The end of El Salvador's bitcoin payments experiment

Back in 2021, El Salvador became the first country in the world to require its citizens to use bitcoin for payments. Last month, four years later, it notched another record: it became the first country to rescind bitcoin's status as required tender. This backtracking was the result of the IMF's threat to pull billions of dollars in assistance if El Salvador didn't put an end to bitcoin's special status.

What have we learnt from El Salvador's four-year bitcoin experiment? I would suggest that it definitively proved that bitcoin is not destined to be money. As far as making payments goes, bitcoin will always be an unpopular option, even when the government gives it a helping hand. And don't blame the IMF for this; bitcoin sputtered-out long before the IMF pressured El Salvador to drop it, as I'll show.

The original motivation behind El Salvador's Bitcoin Law was to harness bitcoin as a means for reaching the unbanked, those without bank accounts, who in El Salvador make up the majority. Cash is still by far the dominant payments choice in El Salvador, but it was believed that an electronic form of cash might complement that. Another goal was to make remittances cheaper by sponsoring a new bitcoin remittance route—few countries are as dependent on remittances from family living overseas as El Salvador. 

President Nayib Bukele made the announcement at a major bitcoin event and El Salvador’s Congress ratified the Bitcoin Law a few days later. Bitcoiners literally cried for joy. For longtime Bitcoin watchers like me, it seemed like an awful idea. But at least it was going to be a fantastic natural experiment.

Satoshi Nakamoto, bitcoin's founder, saw bitcoin as electronic cash, but his dream generally hasn't come to fruition. In practice, 99% of bitcoin adoption is about gambling on its volatile price, with payments being a niche 1% edge case. Bitcoin disciples who continue to believe in Satoshi's electronic cash dream often blame what they see as government meddling for the failure of bitcoin to gain widespread usage as a payments medium. For instance, they say that capital gains taxes on bitcoin makes it a hassle to pay with the orange coin, since it leads to a ton of paper work anytime one buys something with bitcoin. Or they criticize legal tender laws that privilege fiat currency. 

But here was a government that was going to champion the stuff, nullifying all of the headwinds against bitcoin in one stroke! The government meddling hypothesis would be put to test.

The Salvadoran government used a combination of sticks and carrots to kick-start adoption. First, let's list the carrots. The capital gains tax on bitcoin was set to zero to remove the hassle of buying stuff with bitcoin. The government also built a bitcoin payments app, Chivo, for all El Salvadoreans to use. (Chivo also supports U.S. dollar payments.) Anyone who downloaded Chivo and transacted with bitcoin would receive a $30 bitcoin bonus—that's a lot of money in El Salvador. Gas stations offered $0.20 off of a gallon of gas for customers who paid with the app. People could also use Chivo to pay their taxes with bitcoin.

The biggest carrot was zero-transaction fees. Any payment conducted with Chivo was free, as was converting bitcoins held in the Chivo app into U.S. dollars and withdrawing cash at Chivo ATMs. These Chivo ATMs were rolled out across El Salvador and in the U.S., too, to encourage the nascent U.S.-to-El Salvador bitcoin remittance route. Bitcoin ATMs are usually incredibly pricey to use, but in El Salvador the government would eat all the transaction fees. What a fantastic deal.

As for the stick, Bukele introduced a forced-tender rule. Beginning in 2021, businesses were required to accept the orange coin or be punished. This was costly for them to comply with. They would have to update point of sale software, signage, train employees, and set up new processes for handling bitcoins post-sale.

By all rights, this combination of sticks and carrots should have led to a flourishing of bitcoin payments. But it didn't.

The evidence of failure

The first incrimination of the experiment is Figure 1, below. In the Bitcoin Law's initial months, remittances carried out by cryptocurrency wallets exploded, accounting for an impressive 4.5% of all incoming remittances to El Salvador. Not bad! People were actually using the Chivo app to send bitcoins to relatives back home. 

Figure 1: Data from El Salvador's central bank shows that cryptocurrency remittances from wallets like Chivo have steadily shrunk over time from 4.5% of all remittances to 0.87% of all remittances in 2024.

But instead of continuing to gain market share, crypto-linked remittances steadily deteriorated over the next four years to 0.87% of the total by December 2024—hardly a sign of success.

The data for this chart comes from the Banco Central De Reserva (BCR), El Salvador's central bank. The BCR is coy on how precisely it collects this data, but it is almost certainly dominated by Chivo-related transactions. (My note at the bottom explores the data more.)

The second indictment of El Salvador's bitcoin effort comes from survey data compiled by economists Alvarez, Argente, and Van Patten in their 2022 paper, Are Cryptocurrencies Currencies? Bitcoin as Legal Tender in El Salvador. The authors carried out a survey of 1,800 Salvadoran households to get insights into their use of the Chivo Wallet. This wasn't a lazy online survey, but an in-person survey.

The survey found that just over half of Salvadoran adults had downloaded Chivo, which is impressive (see Figure 2, below). Most hardly used it, though. While over 20% of the population continued to interact with Chivo after spending their $30 bitcoin bonus—which isn't a bad adoption rate for an app—the majority of Chivo usage was only occasional, the median Chivo user reporting no bitcoin payments sent or received in any given month, and just one payment per month in U.S. dollars. Payments tools like apps and cards are supposed to be used a few times each week; not once every two or three months.

Figure 2: While awareness of Chivo was high, most Salvadorans did not use Chivo's bitcoin functionality after receiving their $30 bitcoin bonus. Source: Are Cryptocurrencies Currencies? Bitcoin as Legal Tender in El Salvador [link]

The dominance of the app's dollar functionality over its bitcoin functionality also stands out. Chivo was supposed to be a bitcoin payments app, after all, not another version of PayPal of Venmo. For instance, the survey found that of all households who had downloaded Chivo, only 3% had ever received a bitcoin remittance via Chivo, while 8% had received a U.S. dollar remittance via the app (see Figure 3 below). If Chivo was primarily being used for fiat payments, and not bitcoin, then why go through with the whole effort of changing the law for bitcoin's sake?

Figure 3: When Salvadorans did use Chivo for remittances, they preferred it for U.S. dollar remittances over bitcoin-based ones. Source: Are Cryptocurrencies Currencies? Bitcoin as Legal Tender in El Salvador [link]

Moreover, those few citizens who did continue to use Chivo regularly were not the unbanked majority that the Bitcoin Law had originally targeted. The survey found that they were most likely to be from the already-banked minority, young, educated, and male.

By mid-2022, downloads of Chivo had pretty much dried up. Using blockchain tracing, the economists found that $245,000 per day worth of bitcoins were flowing into the Chivo app, which sounds like a lot, but in the payments business, that's peanuts.

It's also worth considering how businesses treated bitcoin after the passing of the Bitcoin Law. Despite the requirement that all businesses  accept bitcoin, just one-in-five actually did so. The survey found that acceptance was driven by large businesses—i.e. McDonald's, Starbucks, Pizza Hut and Walmart—presumably because they couldn't easily evade the consequences of ignoring the law. Bitcoin was not popular with these businesses; the survey found that of those that received bitcoin from their customers, 88% quickly converted them into dollars.

This is problematic. For bitcoin to become money, a circular economy must be kickstarted as the bitcoins spent by consumers are re-spent by businesses on inventory and salaries, which gets re-spent by consumers, and on and on. This wasn't happening.

With just 20% of the population using the app, and mostly for an occasional U.S. dollar transaction, the entire bitcoin experiment can hardly be seen as a wild success. Businesses were not keeping the bitcoins they received, and consumers who were using the app regularly were not the unbanked originally targeted by the Bitcoin Law.

The third and last bit of evidence of the experiment's failure comes from an annual survey from José Simeón Cañas Central American University (UCA) entitled La población salvadoreña evalúa la situación del país. In 2021, the survey began asking Salvadorans whether they had ever used bitcoin to buy or pay for something. This question is more open-ended than the one asked by the three economists, who focused more narrowly on bitcoin transactions conducted via Chivo. 

Figure 4: According to a survey from the UCA, while over 25% of survey participants reported using bitcoin (and not just Chivo) for payments in 2021, only 8.1% used bitcoin for payments just three years later in 2024.

In the first year of the Bitcoin Law, 25.7% of respondents said they used bitcoin for payments. That's a fantastic result, although the $30 Chivo bonus no doubt drove that large number. But over the next three years, bitcoin's usage for payments crumbled, with only 8.1% of Salvadorans reporting that they'd paid with bitcoin by 2024. This is the same downward pattern that we saw in the CBR's remittance data. That's not adoption. That's giving up on bitcoin.

