds rather boring — bitcoin is also a new way to get obscenely rich without being smart, working hard, or being well-connected — and that doesn’t sound boring to anyone. To be fair, it’s impossible to fully assess a person’s motives when she buys bitcoin. If someone was using her bitcoin as a way to circumvent a dictatorial regime, I would not suspect avarice as a motive. If two Harvard-educated twin multi-millionaires buy bitcoin, I would suspect they were more interested in exchanging the multi-M for a multi-B.

Every bitcoin customer is also a bitcoin investor. Every satoshi is both money and a lottery ticket. With bitcoin, this is an unavoidable duality and it’s not bitcoin’s fault. You can’t discount bitcoin’s usefulness as a product and success in the market just because it might also be a great investment. What you must do, however, is discount the narrative of bitcoin’s use as money by the extent to which bitcoin “customers” are really just “investors”.

In any other startup, if major “customers” were paid with stock it would raise eyebrows. If all the startup’s customers were paid with stock, one might legitimately question whether “customer” was the right word to describe them (“shill” comes to mind). If all the company’s customers were also investors and therefore not impartial consumers of the company’s products (if they were paid by the product rather than paying for the product) it might cause you to examine the market and see how much of the product’s alleged market fit was the market to fix a real problem vs the market to fix the investor’s retirement account. It might cause you to look and see whether the product’s actual usage in the market was comparable to its stock valuation.

If investor sentiment was a genuine measure of product-market fit, then the hundreds of millions invested in WeWork or Jawbone would have been a clear sign that large numbers of end users were willing to spend their hard-earned resources to buy these company’s products — but of course it was nothing of the sort. What it proved instead is that no amount of investment (no matter how large) is a clear sign of market acceptance. This makes estimating bitcoin’s market acceptance muddy and difficult because with bitcoin the users are the investors and the investors are the users. The people who are most excited about switching to bitcoin are those who think owning bitcoin is going to make them unspeakably wealthy. But the prospect of unspeakable wealth doesn’t tend to make one the most impartial product tester. It’s also frequently the case that the most enthusiastic investor doesn’t really have any interest in or need for the product on which the investment is based.

For anecdotal evidence of this phenomenon, go on Twitter, search for bitcoin, and start scrolling. Go on, I’ll wait but remember to come back. You’ll see 10 or 20 posts about bitcoin’s price, 10 or 20 posts about how bitcoin is going to make the world a better place, and 5 or 10 posts about some aspect of the bitcoin technology (how secure it is, how it’s limited to 21 million coins, how miners mine, etc.). Along the way, you might see one or two posts about how to use bitcoin and maybe one about some particular aspect of the bitcoin user experience in the marketplace. If you see news about how bitcoin is now accepted as payment (which you probably won’t without some digging), it’s likely referring to a no-name internet gaming site, not amazon.com.

Search for “bitcoin” on Google News. Same drill: you’ll find investor news from the mainline media, brash proclamations from the bitcoin propaganda organs, and some explanations of bitcoin’s technical architecture. No case studies. No user reviews. Almost nothing on why an average person would want to make or receive a payment with bitcoin instead of the alternatives. If you dig, you find that sometimes big international payments are faster and cheaper on bitcoin…just like the big international payments you do every day.

I’ve built a lot of investment pitch decks. Probably thirty or forty by this point that have been used to raise $100M+ in venture funding for a dozen different companies. If you treat all the bitcoin material on the web like a pitch deck or a marketing website, the weakest point is the social proof — real customers representative of a large market describing real problems which bitcoin solves.

But maybe Twitter and the rest of the web is not an accurate indicator of bitcoin’s usage. Maybe the conspicuous dearth of stories and details about people (you know) buying stuff with bitcoin is due to the media’s bias toward sensationalism. Or perhaps bitcoin’s usage really is completely disconnected from its price.

Bitcoin in the Market for Money

So what is bitcoin’s usage as a medium of exchange? And is that usage roughly in line with what you would expect based on bitcoin’s aggregate valuation? If its use in the market is comparable to its value, then that would go a long way toward justifying bitcoin’s price. If its use as a medium of exchange is nascent relative to its $400 Billion valuation, then that would suggest that there are a lot of enthusiastic bitcoin investors, but wouldn’t tell you much of anything about whether bitcoin makes good money.

As best I can estimate, the number of bitcoin transactions per day is a tiny fraction of what you would expect if bitcoin were being used like other currencies. Let me share some numbers with you and what they means.

The number of non-cash payments (transactions) in the US is about 200 Billion per year. Since most American’s use non-cash payments most of the time, this accounts for a majority of transactions. If you divide these 200 Billion yearly transactions by 365 days, you get about 550 million transactions per day. Factor in the US population of 330 million, and add a few tens of millions of cash transactions and you find that each person in the US conducts a rough average of 2 transactions every day. Intuitively this makes sense. You pay your bills, pick up groceries, buy a coffee. Some days you don’t make payments and some days you make five or ten. My checking account lists 50 transactions for this January and that’s probably a third to a half of my household’s transactions during that period. Your patterns are probably similar.

