My observations on the electricity demand associated with Bitcoin made it into the ABC News Quiz last week, which is a kind of fame, I guess.
Meanwhile, I had another piece in the Guardian, this time looking at the fact that, despite being called a “cryptocurrency”, Bitcoin is used even less as a currency now than it was several years ago. The core problem is that the system is so overloaded by miners creating new coins that processing transactions is slow, costly or both I mentioned the fact that game company Steam had stopped accepting coins and that the list of merchants accepting Bitcoin is small enough to fit on one page. Checking further I concluded that this list is out of date, but not in a good way. Lots of those included, such as Expedia, no longer accept Bitcoin, if indeed they ever did. Here’s one person’s experience. Bitcoin is now a “crypto asset” which is even more obviously a Ponzi fantasy than the original currency story.
One response I got was that transaction speed would soon be greatly improved by something called Lightning. Checking on this it appears that this is software in an alpha (very early) stage of development, which would allow any two parties to set up a transactions account separate from the main Bitcoin blockchain, and only occasionally update the main account. An analogy, for readers of a certain age, is the era before Bankcard, when, if you wanted to do something other than paying cash, you maintained a separate credit and debit account with every store you dealt with. This does not seem like the dawn of a new era to me.
John Quiggin 
I heard Roger Montgomery on the radio.
He could buy bitcoin but could not sell it!! no wonder he is wary of it.
The fact that libertarians are all over the blockchain like a cheap suit should ring big alarm bells. It’s a con built on a paranoid fantasy, just like how the US government is any day now going to take all the guns away from good ole boys in the bayou which pushes American gun sales through the roof.
The obsession with it in venture capital circles is caused (IMO) by the near-complete lack of power in antitrust efforts since the days when Microsoft got neutered, which has led to an era without much in the way of real innovation by grass roots startups.
As more bitcoins are “mined” the blockchain gets longer, slowing the rate of transactions and increasing transaction costs. The trading of shorter, incomplete blocks has been around for a while, but hasn’t fixed the problem. Maybe this “lightning” thing will make it easier, but the problem with blockchains are, as there is no central registry, it is really quite difficult to make changes.
The most commonly thrown around solution for slowing transaction speeds is improving technology will increase processing speeds. Unfortunately, what with all the supercomputer time stolen to mine bitcoins, scientific and technological advancement has probably been significantly retarded.
Hasn’t anyone else seen this:
https://www.theguardian.com/technology/2017/dec/15/japanese-company-paying-employees-bitcoin
Also in the Guardian, published on the same day. Without knowing what is motivating the company it seems to contradict JQs point to the headline scanners. Is it a coincidence both articles appeared the same day or an attempt at “balance” by the guardian?
Is bitcoin mainly a medium of transaction among crooks? For the rest of the economy, it is an intrinsically worthless asset that is costly to produce, just like gold but which operates as a store of value. My guess is that the bubble will subside but the demand for bitcoin will continue.
Cameroon, that sounds suspiciously like company currency, able to buy goods only at the company store. An old dodge used in the past to lure people into a company town with apparently very attractive wages which in fact are not as attractive as they seem.
hc,
An essay on “bitcoin” (re-posted):
https://www.theautomaticearth.com/2017/12/bitcoin-doesnt-exist-the-full-story/
On crooks, the author writes:
“This leads into one of the spurious claims on Bitcoin: that it’s a refuge for drug smugglers and illegal activities. I assure you mathematically, that is not true. According to the U.N. the world drug trade is $435B, 4 times the total, and strictly theoretical value of Bitcoin, coins locked, lost, and all. Besides if you owned $160B coins, who would you transfer them to? You’re the only user. $435B/year can only be trafficked by major banks like as HSBC, who have paid public fines because money flows that large can’t be hidden. This is so well-known the U.N. suggested the drug-money flows may be one reason global banks were solvent in ‘08. Even $160B misrepresents Bitcoin because it had a 10-fold increase this year alone. So imagine $16B total market cap. That’s half the size of the yearly budget of Los Angeles, one city. Even that overstates it, because through most of its life it’s been around $250, so imagine a $4B market cap, the budget of West Virginia.
So you’re a drug dealer in illicit trades and you sell to your customers because all your buyers have Bitcoin accounts? Your pushers have street terminals? This doesn’t make sense. And remember as much as the price of Bitcoin has risen 40-fold, the number of participants has too. Even now, even with Coinbase, even with Dell and Overstock, even with BTC $10,000 almost no one has Bitcoin, even in N.Y.C. or S.F.. So who are these supposed illegal people with illegal activities that couldn’t fit any significant value?