The UCA survey found that the 8.1% who reported using bitcoin for payments in 2024 were not using it for day-to-day payments. Of this group of bitcoin payors, 55% used bitcoin just 1-3 times in 2024. Only 8% made bitcoin payments on a weekly or bi-weekly basis. (See Figure 5 below). I really want to highlight this last data point: in 2024, just 1 in 200 Salvadorans paid for something each week or second week with bitcoin.

Figure 5: In a 2024 survey by UCA, 8.1% of Salvadorans reported using bitcoin for payments that year. This group was then asked how often they used it, with the responses visualized in the above chart. Most used bitcoin just once in 2024, with 55% using it one to three times. 8% used it 20 or more times, which would suggest that almost no one is using bitcoin for day-to-day payments, despite that being the goal of El Salvador's 2021 Bitcoin Law.

Summing up these three pieces of evidence, despite a potent combination of subsidies and coercion, the adoption of bitcoin for payments hasn't occurred. Bitcoin usage in El Salvador is, if anything, regressing. Now that required acceptance of bitcoin is being rescinded, I suspect that it's only a matter of time before all the large businesses that introduced bitcoin payments, like McDonald's and Walmart, drop that option. With the government no longer coercing them to accept bitcoin payments, there's no commercial incentive to continue down that path.

It was IMF pressure on Nayib Bukele that finally got him to give up his bitcoin experiment. But the IMF was doing Bukele a favor, really, because the whole thing was already a failure, as I've explained with the charts above. Cancelling it outright would have been embarrassing to Bukele, but now he can deflect attention from himself and blame the IMF.

Why the failure, and what have we learned?

There is a very big hurdle that has prevented El Salvador's one-two punch of subsidies and coercion from working: bitcoin is intrinsically ill-suited to perform as money. 

The stuff is innately volatile, and so risk-shy individuals don't dare hold it or use it for payments. Risk-seekers can tolerate that volatility, but they expect to be rewarded by a dramatic price rise, and so they refuse to use their bitcoins for payments because they could miss out on the jump. The net result is that no one, neither society's risk-seekers nor its risk-avoiders, ends up paying with bitcoins. Only a tremendous amount of subsidies and coercion will ever overcome their natural preferences, but no sane government would ever try to bring those levels of coercion to bear. (And speeding things up with options like Lightning doesn't change this equation.)

The saddest thing about El Salvador's bitcoin experiment is that all sorts of time and resources have been wasted. El Salvador is not a rich country. The money spent on building and operating Chivo, compliance by businesses, bitcoin signage, and subsidies could have been better deployed on more important things like health and education.  One hopes that other countries learn from this experience and avoid going down the same route that El Salvador did. Brazil, which deployed its wildly popular PIX payment system around the same time as El Salvador launched its Bitcoin Law, provide helpful guidance.

More broadly, I'm hoping that El Salvador's failure finally kills off Satoshi's very misguided dream of bitcoin as electronic cash. I once was a believer in that dream, but for all the reasons I wrote in December, I've long since given up on any chance of bitcoin becoming a widely-circulating currency. But a lot of people continue to sacrifice their careers, time and resources to following Satoshi. Many of these are brilliant people. We want them to be creating valuable things for society. Alas, despite all sorts of evidence that bitcoin payments are a dead end, they continue to hit their heads against the wall, using excuses like government interference. 

Guys, Satoshi's dream is a mirage, a delusion, a hallucination. A government just flexed its muscles for four long years to get bitcoin into circulation, and that still didn't work. The lesson here: bitcoin is a bad payments tool and will never become widely-used electronic cash. It's time to move on.

---

*The BCB won't say how it collects this data -- according to the Salvadoran press there are legal limits on how much it can disclose -- but it describes the series as being compiled from administrative records that it receives from "cryptocurrency digital wallets." Reading between the lines, this probably includes Chivo data and any other regulated cryptocurrency service that stores customer crypto and reports to the BCB. (Because Chivo allows both U.S. dollars and bitcoins to be transferred, the BCR's data may be a mix of the two units, muddying the waters.) I think it's safe to assume that the BCR data does not include bitcoin remittances made via non-custodial services, say like Muun wallet or Blue wallet. However, since most of the governments carrot's (i.e. no fees) require the use of Chivo, it's probably a safe assumption that the average Salvadoran uses Chivo for bitcoin transfers, so the BCR data--which almost certainly includes Chivo--is fairly representative of overall usage.

Trump claims US banks can't open in Canada—US banks disagree

In what seems to be an effort to extort Canada for additional benefits, Donald Trump complained yesterday on social media that CANADA DOESN'T EVEN ALLOW U.S. BANKS TO OPEN OR DO BUSINESS THERE. And so according to Trump, Canada doubly deserves to be disciplined with tariffs.

Well, if it's true that U.S banks aren't allowed to do business in Canada, then why in god's name is one of the U.S.'s largest banks doing business in downtown Toronto?

Citigroup Place, 123 Front St. West, Toronto, Ontario, Canada

Citi has been operating in Canada since 1919 and currently has 1,700 Canadian employees. According to OSFI, Canada's bank regulator, the bank earned C$35 million in Canada in the first three quarters of 2024 and has C$5.49 billion in Canadian assets as of September 30, 2024. 

In short, Trump was either lying, misinformed, crazy, or some combination of those three.

Canada allows foreign banks to enter our banking industry by requiring them to set up a domestic subsidiary and applying for a Schedule II banking charter. Schedule II banks can operate in all of the same lines of business as mainstay Canadian banks (i.e. Schedule I banks) like Royal Bank or Bank of Montreal. There are 16 Schedule II banks in Canada, three of which are American. (In addition to Citi, the other two are Amex Bank and JP Morgan.)

Some folks on social media tried to reinterpret Trump's complaint: "But JP, what Trump really meant to say is that Canada doesn't allow U.S. banks to serve retail customers." As proof they cited the fact that if you walk into a Citi office in Canada, Citi won't allow you to open a personal chequing account.

The reason that Citi won't give you a personal chequing account isn't because the rules prevent them from doing so. Rather, Citi (along with Amex and JP Morgan) have chosen not to enter the Canadian retail banking market, preferring to focus instead on other types of Canadian banking, like commercial and investment banking. If Citi, for instance, wanted to set up a retail branch network, it could. In fact, Citi once had a small five-branch retail banking network in Vancouver and Toronto, offering personal chequing and savings account, term deposits, loans, mortgages, mutual funds and RRSPs. But it sold out in 1999 to Canada Trust, which was ultimately bought by TD Bank.

Other foreign banks have also set up Schedule II banks with a retail presence, only to sell out to domestic banks. HSBC Canada, owned by its British parent, became Canada's seventh largest bank—one that was notably successful in offering mortgages to retail customers—but was recently offloaded by its parent to Royal Bank, a Schedule I bank. ING Canada, owned by Dutch-based ING Bank, created one of Canada's most popular discount retail banks, ING Direct, but sold it to Scotia Bank in 2012, which rechristened the discount bank Tangerine Bank.

The lone Schedule II foreign bank I'm aware of that still serves retail customers is ICICI Bank, which is owned by its Indian parent.

Why are U.S. and foreign banks reticent to compete in Canada's retail banking market? Contrary to perceptions that Canadian banking is slow and lazy, it's actually quite difficult to make much headway in Canada. The Big-5 banks, plus National Bank, which counts as half a big bank, have built strong retail branch networks that span the entire country. They compete rigorously for consumer deposits, offering higher interest rates than U.S. banks offer to Americans, suggesting a more cut-throat market than south of the border. In short, U.S. banks don't have the cojones to cross the border and compete head-to-head against Canada's more competitive behemoths. Citi already tried. It gave up.

By contrast, the U.S. is an easier market for a foreign bank to enter because its banking industry is more fragmented. And many Canadian banks have entered, with TD Bank and Bank of Montreal occupying 10th and 13th spot respectively on the list of largest U.S. banks. This fragmentation is the residue of the U.S.'s refusal (until recently) to allow banks to set up branches across state lines. By contrast, Canada has always had fairly permissive rules about establishing cross-country banking networks. The irony here is that Trump's complaints about lack of openness best apply to the U.S., historically the culprit when it comes to tamping down the spread of banking.

Canadian banks' U.S. and international exposure has increased over time. A recent Bank of Canada study finds that our banks now have more foreign liabilities (i.e. deposits) than domestic liabilities. (See chart below). More precisely, 57% of all Canadian banks' liabilities are now foreign. As for our banks' asset mix, foreign assets are poised to surpass domestic assets in the next year or two, if trends continue.

