Read on Twitter
Bitcoin User Adoption Thread
How many Bitcoin users are there?
This is an important question because the number of users has a direct impact on the value of the network. But there are no good ways to directly measure the number of users.
1/19
How many Bitcoin users are there?
This is an important question because the number of users has a direct impact on the value of the network. But there are no good ways to directly measure the number of users.
1/19
Some discussions, like this one from @FEhrsam garner responses of disbelief or confusion.
https://twitter.com/FEhrsam/status/1355582084048433162?s=20
This has a lot to do with the way most Bitcoin user estimates are created: from limited user surveys.
2/19
https://twitter.com/FEhrsam/status/1355582084048433162?s=20
This has a lot to do with the way most Bitcoin user estimates are created: from limited user surveys.
2/19
1. The good surveys carefully choose their population sample, and, as a result, wind up being narrow measures, e.g., from one country.
2. The poor surveys seem to be less careful, blasting their surveys widely, but then suffering from a bias in the sample of survey takers..
3/19
2. The poor surveys seem to be less careful, blasting their surveys widely, but then suffering from a bias in the sample of survey takers..
3/19
Want to reach as many people as possible? Send the survey out on the internet! But then who is going to take it... Heavy internet users who are comfortable or savvy enough to take an internet survey, i.e., those who would be more likely to be early adopters of Bitcoin.
4/19
4/19
Because of these trade-offs (and because Bitcoin is growing quickly), there are wide-ranging figures for user adoption.
For example, you can find estimates for the adoption of Bitcoin in the US ranging from 5% to 14.4% ( https://libertystreeteconomics.newyorkfed.org/2019/03/deciphering-americans-views-on-cryptocurrencies.html and https://www.finder.com/how-many-people-own-cryptocurrency)
5/19
For example, you can find estimates for the adoption of Bitcoin in the US ranging from 5% to 14.4% ( https://libertystreeteconomics.newyorkfed.org/2019/03/deciphering-americans-views-on-cryptocurrencies.html and https://www.finder.com/how-many-people-own-cryptocurrency)
5/19
Some of the most rigorous of these surveys seem to be those from the UK FCA (3.9% of UK adults in 2019, https://www.fca.org.uk/publication/research/research-note-cryptoasset-consumer-research-2020.pdf), the Bank of Canada (3% of Canadian adults in 2018, https://www.fca.org.uk/publication/research/research-note-cryptoasset-consumer-research-2020.pdf), and the NY Fed (5% of US adults in 2018; see the last tweet)
6/19
6/19
But the best survey/report BY FAR is from the University of Cambridge.
For three prior years (2020, 2018, and 2016), they have produced a global crypto survey that attempts to measure user adoption, among many other statistics and useful trends.
https://www.jbs.cam.ac.uk/faculty-research/centres/alternative-finance/publications/3rd-global-cryptoasset-benchmarking-study/
7/19
For three prior years (2020, 2018, and 2016), they have produced a global crypto survey that attempts to measure user adoption, among many other statistics and useful trends.
https://www.jbs.cam.ac.uk/faculty-research/centres/alternative-finance/publications/3rd-global-cryptoasset-benchmarking-study/
7/19
The Cambridge team gets around the pesky user sampling issue with a clever workaround. Since almost everyone who uses crypto uses a service provider (a crypto exchange, a wallet provider, etc.), the team surveyed those middle-men instead of surveying users.
8/19
8/19
They tried to get a comprehensive survey from all crypto service providers (but that's tough).
To estimate the number of crypto users, they asked each service providers for how many accounts they had open. This is clever, although it has some shortcomings.
9/19
To estimate the number of crypto users, they asked each service providers for how many accounts they had open. This is clever, although it has some shortcomings.
9/19
On the one hand, this would underestimate the no. of ACCOUNTS in the crypto space, because the team could not include all service providers.
On the other hand, it would overestimate USERS because each user tends to have more than 1 account at different servicer providers.
10/19
On the other hand, it would overestimate USERS because each user tends to have more than 1 account at different servicer providers.
10/19
But the team also did something else. They asked each service provider "For your accounts, on how many did you perform KYC or AML checks?" And the answer came back at a little more than 50% for the 2020 survey.
They called these checked accounts "ID-verified" users.
11/19
They called these checked accounts "ID-verified" users.
11/19
For 2020, the team found 191mn accounts across the service providers they surveyed (and they used public-data to estimate account totals for some of those that didn't participate). And they found 101mn ID-verified users.
12/19
12/19
Then, the final piece. The team estimated that this ID-verified-to-total-account measure did a decent job of proxying for the fact that most users have more than one account.
But there was still the issue of not capturing some non-participating service providers.
13/19
But there was still the issue of not capturing some non-participating service providers.
13/19
So, the team estimated that 101mn was a lower-bound estimate but also "offers a reliable approximate figure of the total number of cryptoasset holders globally."
14/19
14/19
Today, there are approximately 7.8bn people on Earth, of whom 5.8bn are adults, and 4.66bn are internet users.
If we take 101mn as a reliable estimate of Bitcoin users, that means it has been adopted by 1.7% of the world's adults and 2.2% of internet users, as of Q3 2020.
15/19
If we take 101mn as a reliable estimate of Bitcoin users, that means it has been adopted by 1.7% of the world's adults and 2.2% of internet users, as of Q3 2020.
15/19
But wait! There's more!
The Cambridge team also has an estimate using the same methodology from Q3 2018.
At that point, they estimate there was a lower-bound of 35mn crypto adopters around the world.
16/19
The Cambridge team also has an estimate using the same methodology from Q3 2018.
At that point, they estimate there was a lower-bound of 35mn crypto adopters around the world.
16/19
Figures for the growth of Bitcoin users over time are especially valuable because they allow us to:
1. Compare Bitcoin adoption in the past to the value of the network, and...
17/19
1. Compare Bitcoin adoption in the past to the value of the network, and...
17/19
2. Make model estimates for how users might grow into the future and what that might mean for the value of the Bitcoin network.
More to come on that in a future thread.
18/19
More to come on that in a future thread.
18/19
Thank you to many people whose work inspired this thread, including @FEhrsam, @woonomic, @nic__carter, @ApollineBlandin, @ProfPieters, and @nsquaredcrypto among many others.
19/19
19/19