The majority of these transactions take place in the market of goods and services. You get paid by your employer with a bi-weekly ACH, you then make bank payments to your utilities, landlord, and cell phone provider and card payments to retailers, suppliers, etc. You use cash to pay the babysitter or the neighbourhood lawn guy. You’re not just trading dollars for other currencies, you’re using dollars to exchange for other things.

Two transactions per day per person would suggest that a country like Sweden with a population of 10 million people would make 20 million transactions per day if Swedish buying habits are similar to those of Americans’. The Swedish Krona’s M2 money supply (cash, checking accounts, savings accounts, & CDs) is about $200 Billion in US dollars.

The current market cap of bitcoin (a measure roughly equivalent to M2) is about $450 Billion in USD. One might reasonably expect that a currency over double the value of the Swedish Krona would be used in double the number of transactions — enough to represent a population of 20 or 25 million people. A population this size would likely conduct 40 or 50 million transactions per day. Keep that number in mind.

Where to Find Bitcoin Transactions

Just as with USD transactions, it’s impossible to know precisely how many transactions take place with bitcoin because a transaction can take place off the blockchain without a record of the transaction. However, there are a limited number of places where transactions may be hiding and ways to at least estimate their scale.

Here are the possible places you might find bitcoin transactions:

1. On the bitcoin blockchain — this is the easy place to start
2. Centralized Exchanges like Binance
3. Layer 2 transactions off-chain. The Lightning Network is by far the largest and most mature L2 network
4. Transactions using crypto-backed debit cards on standard payment networks like Visa

Let’s search these places one by one.

1. The Blockchain

The bitcoin blockchain is only processing about 300,000 transactions per day (its limit is about 500 thousand per day). A single Walmart Supercenter might process 10 thousand daily transactions and as we’ve seen, you’d expect a town of 150,000 people to make that number of daily transactions. As Saifedean Ammous puts it in The Bitcoin Standard,

“300,000 daily transactions is the number of transactions that takes place in a small town, not in a medium-sized economy…This suggests that bitcoin adopters value it more as a store of value than a medium of exchange.”

2. Centralized Exchanges

When two parties use a cryptocurrency exchange like Binance or Coinbase the transaction may be handled by the exchange off-chain on a private ledger. Eventually ownership would need to be settled on the blockchain and the exchange might also use the Lightning Network. However, for the purposes of this discussion, the number of transactions on exchanges is not that relevant since it represents trading rather than market usage (trading assets vs buying goods). The trading volume of the NYSE might show up in the Fed’s data as ACH payments, but it would normally only account for a small percentage of the total. The average person trades assets at a much lower frequency than she buys things like toothpaste and gym classes.

3. The Lightning Network

Second-layer off-blockchain protocols like the Lightning Network account for some number of additional bitcoin transactions. It’s impossible to say exactly how many transactions take place on the Lightning Network because transactions are private between two peers on the network. You can, however say with certainty that Lightning currently has less than 20 thousand nodes trading about $130 million worth of bitcoin. This would indicate that 0.025% of bitcoin’s market cap (the layer 1 capacity ratio) is available for exchange on the Lightning Network. The number of users of the Lightning Network is likely much greater than 20 thousand through the use of custodial wallets like Wallet of Satoshi that offer the use of a Lightning node without running a node yourself.

The only reliable estimate I’ve found of the number of transactions on Lightning is from a report published in April 2022 by Arcane Research, research firm on a mission to “…give people the confidence to trust digital assets and thereby enable worldwide adoption”. Their estimates, based on surveys with Lightning payment providers, suggest that the number of payments on Lightning is a little under a million payments per month and only 20% of those payments were to merchants (the rest being person-to-person and trading). If those estimates are true, then Lightning only adds another 30,000 transactions per day to bitcoin’s total only a few thousand of which are used to buy or sell goods and services.

Crypto-Backed Payment Cards

Estimating the number of transactions on Visa and MasterCard crypto-backed debit cards is even more difficult but it’s possible to get a ballpark. Visa mentioned in their earnings call on January 27, 2022 that in the previous quarter, Visa’s 65+ crypto partners had conducted $2.5 billion in payments volume, which amounted to over 70% of the crypto-backed volume in all of Visa’s fiscal year 2021 (ending Sept 30, 2021). If you annualize this $2.5B, you could conclude that $10B in bitcoin transactions occur every year on Visa alone. Assume Mastercard adds transaction volume proportionate to its network size and you might throw in another $3B. Wouldn’t this indicate significant bitcoin usage as a medium of exchange? It would certainly help. But you need to add a few footnotes:

Other cryptocurrencies: The total market value of all cryptocurrencies is about $1T. This means that bitcoin is only 40% of the market. Since crypto-backed Visa cards might use other cryptocurrencies, a conservative estimate would assume at least half of the $2.5B in quarterly volume was other cryptos. This would cut annual bitcoin-backed payment volume to something like $5B rather than the full $10B.