Rising Canadian bank exposure to the rest of the world. Source: Bank of Canada

The reason for this outward migration is clear. Canada's saturated retail banking market offers few opportunities for growth, but other parts of the world are less saturated, and so these jurisdictions offer Canadian banks ideal avenues for acquisitions and growth.

This gives us an additional vantage point for viewing Trump's absurd comments about Canadian banking. He may not be saying that Canada's banking system is closed, but that the U.S. banking system is now effectively shut off to additional acquisitions by Canadian banks, as part of some sort of America First banking policy. This implicit threat of a foreign banking blockade may explain, in part, why the price of Canadian bank stocks fell so much more than the broader Canadian market yesterday. Their avenues for growth may have just narrowed.

Is it better to bribe Trump by purchasing his memecoin or his stock?

Noah Smith writes a provocative article about memecoins as a novel mechanism for bribery payments. A foreign dignitary looking to gain influence over Donald Trump would like to pay him a giant bribe, but doing so directly is prohibited by all sort of laws. Luckily, Trump has just issued his own memecoin, TRUMP, of which Trump owns 80% of all coins. So why not just buy the TRUMP token, thereby pushing its price up and gifting Trump with even more wealth, in return gaining a degree of influence over policy?

The best part is that no money actually changes hands, so it's probably less risky from a legal perspective. The dignitary can just plead "I thought it would go up!", says Noah.

Now, I'm not so sure that crypto is ushering in anything unique here. Consider that Donald Trump also owns shares of Trump Media & Technology Group Corp (DJT), which are NASDAQ-listed "tradfi" shares that predate crypto. Why not just buy DJT shares, pump their price higher, and collect favors from Trump? No crypto involved. 

In fact, a year before Noah wrote his article about memecoin bribery, Robert Maguire of Citizens for Responsibility and Ethics in Washington (CREW), worried about precisely such a scenario. Any entity wanting to "cozy up" to Trump need only buy a bunch of DJT shares on the NASDAQ, enough that they "get Trump’s attention, but low enough that it doesn’t break the five-percent threshold that triggers SEC disclosure."

Consider that Donald Trump and family members hold a 59% ownership stake in DJT equity, which isn't too different from the 80% of TRUMP that they own. Both assets have market caps of around $7 billion. So pushing up the price of DJT will certainly enrich Trump just as much as trying to nudge TRUMP higher. So here's my question: What's the best way to bribe the current President of the United States of America, by pumping the TRUMP memecoin or pumping old-school DJT shares?

Before answering it, I want to pause for a moment to reflect. The fact that I am even writing a blog post on the topic of bribing an American president shows how far along a certain dystopian financial timeline we have gone. Back to the timeline.

I see two reasons why the memecoin probably presents a better pseudo-bribery option than the tradfi stock. 

The first reason is that it's safer to pull off. DJT is listed on just one exchange; the NASDAQ. And the NASDAQ exists in the U.S., which has the most robustly-regulated and well-trusted securities markets in the world. One duty the government requires of NASDAQ is that it surveil transactions in real-time for abusive trading behaviour, so any sketchy DJT purchases could end up being reported by NASDAQ to the authorities. Furthermore, to get access to NASDAQ-listed shares, a brokerage account is required, and that'll require the would-be briber to pass through the brokerage's identity checks.  

On top of that, systems like the Consolidated Audit Trail, a government-mandated system tracking U.S. equity and options trades, gives regulators themselves a means to monitor market activity and investigate potential misconduct.

So a foreign dignitary is taking a bit of a risk if he or she goes the DJT route.

By contrast, the TRUMP memecoin is hosted on a blockchain, basically a borderless and open decentralized database, not a carefully-guarded database confined to the U.S. The result is that TRUMP can be listed anywhere, including on shady offshore crypto exchanges like ByBit or KuCoin, which surely aren't checking customers for pumps. To boot, these offshore exchanges perform only cursory identity checks, if any.

To further protect him or herself, a would-be briber can initiate the pump by sending funds from an offshore exchange, say KuCoin, to a decentralized exchange, or DEX, and only then push the price of TRUMP higher. DEXes are even more hands-off than offshore exchanges; they don't perform any surveillance or identity checks.

The riposte to this is that all blockchain transactions are public and observable, so a bribe conducted on a DEX could be traced. Ok, sure. But while blockchain transactions are visible, they aren’t directly tied to real-world identities. Blockchains are pseudonymous. It's a bit like going to a masked ball. Everyone can see who the dancers are, but as long as everyone has their mask on a degree of anonymity is preserved.     

So to safely get away with bribing Trump, it sure seems that his memecoin is the better option than NASDAQ-listed DJT.

Now for the second reason why the memecoin is better for bribery: it packs more punch per dollar.

A memecoin lacks what equity researchers refer to as fundamental value. Its price is solely a function of Sam's expectations of what future buyers like Jill will pay for it, with Jill's expectations conditioned on what she thinks Sam will pay. They are pure bundles of speculative energy. As I've referred to them in other posts, memecoins are decentralized ponzi games, zero-sum lotteries, or Keynesian beauty contests.

By contrast, DJT is a stock, and stocks provide their owners with a claim on the underlying firm's 1) profits and 2) its assets in case it is eventually wound-up. There is a "something" that buyers and sellers can coordinate on, so that unlike a memecoin, a stock is more than a pure nested expectation games. That's not to say that stocks don't have a big "meme" component (think Gamestop), but the degree to which this guessing game is played with stocks is unlikely to ever reach that of memecoins.

The existence of fundamentals makes pumps less effective. As a pump begins to drive the price of DJT higher, the underlying fundamentals will start to give certain existing investors a reason to sell (i.e. "it's now too expensive relative to earnings"), and that selling will dull the pump. Since there are no fundamentals for TRUMP memecoin buyers to latch on to – any price is as good as another – a memecoin pump never gets throttled by fundamental sellers.

To sum up, someone who has $10 million to bribe Donald Trump will want to demonstrate to the President that their purchases drove the price of the target asset higher: it'll be far easier to demonstrate this by pumping the frictionless memecoin than the burdened-by-fundamentals stock.

Now, if you've gotten this far and think this post is actually about how to bribe Trump, it's not. It's about the often fascinating differences (or not) between crypto and traditional finance. In my view, they aren't really so different. Crypto fans may think there's a financial revolution going on, but there's nothing new under the sun.

You might wonder: is the frictionlessness of a memecoin, its lack of fundamentals, and the ensuing incredible ease by which it can be bribe-pumped a new feature that crypto has brought to the table? Not really. There's no technical hurdle preventing the NASDAQ from listing a non-blockchain version of the TRUMP memecoin on its own old-fashioned Oracle database. People could buy and sell this NASDAQ-listed meme-thingy instead of that blockchain version of TRUMP. But securities law gets in the way. Listing an unadulterated ponzi game on a national stock market has never been legal, at least not in my lifetime. Why putting one up on a blockchain is legal is beyond me, but look over there, the President just did it.

At the speed the train is leaving sanity station and heading to financial silly land, I suspect listing pure ponzis on the NASDAQ will soon be an accepted thing. Memeassets everywhere! Bribes for everyone!

Canadian guilt, Russian oil

We Canadians are overwhelmingly pro-Ukraine and anti-Putin, so when the CBC published an expose last week about "banned Russian oil" sneaking into Canada, it was read in despair by most of us. What an awful failure of Canadian sanctions policy. 

As with a lot of sanctions media coverage, I saw things a bit differently: "Not bad. We're doing our part!" That's because if you add some more context to the CBC article, the data that it presents can be read as good news.

The article takes issue with 2.5 million barrels of refined oil products made from Russian-produced crude that have been indirectly imported into Canada since the start of Putin's invasion of Ukraine in 2022. Given that around 1,000 days have passed since the invasion, that works out to roughly 2,500 barrels per day of Russian-linked refined oil products arriving on Canadian shores. (Analyzing oil flows on a per-day basis is industry standard and also makes it easier for our brains.)

In the grand scheme of things, 2,500 barrels per day is a drop in the bucket. Canada consumes around 1.6 million barrels of refined oil products per day, according to CAPP, which includes stuff like gasoline, diesel, and jet fuel. So just 0.1% of our consumption is Russia-tainted. Even so, every barrel matters, and we should strive to avoid any contribution to Putin's war chest.

But there's more context. 2,500 barrels per day of Russian refined oil products is far less than what we imported prior to the war. According to the Canada Energy Regulator (CER), between 2017 and 2022 Canada was regularly importing around 10,000 barrels per day of refined petroleum products directly from Russia (see chart below). After banning imports of Russian crude and refined oil products, Canada's direct imports fell to zero in 2023. Into this void, indirect imports of 2,500 barrels per day of Russian-linked refined products, the flows that the CBC spotlights, have emerged.