That’s not to say illegal activities don’t happen, but it’s the other half of the spurious argument to say people don’t do illegal acts using cash, personal influence, offshore havens, international banks like Wells Fargo, or lately, Amazon Gift Cards and Tide Detergent. As long as there is crime, mediums of value will be used to pay for it. But comparing Bitcoin with a $16B market cap to the existing banking system which the U.N. openly declares is being supported by the transfer of illicit drug funds is insanity.”
John – I really think you need to learn a bit more about bitcoin before pontificating about it.
Transactions aren’t slow because ‘the system is so overloaded by miners creating new coins that processing transactions is slow, costly’ but because more people want to transact and for a complicated reason the main development team is resisting making each block bigger so that more transactions will fit.
Lightning also doesn’t work like what you are suggesting. Not in the least.
@Joe
While that was inaccurate (miners creating coins is not an obstacle to processing transactions, but how transactions are processed), making the blocks bigger is not much of a solution. O(n/2) is the same as O(n), after all. Increasing blocksize would somewhat decrease transaction cost/lag right now, but yet another increase would be necessary if Bitcoin continues to grow, and another, and another… and right now Bitcoin is extremely marginal in the financial system.
As n grows large, Bitcoin increasingly becomes unfit for purpose as a currency. The various workarounds inevitably consist of taking things off the blockchain, losing the benefits of the blockchain, and are themselves largely unusable as a currency.
Joel – I don’t think anyone who has a bit of an understanding of how bitcoin works would dispute that the block size can’t grow forever. However, until level 2 solutions like lightning and side chains become available a small increase would clear the backlog. It wasn’t so long ago that you could transact for a few cents and blocks could grow much larger quite easily for many years yet.
My main point is that Prof Quiggin obviously knows nothing at all about bitcoin and it shows, badly. His ideological blinkers have blinded him and quite a few people here to its potential use cases. Even his complaint about bitcoins energy usage shows a complete lack of understanding of exactly what we are getting in return.
@Joe
Exactly what are we getting in return for bitcoin and its energy expenditure? You need to spell it out. Otherwise, your claims are completely unsupported.
Speaking more broadly, the bitcoin bubble is just another example of the tendency of unregulated capitalism to create speculative bubbles under certain conditions. The current applicable conditions revolve around secular (long-run) stagnation. We have wage stagnation, over-borrowing for consumption, surplus production (relative to the buying power of wages) and the concomitant capacity under-utilisation.
The bitcoin bubble is another symptom of a known type, the type being the speculative mania or financial bubble. Rather than obsess about particular bubbles, like the bitcoin bubble, we need to look for the systemic cause of all bubbles.
Ikon – to be honest I couldn’t be fagged explaining it to anyone here, except to say that the high energy usage of bitcoin is not some unintended consequence, but there by design.
Joe: “the high energy usage of bitcoin is not some unintended consequence, but there by design”
You’re saying it was intended that for every Australia, it would take a country the size of New Zealand just to power its payment platform?
‘Cause that’s pretty much the scale of the figures we’re seeing.
I’ve had lots of comments like those from Joe, based on self-proclaimed superior expertise. Interestingly, no two of them have ever agreed on what it is that I’ve got wrong.
Notably, when it was still possible to deny Bitcoin’s massive energy use back in 2015, most of the self-proclaimed experts did so, ignoring the basic economics where I was on safe ground. Now, like Joe, they say it’s a feature not a bug, or else duck the issue altogether.
is a typical move in discussions of this kind. Rough translation:
Remember kids! When something soars in price that’s an indication that you should buy more of it! If anyone tells you different it means they don’t want you to get rich and not that they are actually concerned about your well being.
@Ronald
You are a bit early for an April fools day joke.
You get ‘rich’ in whatever unit of account you are measuring the value of marketable things if you buy cheap and sell expensive. Watch kids trading things and you’ll discover many of them understand this.
*Sigh*
The claim that “the system is so overloaded by miners creating new coins” is simply false, that is not the cause of the Bitcoin blockchain’s slowdown. The cause is on-chain transactions simply do not scale well with size, which means simply increasing the block size is simply delaying the problem. The team supporting Bitcoin is well aware of this issue, and have a number of proposed fixes that should alleviate the problems in the short term, which include the adoption of such things as Segregated Witness (SegWit). These fixes take time to roll out, as is the nature of any decentralized and consensus-driven system, and the team has been cautiously conservative in their approach to avoid causing splits in the network. This is exactly as it should be.
