Is Bitcoin Really Used 606 Times Less than USD To Launder Money?
Is Bitcoin Really Used 606 Times Less than USD To Launder Money?
The Base Rate Fallacy Strikes Again
Let’s say your 4th grader comes home and asks you for help with the following question:
Which one is bigger: ½ or 19/40?
If your child says 19 > 1 therefore 19/40 > 1/2, you would look at them funny. One must first find the common denominator, and then compare the numerators to figure out how the fractions compare.
Yet when Joe Kernen, longtime CNBC anchor and co-host of Squawk Box, makes what amounts to the same mistake, very few people point it out. Certainly, the SEC Chair Gensler had a chance to correct the record, but he didn’t.
The context in which Joe Kernen committed this error involved money laundering. The question is rather simple: Is there more or less money laundering with Bitcoin when compared to USD?
Joe Kernen appears not to have done much more than form an opinion after reading a tweet without fact-checking it. As far as we can tell, the $20 trillion figure referenced in the tweet (supposedly the amount of money laundered via USD) is not even real. That an unverified statistic can make it to CNBC in a matter where the stakes are very high is mind-boggling, to say the least. However, even if the figure were to be correct, this comparison would be as meaningless as concluding that 19/40 is greater than ½ because 19 is greater than 1 in the above example.
How much money laundering is actually occurring with USD vs. Bitcoin? Looking at the image below, this is how much, according to Kernen (the one on the left is USD and the one on the right is Bitcoin):
At a minimum, this is the second time Kernen compared USD vs. Bitcoin in the money laundering context. He also talked about it in this video back in January. That day, Kernen didn’t say that the USD is used 606 times more than Bitcoin to launder money, but he did cite 20,000 for USD and 33 for Bitcoin, the very numbers that came from the tweet linked above. Divide the former by the latter, and voila! You get 606. So, that is Kernen’s reference to “a multiple”; without naming it, he cited the figures that lead to the ratio 606. Clearly, that is the position of the CNBC anchor.
This is, of course, yet another example of the good old base rate fallacy. Many people succumbed to it in the context of COVID-19 hospitalizations and vaccinations. Even the brightest minds are not immune to it; six law professors from prominent schools submitted an amicus brief in SEC v. Coinbase in which they conflated absence of evidence with evidence of absence and drew false inferences about the role of contracts in “investment contracts.” Here, the base rate fallacy strikes again, and this time it’s Joe Kernen who draws a false inference about the role Bitcoin plays in money laundering because he does not seem to appreciate that the “study” he cites is deeply flawed.
How would one visualize the base rate fallacy? Glad you asked! What you see below on the left is how Joe Kernen sees the world:
We are not the first ones to point out this flawed comparison. Alison Jimenez, a Bates consulting and testifying expert, has touched on it in a recent congressional testimony. John Reed Stark has an X post on it. The problem with this testimony, in our opinion, was not that Jimenez’s points are not valid but that they lack a certain focus. The real issue is this one, a bullet point that is buried under four others:
Global GDP and “all cryptocurrency transaction volume” are not interchangeable and do not measure the same thing. Global “transaction volume” is orders of magnitude greater than Global GDP.
That’s the ballgame. Jimenez correctly highlights the fact that by using the transaction volume for Bitcoin and GDP for USD, respectively, in the denominator (see, for example, this piece) the comparison stopped being apples-to-apples. The issue is finding the common denominator, and making sure one is ready to compare ½ and 19/20. By providing a laundry list of items, Jimenez prevented the real money laundering issue from becoming front and center in this discussion.
In our visualization of the base rate fallacy above, it is clear the left-hand side is the wrong picture. But what is the correct picture that one should place on the right? To be clear, what you see on the right above is not yet an estimate of the role of Bitcoin vs. USD in money laundering. It’s a placeholder image that is simply intended to highlight that the comparison making the rounds is deeply flawed and resembles what you see on the left. We still need to figure out how to make the apples-to-apples comparison with our goal being to estimate the respective frequencies of Bitcoin vs USD in money laundering. We will share our analysis with you in our next post.