A 75% decline from 10,000 barrels per day to 2,500 barrels per day is not too shabby.

Canada's direct imports of Russian refined petroleum products, which hit zero in 2023. Source: CER

2,500 barrels is still not zero. But we can also take comfort from the fact that those barrels are not as profitable for Russia as they used to be. In the pre-war era, Canada was importing refined petroleum products directly from Russia, but in the post-war era we are importing Russian oil indirectly via a third-party, India. More specifically,  oil in its raw form -- crude oil -- is being shipped all the way from Russia to India by tanker, where it is upgraded by Indian refineries, and only then is it onshipped to Canada.

This new workflow is a big downgrade for Russia. Before it can be used, crude oil has to be converted into pricier consumable types of fuel like gasoline for cars and jet fuel for planes. Upgrading crude oil creates extra profits for whoever does it. Russia's refineries used to capture the entire upgrading margin. They refined the raw oil after it was pulled out of the ground and then regularly sent 10,000 barrels per day of the final product to Canada. But now India is capturing those extra profits on the 2,500 barrels per day that are sent to Canada.

So not only has the quantity of Russian-linked refined products imported by Canada shrunk by 75% since the war began, but thanks to the interposition of Indian refiners at the expense of Russian ones, the quality of Russia's revenue stream has been downgraded: pound-for-pound, Russia's indirect exports to Canada are a far less lucrative for Putin than they were back in 2021, because his refining margin has disappeared.

Compounding Russia's woes is the much more circuitous route that its oil must now take. Prior to 2022, Russian refined oil exports were loaded onto boats in Russian ports like Saint Petersburg and shipped via the Baltic Sea to Canada, around 4,000 nautical miles away. That's a 15-day voyage according to Sea-Distances.

These days, that 15-day voyage has tripled, even quadrupled. First, Russian crude oil must travel from the Baltic to India, a 7,500 nautical mile journey that can take 30 days. That's if it goes through the Suez canal. Passing around the southern tip of Africa amounts to a 12,000 mile trip taking up to 50 days. Once refined in India, the product must travel another 8,000 miles from India to eastern Canada. 

What an incredible amount of travel to get a barrel of Russian refined oil to Canadian markets! A good way to visualize these new transportation frictions is provided by the Kyiv School of Economics, which charts the volume of Russian oil being transported by oil tankers over time. Thanks to the forced rerouting of crude to less efficient routes as countries like Germany and Canada close their borders to Putin, Russia's oil on water is 163% higher than the pre-invasion average.

Record volumes of Russian oil on water is not a good thing for Putin. It mean higher transportation costs. Source: KSE

The extra transportation and insurance costs that "oil on water" entails inevitably eat into the final price that Russia can negotiate with buyers like India for its barrels of crude. For these long distances to be financially worthwhile for Indian businesses, they will only buy Russian crude at a discount to the going world price. According to the Dallas Fed, the Russia discount regularly clocks in at around $20 below the market price. This constitutes a big step down for Russia -- prior to the war it was receiving the full world price.

The upshot is that Canadian imports of Russian oil are down, and even though some Russian refined petroleum products are indirectly making their way to Canada, this is only after we've extracted our pound of flesh from Putin by forcing him to give up his refining margin and by obliging him to accept a crude oil price discount on account of distance traveled. So let's take some pride from that.

Does that mean we shouldn't do anything about our indirect imports of Russian oil product?

I want to clarify that Canada isn't importing "banned" products or breaking Russian sanctions. For better or for worse, the coalition's sanction on Russian crude oil have been designed to allow crude to continue to flow around the world, the intent being to avoid a big spike in oil prices while still hurting Russia. The 2,500 barrels of indirectly-refined refined oil we get each day are fair game.   

But that doesn't mean Canadians should do nothing. The CBC article is a good effort to name-and-shame certain Canadian importers that are accepting Russian-linked crude from third-parties, including Everwind Fuel's Point Tupper oil storage facility in Nova Scotia. C'mon, Everwind. Why not choose better trading partners, ones who aren't acting as go-betweens for Putin?

However, the best step we can do to counter Russia is to focus on producing more renewables, crude oil, and other commodities, as well as to find reliable ways to get these resources to market. 

Unlike Europe and the U.S., which have plenty of economic and financial heft, Canada doesn't have any sizable economic chokepoints that we can lever to hurt Russia. We could cut down on the 2,500 barrels per day of Russian-linked oil imports, but as laudable as that might be it doesn't constitute a genuine chokepoint. Canada's edge is that our economy is remarkably similar to Russia. Both of us extract a bunch of resources. The more we compete with Putin in resource extraction, the more we reduce the prices he relies on, thus impairing his ability to fund his invasion of Ukraine.

Here’s why we tolerate fake check scams

Source: Better Business Bureau

The daily news is filled with personal stories about bad experiences with banks. Here’s a recent example. In November 2024, a charity inadvertently accepted a fake check from a would-be donor. The charity's bank allowed the charity to deposit the check, crediting it with the funds. A few days later the bank discovered the fake, but only after the charity had transferred some of the "donated" funds back to the donor, a scammer. The bank then raided the charity’s bank account for the full amount, leaving the charity out of pocket.

Readers will find this story disturbing. Banks are supposed to protect scammed customers, not kick them while they are down, especially charities. Many of us will wonder if the payments system needs to be fixed. 

But payments systems are complex organisms that have evolved over many centuries. When viewed from afar, what appears to be a glitch is often actually an element of a balanced whole. Solving the problem of fake check scams would upset this balance, introducing new complications further down the payments process.

Let’s look a bit more closely at the scam.   

The anatomy of a fake donation scam

Approached by a stranger who wanted to donate money, Motorcycle Missions – a Texas-based charity that helps helps veterans and first responders with post-traumatic stress disorder – was sent a $95,000 paper check in the mail. Motorcycle Missions proceeded to deposit the check at its bank, Chase, which immediately credited the charity for the full amount. A few days later, the stranger asked for some of the money back. His assistant had made a mistake, the stranger claimed, and the check was supposed to be for just $50,000. So Motorcycle Missions helpfully wired $45,000 to the stranger's account.

But it was a scam. The check, donation, and donor were all fake. Unfortunately, the $45,000 that flowed out of Motorcycle Missions's account and into the account of the scammer was very real. Chase promptly seized $95,000 from the charity’s account as compensation for the amount of money that it had created upon accepting the fake check. Because it had paid out $45,000 to the scammer, the charity was left $45,000 out of pocket.

Unjust? It seems so. Charity gets tricked by scammers only to have his fat cat banker, the one who processed the check, refuse to help him. Unfortunately, scams like this are all too common.

Exploiting the check timing gap

In addition to exploiting the constant need of charities for funding, fake donation scammers exploit a weakness in the check payments system. Specifically, they target the timing gap between a bank’s initial crediting of funds to a depositor’s account and the point at which a check’s authenticity is finally verified.

When someone accidentally brings a scammer's fake check to the bank, banks will do their best to catch it. But some fakes sneak through. This is where the timing gap opens up. 

The amount indicated on the face of the fake check is credited to the depositor’s bank account. The customer can then spend it (or be duped into paying off their scammer). But behind the scenes the actual processing and settlement of the fake check grinds on. A few days or even weeks later the check’s true nature is eventually discovered. But, by then, the sneaky scammer has already received their electronic payment.

So why don’t we just fix things by removing the timing gap?

The tradeoff between speed and security

The payment system is a combination of tradeoffs and sacrifices. We can remove the check timing gap, but this means introducing other weak spots into the checking system.

For instance, we could easily put a quick end to all fake donation scams by stipulating that banks only credit funds to depositors’ accounts after the paper check has been irrevocably confirmed to be legitimate. In that case, if Motorcycle Missions were to accidentally deposit a fake check, it needn’t worry. The check will eventually be caught and the charity won’t be hit with a $45,000 charge. Knowing that the system has a perfect defence, scammers would stop check scamming.

But there are consequences to fixing the timing gap. All of us check-users would now be required to wait days, even weeks in some cases, before we can spend our money.  

Speed is an important feature of any payments system. Because Motorcycle Missions was probably a longtime and trusted customer, Chase allowed the charity to use the amount printed on the face of the check immediately, even though the check hadn’t definitively settled. In bank-speak, banks will lend or grant provisional credit to their check-cashing customers. 

This service is important to us. We may have bills due two days from now. We can’t wait weeks for our checks to be 100% settled.

