Up to Two-Thirds of Bitcoin Transactions Have No Economic Value
bloomberg.com
bloomberg.com
If you are going to slice it like that, there are plenty of ACH/SWIFT transfers that are actually just people moving money from account to account or institution to institution too, rather than actual inter-party transactions.
And specifically, if mining pools distributing rewards to participants don't have economic value, then neither does Uber. Mining pools are literally just "the gig economy" for computational power.
I'm as critical of Bitcoin as anyone else but this is just an ad for Elementus's services. "If you want to understand what you're getting into, pay us!".
In that sense mining pool distributions are double-counting - the distribution of the mining reward to the pool and then the distribution of that same reward from the pool to an individual member. (I understand that correctly right?)
A mixer is just shuffling it's coins around so it's non-economic. Spoofed trades are obviously non-economic. 'Market Manipulation' could mean lots of stuff but I wouldn't be surprised if there are bots that enter lots of offsetting trades on both sides of the orderbook to generate volume around specific events. If they're offsetting that's non-economic.
Now maybe 'non-economic' isn't the best term but I do think those sorts of activities should be given a meaningful distinction from the other third of the volume that is 'normal transactions' of individuals buying and selling from/to each other.
In a model where Bitcoin is the only available currency, you could certainly assign an economic value to a mixer - the value of the created privacy (yes, I know that mixers are flawed).
There are also cryptocurrencies that have (a higher level of) built-in privacy, like ZCash, and that people are using it should also be an indicator that there is economic value in those transactions since the only real reason to use ZCash is to get that additional bit of privacy.
Are mixers spoofing?
Here, consider a market with three people: A, B, C all with 100 BTC.
At 12:01 A sells 50 shares to B. Starting at 12:01 and going until 12:00 B and C sell 1 share back and forth, one transaction per minute. So after 24 hours they have:
A - 50
B - 150
C - 100
Is the 'true' trading volume for the day 50 or is it 1490? It's a matter of perspective but Elementus views that 50 as an important number that investors should have. I agree that using the term 'economic' is dumb but investors want to understand the actual dynamics of buyers and sellers in the market, not the dynamics of ultra-noisy zero-netting mixers.
Let's say I have a customer who pays me, and I pay a bunch of employees or contractors who helped me deliver services to that customer, keeping some profit for myself. Does that have no economic value? Because that's what mining pools do.
> I wouldn't be surprised if there are bots that enter lots of offsetting trades on both sides of the orderbook to generate volume around specific events.
There is no order book on Bitcoin, Ethereum, and others. The order books are on web based exchanges, and yeah, there's a lot of fake trading. Most of it is clearly generated by the exchanges themselves and is completely fake. That stuff is so obvious when you look. In those cases the volume dwarfs the depth, so most orders miraculously attract counter parties instantaneously while the rest of the orders just sit there. Exchanges fake it to get a better rank on coinmarketcap, which blindly trusts the trade data from the APIs. And there's wash trades by users of these web based exchanges, but I believe it's significantly less volume than the pure fakes because those guys have to put up real funds so it's only useful to make shitcoins look more popular. In any case, none of that actually makes a transaction on a blockchain. It's just shifting balances in some centralized database.
Again we're really getting hung up on the term 'economic.' Okay you're the pool, you receive 50 BTC and you pay 45 of that out to miners. Have we transacted 95 BTC or 50 BTC? Is the 'economic value' of our transaction 50 or 95? The best thing is it doesn't even matter what you answer because what's best for everyone is if we know both numbers. Right now all parties interested in BTC can see is the 95. But if I'm investing in BTC I don't care about the 95, I care about the 50 because 50 tells me how much BTC actually changed hands here. And this firm is just trying to find all the double counting and completely fake transactions and remove them from our number.
>There is no order book on Bitcoin, Ethereum, and others. The order books are on web based exchanges
I mean yeah? I don't understand what difference you're trying to draw here.
>yeah, there's a lot of fake trading
>It's just shifting balances in some centralized database.
Exactly and it's the same way in a mining pool as per the above example. 45 BTC is getting counted twice because we shifted it twice.
How are they accurately identifying the purpose of Bitcoin transactions? I cannot think on any methodology that would not be wildly speculative.
Elementus seems like perhaps they don't understand what they are getting into either.
Is this two-thirds of the amount transacted? Two-thirds of total transaction count? Two-thirds of bytes written to the blockchain?
I assume this doesn't involve economic activity transacted in bitcoin which isn't recorded on the blockchain; for instance payments made between users on coinbase or trades on cryptocurrency exchanges.
Thats a great analogy. I'm amused by then thinking about what would happen if 51% of the drivers were able to band together and take control of Uber.
Before the web when kids had to interact you could have a grand game of monopoly playing, kids and neighbours kids involved. There could be high stakes rents being 'earned' and all kinds of virtual activity. Meanwhile the gardening-jobs for pocket money, the car washing for pocket money, the preparing of food for actual food might not be getting done due to the distraction of the all important Monopoly game.
A parent could gently infer that the Monopoly game with side bets for tiddlywinks was not real economic value, thinking that getting the car washed, the garden done and the food prepared was 'economic value' as pocket money, albeit in mere 'fiat' currency was involved.
Wanting to perpetuate the game the most irksome of the children could claim that their Monopoly game was economic value, after all the light and heat is on and some distant coal miner is getting paid for that. They could also claim that there is wear and tear going on and that Parker Games/Waddingtons are profiting from making replacement Monopoly games. They could also claim that washing the car was not 'economic value' as that pocket money is merely transferred inside the household.