I also find it rather curious that you fret about Steam not accepting Bitcoin as evidence of anything, but fail to note that Overstock.com has seen the amount of its Bitcoin sales more than double over the past year, and eBay is reportedly seriously considering adding Bitcoin as an option, and even craigslist has given cryptocurrencies as an option to users. It should be noted that there also services being built that will allow for the payment of cryptocurrencies without requiring merchants to hold onto them.
Your analogy of the Lightning Network is also incorrect. Peers on the Lightning Network can pay any other peer even if they don’t directly have a channel open between each other, and channels can be kept open indefinitely (with no added cost) or closed immediately. Lightning Network also has other benefits and enables such things cross-chain atomic swaps without requiring a trusted intermediary like an exchange. A lot of people mistake Lightning Network for sidechains, but this is incorrect.
Folks honestly interested in learning more about what is being done to alleviate Bitcoin’s scaling issues should consult the Bitcoin capacity increases FAQ (I won’t leave a link here in case this comment gets flagged as spam). But please, don’t sit around implying that somehow the developers don’t know about this problem or aren’t doing anything about it, or that somehow you know more about the technology or the challenges than they do.
@VM Again with the weary technological superiority. It would be great if people making points like this could demonstrate that the energy use problem is an illusion or that Bitcoin is actually making progress as a currency, rather than offering technical quibbles.
Searching on some of the firms you mention, I read the following
That’s consistent with what I’ve written. So, it would be helpful for you to spell out why Bitcoin is facing scaling issues when its use as a currency is decreasing. You say it’s not because mining-related transactions are using up the capacity of the system. What then, is the problem.
With Lightning, I take your point to be that, rather than having a direct channel between parties, it’s possible to pass the transaction through a number of peer-to-peer connections, each of which you must trust to work reliably. That seems even worse than the simple case I describe.
I know SFA about bitcoin but I can see how it might appeal to those who are frustrated by their perceived limits of their own production. As the number of bitcoins is finite some analysts are predicting enormous values, which should be enough keep the whole shebang roling along for quite some time.
@John Quiggin
I’ve attempted to respond to this, but your blog is not allowing my comments through.
John I will explain why I really couldn’t be fagged. Based on what you wrote above it is pretty obvious that despite writing about bitcoin for several years you haven’t really bothered to learn much about it. These aren’t just ‘technical quibbles’, what you said above are frankly ‘whoppers’. Most of your commenters are the same.
So now you want me to explain to you a fairly technical but important aspect about decentralised blockchains and bitcoin in particular? I think not.
By now you are probably thinking I am shilling for bitcoin. Yes, I have some skin in the game but I actually enjoy watching the price go down as well as up, as people try to figure out just what they are dealing with. At this point it is basically just an experiment and it could go anywhere.
I am quite fascinated to see here that most people freely admit they know SFA about bitcoin and are totally fine with remaining that way, but know it can’t be good, or can’t work, can’t scale, is a ponzi, uses to much energy etc. Weird!
@Joe I thought I was the only one admitting to ignorance?
Anyway, IMVHO the problem with bitcoin is that it doesn’t explain itself very well – I don’t fully undestand it so why buy it? (thinking Buffett now).
@VM
I don’t know what would be causing that. There are words that will stop a comment in ways I can’t control. Email me at j.quiggin@uq.edu.au and I will post the comment for you.
@Joe As you say, I’ve been writing about this for a while. So, if you can show that what I wrote in 2015 underestimated Bitcoin’s electricity demand, I’ll be happy to concede error. Otherwise, all you have are (unnamed) silly quibbles.
Interesting and thoughtful analysis. But Bitcoin is still too new for any analysis to be anything but speculative.
I would take anti-bitcoin views much more seriously if those who espouse them were actually short them with their own non-bitcoin currency.
I apologize in advance for the wall of text.
“Again with the weary technological superiority. It would be great if people making points like this could demonstrate that the energy use problem is an illusion or that Bitcoin is actually making progress as a currency, rather than offering technical quibbles.”
I didn’t mention anything related to Bitcoin’s energy use. I agree it is enormous and it is a problem that needs to be addressed, and I agree that it is also a point of vulnerability for the network to have the majority of mining in China. However, this could be addressed by Bitcoin switching to a less energy intensive algorithm, such as Proof-of-Stake, away from Proof-of-Work. Proof-of-Stake is thousands of times more energy efficient than Proof-of-Stake. There is nothing intrinsic in Bitcoin that would make such a switch impossible, provided sufficient consensus is achieved.