In fact, check speed is considered so important that according to U.S. law, specifically Regulation CC, all checks deposited must be made available for withdrawal by the business day after the day of deposit. Since the only way for banks to meet these standards is to grant provisional credit, the timing gap is legally baked into the system. 

And into this gap scammers stream. We accept these chinks in the check system because we want the overall process to move more quickly.

Who bears the costs of speedy checks?

If society has decided to tolerate the fake check problem in order to get more speed out of the check system, someone has to bear the extra credit risk of these fakes. Which unfortunate party is held responsible?

Commercial law places this risk squarely in the lap of bank customers. (See also). When a bank puts money in a customer’s account upon deposit of a check, it is lending to them. As with any loan, the lender can collect should the borrower default (say, because the check was fake). 

That’s exactly what happened with Motorcycle Missions. It was granted $95,000 in provisional credit after depositing a fake check, only for the loan to be called when the fake was discovered.

We might not think this is fair. Surely banks are better at evaluating whether a check is fake or not than customers. So why not shift the burden of absorbing the cost of fake checks onto banks and away from the public? 

We could certainly design a payments system along these principles. Now when Motorcycle Missions deposits a fake $95,000 check, and its bank credits it the amount, Chase must absorb the $45,000 expense when the fake is discovered.

In this system, not only would banks make check payments go fast by offering provisional credit. They would also take on all of the risk of fake checks. What a win for bank customers! We’d get maximum speed and complete safety. 

But it’s not that easy. 

To absorb the costs of extra credit risk, banks like Chase would probably increase monthly checking account fees. Rather than passing on the costs of fake checks exclusively to the scammed customers, as in the current system, every customer would bear part of the burden in the form of higher fees. 

This spreading-out of costs is a win for vulnerable customers who, given the precariousness of their business or personal lives, are more likely to fall for fake check scams and be hurt by associated penalties. But the rest of the bank’s customers may not be as thrilled, preferring charges fall on those who make mistakes. 

In sum, what happened to Motorcycle Missions is unfortunate. But solving the problem of fake checks isn’t as easy as one might think. Changes to a tightly-wound system like the check system involve tradeoffs. You don’t get something for nothing.

P.S.: Please consider donating to Motorcycle Missions here.

[A version of this article was originally published at the AIER's Sound Money Project.]

Stablecoins are non-fungible, bank deposits are fungible

On Twitter/X, I recently suggested that the network effects of the stablecoin market are massive. Tether, which has four times more wallets than all other stablecoins, is locked-in as the stablecoin lingua franca, just like English has been locked-in as the global language of business. 

In case you've missed the trend, stablecoins are fiat money (primarily U.S. dollars) that are issued on a new type of database called a blockchain. The total value of stablecoins in circulation has grown from $0 to over $200 billion in a decade, with Tether dominating at $138 billion.

When I said at the outset that the stablecoin market is governed by network effects, what I meant is that a positive feedback loop exists whereby the value that a network (i.e. languages or stablecoins) provides to users increases as more users join the network. Once a given stablecoin has entered into this virtuous loop, other issuers cannot join in, and will have troubles competing. It's a winner take all market that Tether and its stablecoin USDt (and perhaps smaller competitor USDC, issued by Circle) have already won.

Larry White, a monetary economist who I've mentioned a few times on my blog, asked me why I think network effects are present in the stablecoin market. We don’t see network effects with other U.S. dollar payment media like checkable deposits, Larry points out (and I agree), so it's not clear why we should see this with stablecoins.

Here's my logic.

Stablecoins aren't fungible, bank deposits are

The key is that while U.S. dollar stablecoins—Tether's USDt, Circle's USDC, PayPal USD, etc—are pegged to the dollar, and thus seem to be alike, they are not actually completely alike. That is, they are not fungible with each other. 

Fungibility is one of my favorite words, and I write about it quite often on this blog. It means that members of a population are interchangeable, or perfectly replaceable with each other. All grams of pure raw gold are interchangeable. Not all grams of pizza are alike—pizza is non-fungible.

U.S. dollar bank deposits (say Well Fargo dollars and Chase dollars) are fungible with each other. Rather than being independent, they are fused together as homogeneous and singular U.S. dollars. A Chase dollar is just as good as a Wells Fargo dollar for the purposes of making payments.

That's not the case with stablecoins, which are like pizza. Or better yet, in the same way that Chinese yuan and UAE dirham are pegged to the dollar but remain independent currencies, each U.S. dollar stablecoin is pegged to the dollar but functions as its own distinct non-fungible currency. For the purposes of making payments, one stablecoin is not as good as another one, just like how dirham balances aren't perfect replacements for yuan.

The reason behind this difference is that U.S. banks cooperate with each other by accepting competitor's money at par on behalf of their customers. For instance, I can take a Wells Fargo check to my Chase branch and Chase will accept it 1:1 even though it represents a competing bank's dollar. Or I can send an ACH payment directly from Wells Fargo to Chase, and Chase will accept that Wells Fargo dollar at par and convert it into a Chase dollar for me. 

The effect of this reciprocal acceptance is that all U.S. banking dollars are tightly knit together, or interchangeable. A fungible standard has been created.

I can't perform these same actions with stablecoins. I can't send 100 USDC to Tether to be converted into 100 USDt, nor send 100 USDt to Circle, which issues USDC, to be converted into 100 USDC. Stablecoins issuers are loners. They've chosen to avoid banding together to weave a unified U.S dollar stablecoin standard.

This lack of standardization explains some weird things in the stablecoin market, like why there are so many markets to trade USDt for USDC (see below). Notice that the clearing price in these stablecoin-to-stablecoin markets is never an even $1, but always some inconvenient price like 0.991 or 1.018.

Some of the multiple markets for trading USDt for USDC, all at varying prices Source: Coingecko

 

There is no equivalent trading market for Chase-to-Wells Fargo balances or TD-to-Bank of America dollars. These banks' dollars are perfectly compatible and don't require such markets.

The advantages of a single dollar standard

Harmonization is useful. Anyone can walk into a McDonald's and purchase a Big Mac for $5.69 with whatever brand of bank dollar they want. Money held at small banks is just as useful as money at massive ones: the Bank of Little Rock may only have five branches, but its dollars are accepted at McDonald's all across the world, on par with those of Chase, America's largest bank.

McDonald doesn't accept stablecoins, but if it did, it would have to offer multiple prices for a Big Mac: i.e. 5.73 USDt and 5.68 USDC. Each stablecoin serving as its own particular unit of account is inconvenient, both for McDonald's and its customers. PayPal USD probably wouldn't even be accepted at McDonald's: it's too small.

The lack of standardized stablecoin market becomes even more awkward in asset markets. If you want to buy $1 million bitcoins on, say, Binance, there's a whole array of different U.S. dollar stablecoin markets available, including bitcoin-to-USDt, bitcoin-to-USDC, and bitcoin-to-FDUSD. (FDUSD refers to First Digital USD, a medium sized stablecoin.)

The table above shows the prices of bitcoin and ether on Binance, the world's largest crypto exchanges. Notice that liquidity in both Binance's bitcoin and ether trading market is compartmentalized into different stablecoins rather than being fused into a single homogeneous US dollar-to-bitcoin market. Source: Coingecko

You can forget about easily buying bitcoins with PayPal USD stablecoins. No crypto exchange offers that trading pair; PayPal USD is too small to be worth the hassle.

This has the effect of fragmenting the liquidity of the stablecoin market into different buckets. Instead of stablecoins-in-general having a certain level of marketability, each individual stablecoin has its own distinct liquidity profile in asset markets.

In contrast, the liquidity that a Wells Fargo dollar, a Bank of Little Rock, or a Chase dollar provides to their owner in the context of asset markets has been unified into a collective whole. If you want to buy shares of Blackrock's iShares Bitcoin ETF, there isn't a separate market for Wells Fargo-to-bitcoin or Chase-to-bitcoin. As for Bank of Little Rock dollars, they are just as fit for bitcoin purchases as its much largest competitors.

A winner-takes-all market

Now we can understand why network effects dominate the stablecoin market.

If you want to start using stablecoins to trade crypto or buy stuff, you will always be arm-twisted by market logic into choosing the largest most liquid stablecoin. And your decision to go with the largest one makes that stablecoin a little more liquid, thus solidifying its pole position.

Selecting a smaller stablecoin like PayPal USD makes little sense. McDonald's will never accept it, and there are many crypto assets that you won't be able to buy with it. Even when certain PayPal USD trading pairs are available, the bid-ask spreads will be wide, imposing much larger costs on you than if you simply went with a larger stablecoin. Thus network effects, working in reverse, repel uptake of PayPal USD.