Give the article some credit, we have got some rather self absorbed individuals devoting their lives to playing around with digital tiddlywinks, ponzi-kittens and the coin to rule them all - Bitcoin. In the broader sense, after the closing of The Silk Road, evidence is extremely thin on the ground that any of this activity is economically useful.
Is Uber 'economically useful'? I'm as critical as the next person of Uber however, love it or loath it, Uber does get people from A to B and provides a means for people to earn a living. That much is clear and I don't have to ask my inner nine year old child for spurious reasons to argue otherwise.
You're transferring banking systems/locations. That affects money supply, transfer fee's, lending, etc. It definitely has an economic impact, far larger than transferring bitcoins around.
Two brokers walk on the street and suddenly one of them proposes: if you eat that shit on the ground you get a million dollars. He eats it very quickly and is amazingly happy with his profit.
A moment later the guy who had eaten the shit proposes the other: if you eat this shit you get a million from me. He also eats it without thinking. After 10 minutes one of them asks: why have we eaten all that shit?
The other replies: Don't you see we have created two million trading volume?
(Apologies if it isn't funny. Don't remember exactly the original and I'm also not a real joke teller type. )
Ex. for a credit card transaction, the 'economic value' of the transaction (purchase price) would be half or less of the total amount of dollars moving (price from CC to merchant, payments from customer to CC). Even if a customer pays off the balance instantly, two transfers of the amount of the purchase price are made: 50% economic value transactions. If making minimum payments, economic value seems like it could constitute less than a third of total amount transferred.
If CC, merchant, and customer all use different banks, there may be more transfers necessary behind the scenes.
AKA you to CC = X, CC to merchant X - Y. No way to move money like that without 2 simple transactions as 3 accounts are updated.
Even BTC-USD trades have economic value obviously.
The article presents transactions which are basically kinds of accounting functions, i.e. tumblers to hide trails etc. etc..
As far as USD, maybe this would be the equivalent of moving USD from one account to another.
Now, clearly it wouldn't be done if there was no value to the transaction ... but it's generally net neutral stuff.
There's now way to compare to the USD because it's generally not something that is tracked at all with fiat currency because it would be pointless. But it just happens to go on the record with BTC.
Go look up a list of common darknets, go open a few of the top sellers. Look at their listings, look at how many reviews each listing has. Go back 2 weeks later and check again, you'll be surprised. Now look at how many darknets there are with decent user bases and consider how many sellers each has with moderate activity. Most of that commerce is in bitcoin, some in monero or ETH but mostly bitcoin.
I mean, you can pop on and buy several kilograms of cocaine in a single listing (or any other drug you can think of, some even sell guns and human bones, most also have cvv2 dumps and other identity theft stuff as well as stolen uber accounts any all sorts of other things). With the cocaine for example, you can pull up a 1kg+ listing that will have a dozen or more reviews indicating tens, or hundreds, of thousands of dollars in sales just for a single listing on a single market.
People like to make-believe bitcoin is for legit commerce but it's mostly for speculation and criminal enterprise.
>"So about 3 TWh extra electricity is being used each year. This is about the total used each year by Papua New Guinea, which is ranked 133/219 in electricity consumption by country." https://news.ycombinator.com/item?id=17449349
Just measuring single page-loads using a browser extension is far to simplistic of a metric.
All I know is the site has become almost unusable on an old tablet I have, it sits on a loading screen for tens of seconds before loading some simple text content. So whatever they are doing is not very efficient.
I'd love to see better analyses along these lines though. How many TWh per year on pron, how many on netflix/streaming video, how many on drinking/partying. There's all sorts of unnecessary stuff we have no intuition for.
And now I need to go down a rabbit hole to find out the cost / GB currently. I would expect that it has gone down in the last 6 years.
It provides no independently-verifiable evidence of the claim in the title, nor does it even make a small effort to explain the analysis supposedly done by Cryptocompite, a service whose website consists of nothing more than an email capture.
FWIW, determining the nature of Bitcoin transactions is not trivial. The disposable pseudonym privacy model, coupled with CoinJoin transactions and now increasingly Lightning Network gives users a great deal of privacy power if they choose to use it.
Many analyses I've seen are based on transactions whose users throw that privacy stuff out the window, re-using pseudonyms and posting them to social media.
What I would be curious to know is what portion of the remaining third is actually used to buy goods and services vs. speculation. And for those who actually use it as a currency, how much is for buying legal things. My gut tells me that the number must be ridiculously low.
If that change changes your reckoning of the accounting, then the difference between more and less trading volume in the network has come down to how you represent the ledger format. At which point you need to choose how to resolve that discrepancy—for example, by always choosing the most compact ledger format possible.
(Analogy: figuring out how many cycles a series of CPU instructions takes to execute. If some of those instructions are hints or pragmas that the CPU consumes without a full cycle, then representing this CPU’s instructions in a RISC encoding will give you a different cycle-count estimate than encoding the instructions as CISC, where the hints/pragmas just become part of the CISC instruction they’re associated with.)
https://voxeu.org/article/missing-profits-nations
What percentage of US dollar transactions have no economic value?
(And yeah, if my salary counts as an economic transaction then why shouldn't miner payouts? You can avoid double-counting by not counting coinbase transactions.)
One could argue that pure latency arb (what flash boys portray as hft) is fairly useless since it's just trying to catch stale quotes that market makers would move otherwise, but that's a very small portion of the industry.
The question is not how wasteful a system is. The question is, what is the value / waste ratio of it.