It should be noted that cryptocurrencies such as NEO are already using Proof-of-Stake.
Regarding progress towards becoming a proper currency, I’ve mentioned that observers should seek the Bitcoin Scaling FAQ to observe the problem is known and being worked on by Bitcoin’s developers. Unfortunately, the group of developers working on Bitcoin is small and have limited bandwidth, and changing the network requires achieving wider consensus, which has proven contentious in areas such as the block size. This is a non-trivial process.
“Searching on some of the firms you mention, I read the following: “While online retailers like Overstock have accepted bitcoin payments for years, the number of such transactions has reportedly been declining, in part because of the rise in fees.” That’s consistent with what I’ve written. So, it would be helpful for you to spell out why Bitcoin is facing scaling issues when its use as a currency is decreasing. You say it’s not because mining-related transactions are using up the capacity of the system. What then, is the problem.”
The problem isn’t the miners who are generating the coins; the problem is a specific rule in the code that caps the block size to 1 MB. The block is what contains the transactions to be processed, which is what the miners do, and once the processing is done, it is added to the end of the blockchain. Originally, the 1 MB limit was added as a security mechanism to prevent denial of service attacks by malicious parties sending arbitrarily large transactions that could bring the network down. It has had the unintended effect of becoming a hard bottleneck on the network’s ability to process large volumes of incoming transactions.
In periods of high transaction volume, this is what causes the backlog, which in turn causes a spike in fees as people sending transactions choose to pay more for their transactions to get processed quicker.
Now you might say, why not increase the block size? This would require a so-called hard fork of the Bitcoin blockchain, and this is indeed the approach that was taken by the Bitcoin Cash hard fork of Bitcoin. But Bitcoin’s developers have feared that an unnecessary hard fork without sufficient consensus would split the network into two, and have chosen to defer the risk for now and fall back on software updates to increase capacity, such as with SegWit. However, these have been slow to roll-out due to the network requiring consensus and wallets to be updated to support the new functionality. The Bitcoin team is aware that a hard fork will become inevitable, but are playing it cautiously, and not without reason.
The other thing to note is increasing the block size introduces other technical problems, the full scope of which I cannot treat here.
“With Lightning, I take your point to be that, rather than having a direct channel between parties, it’s possible to pass the transaction through a number of peer-to-peer connections, each of which you must trust to work reliably. That seems even worse than the simple case I describe.”
There is no reason, at least of which I can see, that you should consider this as “worse”. Information is reliably routed over secure channels between points over the Internet all the time, and routing is a well understood problem in computer science and software engineering, so routing securely isn’t the challenge. The challenge for the Lightning Network lies in infrastructure adoption. I’m not going to speculate whether ultimately Lightning Network will resolve Bitcoin’s scaling issues; as a software engineer myself, what sometimes works on paper can have serious problems in practise, and even well thought out implementations can unexpectedly fail. However, I believe it has a fair chance of success.
Additional reading for those wanting additional information:
https://en.bitcoin.it/wiki/Block_size_limit_controversy
https://en.bitcoin.it/wiki/Scalability_FAQ
https://bitcoin.org/en/bitcoin-core/capacity-increases-faq
Some interesting comments on the subject http://www.antipope.org/charlie/blog-static/2017/11/unforseen-consequences-and-tha.html
From my memory of John Quiggin’s past comments (on topics not directly related), I suspect his respoonse to you would be along these general lines, borrowing from Keynes: ‘the market can stay wrong longer than I can stay solvent’. In order to make money by going short, you need not only to be right that something is going down, you need to be right about when it is going down, which is much harder.
@J-D Exactly right. True in the dotcom bubble and true today. If i’d gone short in 2015, where would I be now? IIRC, the Keynes quote is apocryphal, though also very apposite.
@27 I’m aware of the block size issue and of Bitcoin Cash. But, as I understand it, Bitcoin Cash doesn’t seem to have fixed the problem. More to the point, why is block size a problem now more than in the past?
As I understand it, it’s because transactions volumes are increasing, with the result that payments for priority processing are rising. Since we know goods and services transactions aren’t increasing, the increased volume must be on the asset side (mining or speculation).
To sum up it seems to me that our disagreement rests on the fact that I’m looking at the growing demand for (non-currency related) transactions while you’re looking at the fixed supply, constrained by blocksize.
Happy to be corrected on this.