The unsafe stablecoin is the largest

Tether remains the largest stablecoin, despite being one of the most unsafe stablecoins. (USDC does not get top marks for safety, either.) Network effects explain this.

Stablecoin rating agency Bluechip awards Tether a D rating, noting that it is "less transparent and has inferior reserves... USDT is not a safe stablecoin". Under normal conditions (i.e. those not characterized by network effects) the safest stablecoins would have long-since displaced Tether from its leading spot. But in stablecoin markets, the safest stablecoins—Gemini USD, PayPal USD, and USDP, all rated A or A- by Bluechip—remain insignificant players. The virtuous circle in which Tether is locked dominates all other factors.

These are the best-ranked fiat stablecoins according to Bluechip. But they are also tiny, with market capitalization below $1 billion. There appears to be no point trying to be a safe stablecoin, since the network effects arising from liquidity completely dominate any safety concerns that users might have.

Eyeing Tether's profits, new competitors are entering the stablecoin market. But this is a game they probably shouldn't bother playing. PayPal arrived last year with PayPal USD, but to date it remains mostly irrelevant, despite huge growth in the overall stablecoin market over the same period. Ripple and Revolut are also slated to bring out their own products. They're also destined to mediocrity, because they're too late to make the jump into the virtuous loop that Tether and (to a lesser extent) Circle occupy. 

(There is one caveat. Should one of the two leaders eventually be shutdown for money laundering offenses or sanctions evasion, one of these also-rans could be vaulted into their spot.)

Might the stablecoin sector eventually migrate over to the unified fungible standard that characterizes banking deposits? 

No, that's probably not going to happen. For a fusion to occur, Tether and runner-up Circle, which issues USDC, would have to start accepting their competitors' stablecoins at par. But they won't go down this path, since that would kill off the network effect that gives them their unrivaled dominance over the rest of the pack. No, it's in the interests of the leaders for chaotic non-fungibility to continue. 

Alas, this lack of standardization may limit the stablecoin sector's broader potential to serve as a cohesive global payment alternative to the better-organized banking standard. Sometimes a bit of cooperation trumps competition.

Someone is wrong on the internet about the SWIFT network

There's a chart that has been circulating for a while now on social media that shows payments traffic on SWIFT, a key global financial messaging network. Below is a version from the Economist, but I've seen other versions too.

Source: The Economist

When banks make cross-border payments between each other, say euros to dollars, they need to use a communications network to coordinate the debiting and crediting of accounts, and SWIFT is the dominant network for doing so. Think of it as WhatsApp for banks.

Here's the problem. The main conclusion that pundits are taking away from the chart is the wrong one. Most of them seem to think that the chart illustrates an erosion in the euro's global popularity (i.e. de-euroization) and a simultaneous move towards the dollar for global trade. The Economist, which entitles its chart "Dollarisation," is also guilty.

Today I'm going to show you why that's the wrong conclusion; there is no SWIFT-related de-euroization. The reason for going through this effort isn't just because it's fun to dunk on wrong folks. It can also teach us some interesting things about the massive bits of unsung payments infrastructure that underlie our global economy, including not only SWIFT but also Europe's T2 and the U.S.'s Fedwire, two of the world's busiest financial utilities.

Let's dig in. The problem with trying to analyze charts of SWIFT messages across various currency jurisdictions is the data isn't necessarily comparable. As I said at the outset, commercial banks around the world use SWIFT to coordinate cross-border payments with other banks, and that is what people are hoping to measure with the SWIFT chart at top. But muddying the waters is the fact that in the EU, banks also use SWIFT for domestic payments. Here's how:

The most important bit of payments infrastructure in both the U.S. and EU are their respective central bank's large-value payments (or settlement) systems. When commercial banks make crucial domestic payments with each other, typically on behalf of their customers, these payments are settled in real-time using each commercial banks' respective account at their central bank, in the U.S.'s case the Federal Reserve, and in Europe's case the European Central bank, or ECB. The ECB's mechanism for settling payments is known as T2 (and previous to that, Target2.) The Fed uses Fedwire.

To coordinate this "dance of databases," the central bank and participating commercial banks need to communicate clearly and rapidly with each other, and that's where financial messaging networks come in. Fedwire doesn't use SWIFT for this. It comes fitted-out with its own proprietary messaging network for member banks. But the ECB has chosen a different setup. Up until 2023 the ECB had outsourced all messaging to SWIFT, a bank-owned cooperative based in Belgium.

Now you may be able to see why comparing the amount of euro payments made using SWIFT messages to dollar payments made using SWIFT is an apples to oranges comparison. Both data sets include cross-border payments, but the EU dataset also includes a large amount of domestic payments. The U.S. dataset doesn't.

This means that the variations in the amount of euro payments messages that get captured in the chart at top may not reflect dramatic geopolitical shifts like "de-euroization, but may be linked to more banal things like changes in local EU payments habits. And indeed, I'm going to show why domestic and not international factors explain the 2023 drop in the euro share of SWIFT messages. 

In 2023, the ECB replaced its Target2 settlement system with a new system called T2. Two key upgrades were introduced with T2 that ultimately affected SWIFT message flows. 

The first of the upgrades was a new language for constructing messages, with the ISO 20022 messaging standard replacing the legacy MT messaging format. (I wrote about ISO 20022 in an article entitled The Standard About to Revolutionize Payments.) 

This change in payments lingo has had a big effect on the sum of SWIFT data displayed in the chart at top. Both the ECB and SWIFT provide explanations for this, but here is my shorter summary. Prior to the 2023 changeover, a type of euro payment known as a liquidity transfer was regularly captured in the SWIFT chart. A liquidity transfer occurs when a European commercial bank, which often has several accounts at the ECB, must rebalance between its accounts when one of them is running low. These within-bank liquidity transfer messages aren't terribly interesting and have nothing to do with global payments, but were included in the SWIFT dataset nonetheless up until 2023, thus fudging the results. 

With the arrival of ISO 20022, messages related to euro liquidity transfers are now conveyed using a new type of message. Thus the big decline in the euro's share of SWIFT messages in 2023 — liquidity transfers have effectively dropped out of the chart. This is a good thing, though, since the omission of these relatively unimportant within-bank transfers means we're getting a cleaner and more accurate signal.

The second upgrade introduced in 2023 was the opportunity for European commercial banks to choose among multiple messaging networks for accessing T2. Under T2's predecessor, Target2, banks only had one access choice: the SWIFT network. With T2, European banks can also use SIAnet, owned by the Nexi Group. (I wrote about this upgrade here, in which I described T2's switch from an older Y-copy topology to a network agnostic V-shaped topology.)

In that older post, I suggested that adding additional access points was a healthy step for Europe, since it meant more resilience should one network suffer an outage. And in fact, Europe is already reaping the benefits. When SWIFT failed for several hours on July 18, 2024, the ECB issued the following alert:

"T2 is operating normally. However, due to an ongoing SWIFT issue, some incoming messages do not reach T2 immediately. Similarly, some T2 outgoing messages might not reach the receiver immediately... There is no impact on traffic sent or received via NEXI. Participants may continue sending new instructions and queries to CLM/RTGS/CRDM. Updated information will be provided at the latest by 16:30."

Whereas an outage of SWIFT in 2021 or 2022 would have seriously slowed down Europe's financial activity, the addition of Nexi's SIAnet to the mix in 2023 limited the damage caused by the 2024 SWIFT outage. By contrast, the UK's central bank, the Bank of England, remains entirely reliant on SWIFT for messaging, and so the 2024 outage caused more disruption for Brits than Europeans, according to the Financial Times.

Unfortunately, I can't find any data on how many European banks have actually chosen to shift their T2 messaging needs over to Nexi. But I'd imagine that it isn't negligible, given that Nexi's SIANet is already being used by banks to access other key bits of Europe's payments architecture including STEP2, a pan-European automated clearinghouse. And so some non-negligible portion of the drop in the euro's share of SWIFT messages in the top chart is due to a shift away from SWIFT.

If the SWIFT chart at top doesn't mean what people think it means, what is the euro's status as a global trading currency? A 2024 article from the ECB clears this up. In short, the euro's international role hasn't eroded over the last few years. The de-euroization memes are all wrong.

The irony of all of this is that rather than reflecting a decline in Europe's status, the SWIFT chart illustrates the opposite. A bunch of healthy advances are driving the euro's share of SWIFT payments down, including a more accurate classification of financial messaging data thanks to a better messaging language, combined with a much needed de-SWIFTication of European messaging flows. It's not as juicy as euro critics make it out to be.