Crypto prices in late 2021: Sept 30 through Dec 31st, 2021 (the period in which Visa reported $2.5B in crypto-backed volume) saw the value of bitcoin and most other cryptocurrencies hit all-time highs, sometimes two or three times their current price. People tend to spend more when they have more to spend. Tellingly, none of Visa’s subsequent earnings releases have disclosed crypto-backed payment volume for the last few quarters. If bitcoiners had two or three times as much bitcoin to spend, one might reasonably assume they would spend at least 1.5x as much in that period as in later periods. Thus, $5B gets trimmed down to more like $3.5B annually.

Christmas: The last three calendar months of every year see more shopping volume than other quarters by a few percentage points. We’re likely down to about $3B annually in bitcoin-backed Visa payments.

US dollar transactions: Some “crypto” debit cards also allow you to spend in USD. Who knows what portion of the $2.5B was simply spending plain old dollars. We’ll give bitcoin the benefit of the doubt on this one.

Double counting: Some portion of the payments through the Visa network likely take place through the Lightning Network and are settled on the bitcoin blockchain (which would be accounted for in the previous sections). This may be a relatively small number, but it’s not likely zero.

You’re not paying in bitcoin: At the time of purchase, crypto-backed debit cards exchange the bitcoin in your account with USD or whatever local currency the merchant accepts. So it’s not purely a bitcoin transaction and not even primarily a bitcoin transaction. It’s evidence that merchants don’t yet accept bitcoin as well as evidence that some people want to pay in bitcoin. The USD in this exchange is more important than the bitcoin. If you change out the bitcoin for Yen behind the Visa card, you can still buy your latte. If you try and pay the merchant with straight bitcoin, you’re likely to leave empty handed. You can, incidentally, get a gold-backed debit card that works the same way, and the merchant doesn’t get paid in gold either.

It’s still small potatoes: In the same fiscal quarter, Visa announced a total worldwide payment volume of $3.6T. This means that the $2.5B crypto-backed payment volume was less than a tenth of a percent (0.0689% to be exact) and bitcoin payment volume (as only half the market cap of all crypto) was likely only about one third of one tenth of one percent of Visa’s total payments.

Adding It Up

If you run a back-of-the-napkin calculation with the above considerations, you get a likely bitcoin-backed debit card payment volume of $3B to $4B annually factoring in likely Mastercard payments. This equates to about $10M daily, which (at the average Visa transaction price of $17) suggests about 600,000 transactions per day if you give bitcoin 100% credit for merchant exchanges made possible by converting to USD or a local currency. Add this to the 300,000 transactions on the blockchain (assuming no duplication) and add in 30,000 Lightning transactions and you’ve probably graduated from the transaction volume of a small town to that of a medium-sized city: about a million transactions per day (or maybe two million just to pad the numbers with transactions we may have missed).

A million daily transactions is nothing to sniff at, but it is completely disproportionate to the 40 or 50 million you’d expect from bitcoin’s market cap. It is, however exactly what you would expect if most bitcoin owners were merely using bitcoin as an investment (something you buy and hold or trade periodically) rather than as a currency (something you use to buy and sell). These million daily transactions include many of the transactions that are merely trading between investors although it does not include all the trades done off-chain on exchanges.

Big Shoes to Fill

To be clear, none of this proves that bitcoin cannot function technically as a large-scale medium of exchange, nor the impossibility of adopting it for that function in the future. There are legitimate technical questions as to whether it can, legitimate strategic questions as to whether it will, and legitimate practical and philosophical questions as to whether it would be preferable. The fact that bitcoin’s usage as a medium of exchange is relatively low doesn’t answer any of those questions.

The question it does answer is whether bitcoin’s present day market usage as a medium of exchange justifies bitcoin’s market valuation as an investment. It most certainly does not. If it becomes clear that bitcoin’s usage as a product has peaked, its valuation as an investment will collapse. And bitcoin has no other possible usage except as a medium of exchange with which to prop up its valuation. The only possible value bitcoin can store for the long-term is the value of its use as a medium of exchange (some bitcoiners have contested that these two roles are separate. I deal with bitcoin’s use as a store of value elsewhere).

It is possible that bitcoin’s market usage will one day catch up to the expectations set by its valuation. But that day is not today. To get there, bitcoin will have to chart a path through various technical challenges, various government restrictions and regulations, and various competitive threats from both traditional and crypto currencies. More people will have to adopt bitcoin as a product rather than as an investment. It probably also requires that the buy and HODL behaviour of many bitcoiners begins to change to something more pragmatic and less ideological.

The path is not impossible, but it is beset with risks — some of which are existential and very few of which seem to be openly acknowledged by bitcoin advocates although they are confirmed by the volatility of bitcoin’s price in the free market. This path is not optional and bitcoin has certainly not already succeeded. Neither is it destined to succeed through the divine intervention of Satoshi Nakamoto nor is it inexorably drawn up and to the right by the physics of money.

Bitcoin will one day have to get a grown up job helping lots of people exchange one good for another — the only job it’s qualified to do — if it expects to keep getting its grown up valuation.