And here’s a Wired article on the energy issue with various claims -https://www.wired.com/story/bitcoin-global-warming/?mbid=synd_digg
@John Quiggin, apologies for the late response. To respond to your points, no, Bitcoin Cash does not solve the scaling problem. Upping the block size (and ignoring the other technical implications of it) is only a temporary fix. To quote the Bitcoin Scaling FAQ: “For basic Bitcoin to scale to just 2 transactions a day for 7 billion people would require 24-gigabyte blocks”.
Regarding what the rising volume of Bitcoin transactions are about, I agree that a large chunk of it is being driven by speculation and hype. People are buying, selling, and trading their coins in the hope they will get rich quick. There is no doubt a big degree of greed and stupidity in action here, but I think it’s important to also understand that Bitcoin’s developers don’t and can’t control the media representations and hype that outsiders and media personalities may build.
However, speculative trading aside, I do believe there is at least some demand by people to use their cryptocurrency as a means for paying for goods and services. If you check the statistics for Living Room of Satoshi, you can see people are using their cryptocurrency to — at least indirectly — pay their bills and credit cards. Perhaps with time, this demand will continue to grow, or it may vanish.
Also, to correct a rather humorous typo in my previous post, the sentence that reads, “Proof-of-Stake is thousands of times more energy efficient than Proof-of-Stake”, should be, “Proof-of-Stake is thousands of times more energy efficient than Proof-of-Work”.
VM: “There is nothing intrinsic in Bitcoin that would make such a switch impossible, provided sufficient consensus is achieved.”
The chances of Bitcoin ever achieving “consensus” to completely overhaul the entire Core architecture to become Proof of Stake are zero to zip.
You may as well be claiming that “war is bad, but peace is hypothetically possible, so war isn’t all that bad”.
Or – “continued use of coal is justifiable because hypothetically it might become clean one day”.
And NEO doesn’t use Proof of Stake, it uses modified Practical Byzantine Fault Tolerance.
@Nick
You’ll pardon me if I am unconvinced by your speculative assertion. Everyone on Bitcoin Core knows a hard fork involving major changes will eventually be needed, and this could entail adopting a different algorithm, whether PoS or some other algorithm which could be developed in the future.
And you are correct about NEO. I simply did a quick and lazy 30-second Google search for Proof-of-Stake coins. It’s an insubstantial detail.
As a (retired) IT Professional, I’d be shorting the bejesus out of Bitcoin if I were a gambling man. The main reason I don’t is because I’m not quite sure when the bubble will crash.
@VM “I do believe there is at least some demand by people to use their cryptocurrency as a means for paying for goods and services”
Wouldn’t this be nuts in the current environment, with the price of Bitcoin going parabolic? Why would you spend something that might be worth double in a couple of weeks? (unless it’s for nefarious purposes, I guess)
It would seem to have potential for either transactions or speculation, but you can’t use it for both at the same time. At the moment, it’s being used for speculation (and there’s no interest or dividend returned based on its output, so it *is* pure speculation), and so is essentially useless as a currency, and (to me at least) will remain so until I can pay my taxes or buy bananas at the local shops with it.
@Norfolk Enchants, I agree that combination of growing value and high fees would give a strong incentive for people to hold onto Bitcoin rather than spend it, but a look at Living Room of Satoshi’s Stats page states that (as of the time of this writing) Bitcoin accounts for about 56% of use of the platform. Evidently, some folks are choosing to spend their Bitcoins on paying off bills, credit cards, and bank accounts.
Perhaps these folks care less for the high transaction fees, or perhaps they’re timing their spending during periods of lower network activity, or perhaps there is some other reason or combination of reasons. I don’t have an answer.
Other than holding onto one Bitcoin for sentimental reasons, I don’t have a compunction against spending it provided the fees are not excessive. I think the arguments that it should be a “store of value” are silly, cringeworthy, and are probably harmful for Bitcoin’s acceptance. I think it is better for the Bitcoin community to just be forthright and say, “Sorry, we have scaling problems”, than to say, “Don’t spend it, it’s a store of value”.
Everyone on Bitcoin Core knows a hard fork involving major changes will eventually be needed
Which means that current bitcoin is, well, useless, no?
It’s the esperanto problem: esperanto is plainly unfit for the purpose of being an international auxillary language… but fixing esperanto means persuading people who’ve learnt current — defective — esperanto. Who aren’t by-and-large people who are concerned by esperanto’s problems, because if you were you’d never have learned the bloody language.
So there’s no realistic path forward within esperanto. And no point investing in esperanto, because it’s not going to be the basis of any auxlang. But because esperanto has the bulk of the auxlang buy-in… no realistic prospect of progress outside of esperanto, either. Net result, auxlangs Just Ain’t Gonna Happen.