After twelve years of writing about bitcoin, here's how my thinking has changed

What follows is an essay on how my thinking on bitcoin has changed since I began to write on the topic starting with my first post in October 2012. Since then I've written 109 posts on the Moneyness Blog that reference bitcoin, along with a few dozen articles at venues like CoinDesk, Breakermag, and elsewhere.

An early bitcoin optimist

I was excited by Bitcoin in the early days of my blog. The idea of a decentralized electronic payment system fascinated me. Here's an excerpt from my second post on the topic, Bitcoin (for monetary economists) - why bitcoin is great and why it's doomed, dated November 2012:

"Bitcoin is a revolutionary record-keeping system. It is incredibly fast, efficient, cheap, and safe. I can send my Bitcoin from Canada to someone in Africa, have the transaction verified and cleared in 10 minutes, and only pay a fee of a few cents. Doing the same through the SWIFT system would take days and require a $35 fee. If I were a banker, I'd be afraid." [link]

I was relatively open to Bitcoin for two reasons. First, I like to think in terms of moneyness, which means that everything is to some degree money-like, and so I welcome strange and alternative monies. "If you think of money as an adjective, then moneyness becomes the lens by which you view the problem. From this perspective, one might say that Bitcoin always was a money," I wrote in my very first post on bitcoin. Second, prior to 2012 I had read a fair amount of free banking literature—the study of private money—so I was already primed to be receptive to a stateless payments system, which is what Bitcoin's founder, Satoshi Nakamoto, originally meant his (or her) creation to be. 

A lot of bitcoin-curious, bitcoin-critics and bitcoin-converts were attracted to the comments section of my blog, and we had some great conversations over the years. My bitcoin posts invariably attracted more traffic than my non-bitcoin ones, all of us scrambling to understand what seemed to be a newly emerging monetary organism.

My early thoughts on the topic were informed by having bought a few bitcoins in 2012 for the sake of experimentation, some of my earlier blog posts describing how I had played around with them. In 2013 I wrote about the first crop of bitcoin-denominated securities market (which I dabbled in)—predecessors to the ICO market of 2017. I also used my bitcoins to buy altcoins, including Litecoin, and in late 2013 wrote about my disastrous experience with Litecoin-denominated stock market speculation. In Long Chains of Monetary Barter I described using bitcoin as an exotic bridging currency for selling XRP, a new cryptocurrency that had just been airdropped into the world. I didn't notice it at the time, but in hindsight most of these were instances of bitcoin facilitating illegal activity, i.e. unregistered securities sales, which was an early use case for bitcoin.

Although Bitcoin excited me, I was also critical from the outset, and in later years my critical side would only grow, earning me a reputation among crypto fans as being a salty no-coiner. In a 2013 blog post I grumbled that playing around with my stash of bitcoins hadn't been "as exciting as I had anticipated." Unlike regular money, there just wasn't much to do with the stuff, my coins sitting there in my wallet "gathering electronic dust."

 "...the best speculative vehicles to hit the market since 1999 Internet stocks."

What my experimentation with bitcoin had taught me was that the main reason to hold "isn't because they make great exchange media—it's because they're the best speculative vehicles to hit the market since 1999 Internet stocks." But that wasn't what I was there for. What had tantalized me was Satoshi's vision of electronic cash, a revolutionary digital payments system. Not boring old speculation.

In addition to my practical complaints about bitcoin, I also had theoretical gripes with it. The "lethal" problem as I saw it back in my second post in 2012 is that "bitcoin has no intrinsic value." Over the next decade this lack of intrinsic value, or fundamental value, would underly most of my criticisms of the orange coin. Back in 2012, though, the main implication of bitcoin's lack of intrinsic value was the ease by which it might fall back to $0. As I put it in a 2013 article:

"Bitcoin is 100% moneyness. Whenever a liquidity crisis hits, the only way for the bitcoin market to accommodate everyone's demand to sell is for the price of bitcoin to hit zero—all out implosion" [link]

But if the price of bitcoin were to fall to zero then it would cease to operate as a monetary system, which would be a huge disappointment to those of us who were fascinated with Satoshi's electronic cash experiment. Adding to the danger was the influx of bitcoin lookalikes, or altcoins, like litecoin, namecoin, and sexcoin. In theory, the prices of bitcoin and its competitors could "quickly collapse in price" as arbitrageurs create new coins ad infinitum, I worried in 2012, eating away at bitcoin's premium. The alternative view, which I explored in a 2013 post entitled Milton Friedman and the mania in copy-paste cryptocoins,  was that "the earliest mover has superior features compared to late moving clones," including name brand and liquidity, and so its dominance was locked-in via network effects. Over time the latter view proved to be the correct one.

The "zero problem"

Despite my worries, I was optimistic about bitcoin, even helpful. One way to stop bitcoin from falling to zero might be a "plunge protection team," I offered in 2013, a group of avid bitcoin collectors that could anchor bitcoin's price and provide a degree of automatic stabilization. In a 2015 post entitled The zero problem, I suggested that bitcoin believers like Marc Andreessen should consider donating $21 million to a bitcoin stabilization fund, thus securing a price floor of $1 in perpetuity. 

No fan of credit cards, in a 2016 post Bitcoin, drowning in a sea of credit card rewards, I suggested that bitcoin activists encourage retailers that accept bitcoin payments to offer price discounts. This carrot would put bitcoin on an even playing field with credit card networks, which use incentives like reward points and cashback to block out competing payment systems.

My growing disillusionment

By 2014 or 2015, I no longer saw much hope for bitcoin as a mainstream payments system or generally-accepted medium-of-exchange. "For any medium of exchange to displace another as a means for buying stuff, users need come out ahead. And this isn't happening with bitcoin," I wrote in a 2015 post entitled Why bitcoin has failed to achieve liftoff as a medium of exchange, pointing to the many costs of making bitcoin payments, including commissions, setup costs, and the inconveniences of volatility.

In another 2015 post I focused specifically on the volatility problem, which stems from bitcoin's lack of intrinsic value. If an item has an unstable price, that militates against it becoming a widely used money. After all, the whole reason that people stockpile buffers of liquid instruments, or money, is that these buffers serve as a form of insurance against uncertainty. If an item's price isn't stable—which bitcoin isn't—it can't perform that role. 

Mind you, I did allow in another 2015 post, The dollarization of bitcoin, that bitcoin might continue as "an arcane niche payments system for a community of like minded consumers and retailers." I even tracked some of these arcane payments use cases, such a 2020 blog post on retailers of salvia divinorium (a legal drug in many U.S. states) falling back on bitcoin for payments after the credit card networks kicked them off, followed by a 2021 post on kratom sellers (a mostly-legal substance) doing the same. But let's face it, a niche payments system just wasn't as impressive as Satoshi's much broader vision of electronic cash that had beguiled me in 2012, when I had warned that "if I was a banker, I'd be afraid." 

The dollarization of bitcoin

By 2015 a lot of my pro-bitcoin blog commenters began to see me as a traitor. But I was just changing my thinking with the arrival of new data.

Searching for Bitcoin-inspired alternatives: Fedcoin and stablecoins

Bitcoin's deficiencies got me thinking very early on about how to create bitcoin-inspired alternatives. By late 2012 I was already thinking about stablecoins:

"What the bitcoin record-keeping mechanism needs is an already-valuable underlying asset to which it can be tethered. Rather than tracking, verifying, and recording the movement of intrinsically worthless 1s and 0s, it will track the movement of something valuable." [link]

Later, in 2013, I speculated about the emergence of "stable-value crypto-currency, not the sort that dangles and has a null value." These alternatives would "copy the best aspects" of bitcoin, like its speed and safety, but would be linked to "some intrinsically valuable item." A few months later I predicted that "Cryptocoin 2.0, or stable-value cryptocoins, is probably not too far away." This would eventually happen, but not for another few years.

My dissatisfaction with bitcoin led me to the idea of decentralized exchanges, or DEXs, in 2013, whereby equity markets would "adopt a bitcoin-style distributed ledger." That same year I imagined central banks adapting "bitcoin technology" to run its wholesale payments system in my post Why the Fed is more likely to adopt bitcoin technology than kill it off. In 2014 I developed this thought into the idea of Fedcoin, an early central bank digital currency, or CBDC, for retail users.

If not money, then what is bitcoin?

By 2017 or so, even the most ardent bitcoin advocates were being forced to acknowledge that Satoshi's electronic cash system was not panning out: the orange coin was nowhere near to becoming a popular medium-of-exchange. This was especially apparent thanks to a growing body of payments surveys (which I began to report on in 2020) showing that bitcoin users almost never used their bitcoins to make payments or transfers, preferring instead to hoard them. So the true believers pivoted and began to describe bitcoin as a store-of-value, or digital gold. It was a new narrative that glossed over Satoshi's dream of electronic cash while trying to salvage some monetary-ish parts of the story.

I thought this whole salvage operation was disingenuous. In 2017 I wrote about my dissatisfaction with the new store-of-value narrative, and followed that up with a criticism of the digital gold concept in Bitcoin Isn’t Digital Gold; It’s Digital Uselesstainium. (The idea that store-of-value is a unique property of money is silly, I wrote in 2020, and we should just chuck the concept altogether.)

But if bitcoin was never going to become a generally-accepted form of money, and it wasn't a store-of-value or digital gold, then what exactly was it? 

I didn't nail this down till a 2018 post entitled A Case for Bitcoin. We all thought at the outset that bitcoin was a monetary thingamajig. But we were wrong. Of the types of assets already in existence, bitcoin was not akin to gold, cash, or bank deposits. Rather, it was most similar to an age-old category of financial games and zero-sum bets that includes poker, lotteries, and roulette. The particular sub-branch of the financial game family that bitcoin belonged to was early-bird games, which contains pyramids, ponzis, and chain letters. Here is a taxonomy:

A taxonomy showing bitcoin as a member of the early-bird game family

Early-bird games like pyramids, ponzis, and chain letters are a type of zero-sum game in which early players win at the expense of latecomers, the bet being sustained over time by a constant stream of new entrants and ending when no additional players join. Pyramids and ponzis are almost always administered by thieves who abscond with the pot. Bitcoin, by contrast, was not a scheme nor a scam. And it was not run by a scammer. It was leaderless and spontaneous, an "honest" early-bird game that hewed to pre-set rules. Here is how I described it in a later post, Bitcoin as a Novel Financial Game:

"Bitcoin introduces some neat features to the financial-game space. Firstly, everyone in the world can play it (i.e., it is censorship-resistant). Secondly, the task of managing the game has been decentralized. Lastly, Bitcoin’s rules are automated by code and fully auditable." [link] 

This ponziness of bitcoin was actually a source of its strength, I suggested in 2023, because "it's tough to shut down a million imaginations." By contrast, if bitcoin had an underlying real anchor, like gold, then that would give authorities a toe hold for decommissioning it.

Bitcoin-as-game gave me more insight into why most bitcoin owners weren't using bitcoin as a medium-of-exchange. Its value as a zero-sum bet was overriding any functionality it had for making payments. In a 2018 post entitled Can Lottery Tickets Become Money?, I worked this out more clearly:

"Like Jane's lottery ticket, a bitcoin owner's bitcoins aren't just bitcoins, they are a dream, a lambo, a ticket out of drudgery. Spending them at a retailer at mere market value would be a waste given their 'destiny' is to hit the moon." [link]

If bitcoins weren't like bank deposits or cash, how should we treat them from a personal finance perspective? Feel free to play bitcoin, I wrote in late 2018, but do so in moderation, just like you would if you went to Vegas. "Remember, it's just a game."

Bitcoin is innocuous, don't ban it

By 2020 or 2021, the commentary surrounding bitcoin seemed to be getting more polarized. As always there was a set of hardcore bitcoin zealots who thought bitcoin's destiny was to change the world, of which I had been a member for a brief time in 2012. But arrayed against them was a new group of strident opponents who though bitcoin was incredibly dangerous and were pushing to ban it.

A vandalized 'Bitcoin accepted' sign in my neighborhood

I was at odds with both sides. Each saw Bitcoin as transformative, one side for the good, the other for the bad. But I conceived of it as an innocuous gambling device, one that only seemed novel because it had been transplanted into a new kind of database technology, blockchains. We shouldn't ban bitcoin for the same reason that we've generally become more comfortable over the decades removing prohibitions on online gambling and sports betting. Better to bring these activities into the open and regulate them than leave them to exist in the shadows.

Thus began a series of relax-don't-ban-bitcoin posts. In 2022, I wrote that central bankers shouldn't be afraid that bitcoin might render them powerless. For the same reason that casinos and lotteries will ever be a credible threat to dollar's issued by the Fed or the Bank of Canada, neither will bitcoin.

Illicit usage of bitcoin was becoming an increasingly controversial subject. Just like casinos are used by money launderers, bitcoin had long become a popular tool for criminals, the most notorious of which were ransomware operators. My view was that we could use existing tools to deal with these bad actors. Instead of banning bitcoin to end the ransomware plague, for instance, I suggested in a 2021 article that we might embargo the payment of ransoms instead, thus choking off fuel to the ransomware fire. Alternatively, I argued in a later post that the U.S. could fight ransomware using an existing tool: Section 311 of the Patriot Act. Which is what eventually happened with Bitzlato and PM2BTC, two Russian exchanges popular with ransomware operators that were put on the Section 311 list.    

Nor should national security experts be afraid of enemy actors using bitcoin to evade sanctions, I wrote in 2019, since existing tools, in particular secondary sanctions, are capable of dealing with the threat. The failure of bitcoin to serve as an effective tool for funding the illegal Ottawa protests, which I documented in a March 2022 article, only underlined its low threat potential:

"Governments, whether they be democracies or dictatorships, are often fearful of crypto's censorship-resistance, leading to calls for bans. The lesson from the Ottawa trucker convoy and Russian ransomware gangs is that as long as the on-ramping and off-ramping process are regulated, these fears are overblown." [link]

Other calls to ban bitcoin were inspired by its voracious energy usage. In a 2021 blog post entitled The overconsumption theory of bitcoin, I attributed bitcoin's terrible energy footprint to market failure: end users of bitcoin don't directly pay for the huge amounts of electricity required to power their bets, so they overuse it. No need to ban bitcoin, though. The way to fix this particular market failure is to introduce a Pigouvian tax on buying and/or holding bitcoins, which I described more clearly in a 2021 blog post entitled A tax on proof of work and a 2022 article called Make bitcoin cheap again for cypherpunks! 

Lastly, whereas bitcoin's harshest critics have been advocating a "let it burn" policy approach to bitcoin and crypto more generally, which involved leaving gateways unregulated and thus toxic, I began to recommend regulating crypto exchanges under the same standards as equities exchanges in a 2021 article entitled Gary Gensler, You Should be Watching How Canada is Regulating Coinbase. Yes, regulation legitimizes a culture of gambling. But even Las Vegas has stringent regulations. A set of basic protections would reduce the odds of the betting public being hurt by fraudulent exchanges. FTX was a good test case. After the exchange collapsed, almost all FTX customers were stuck in limbo for years, but FTX Japan customers walked out unscathed thanks to Japan's regulatory framework, which I wrote about in a 2022 post Six reasons why FTX Japan survived while the rest of FTX burned.  

So when does bitcoin get dangerous?

What I've learnt after many years of writing about bitcoin is that it's a relatively innocuous phenomena, even pedestrian. When it does lead to bad outcomes, I've outlined how those can be handled with our existing tools. But here's what does have me worried. 

If you want to buy some bitcoins, go right ahead. We can even help by regulating the trading venues to make it safe. But don't force others to play.

Whoops, You Just Got Bitcoin’d! by Daniel Krawisz

Alas, that seems to be where we are headed. There is a growing effort to arm-twist the rest of society into joining in by having governments acquire bitcoins, in the U.S.'s case a Strategic Bitcoin Reserve. The U.S. government has never entered the World Series of Poker. Nor has it gone to Vegas to bet billions to tax payer funds on roulette or built a strategic Powerball ticket reserve, but it appears to be genuinely entertaining the idea of rolling the dice on Bitcoin.  

Bitcoin is an incredibly infectious early-bird game, one that after sixteen years continues to find a constant stream of new recruits. How contagious? I originally estimated in a 2022 post, Three potential paths for the price of bitcoin, that adoption wouldn't rise above 10%-15% of the global population, but I may have been underestimating its transmissibility. My worry is that calls for government support will only accelerate as more voters, government officials, and bureaucrats catch the orange coin mind virus and act on it. It begins with a small strategic reserve of a few billion dollars. It ends with the Department of Bitcoin Price Appreciation being allocated 50% of yearly tax revenues to make the number go up, to the detriment of infrastructure like roads, hospitals, and law enforcement. At that point we've entered a dystopia in which society rapidly deteriorates because we've all become obsessed on a bet.

Although I never wanted to ban Bitcoin, I can't help but wonder whether a prohibition wouldn't have been the better policy back in 2013 or 2014 given the new bitcoin-by-force path that advocates are pushing it towards. But it's probably too late for that; the coin is already out of the bag. All I can hope is that my long history of writing on the topic might persuade a few readers that forcing others to play the game you love is not fair game.