Cryptocurrency

[Mashy]

I want to discuss crypto. Who's holding? Portfolio? Are you bullish/bearish?

[Utisz]

Personally, I can't wait for the day when the 0.5 BTC I turned down from Perdix exceeds my own net worth.

[starla]

Not holding. Am generally suspicious. I think it is going to trade like a tech stock on steroids: when the market is good it will rocket. When the market turns down it will plummet fast. I think it's high right now partly because chips are so scarce.

I am usually wrong about these things though.

I don't understand fully how bitcoin is mined, and why running algorithms on a computer should translate into money. Who is getting value out of whatever the miners are doing? This is what has prevented me from ever buying any, recent runup not withstanding. And I think the lack of understanding on the part of most non-SWE folk is what prevents wider appeal and adoption.

[DJ Drug Problem]

It's a hedge for millennials as gold is to boomers, the more uncertain the markets the more it goes up, could be worth a house in a decade or two.

Good luck cashing an entire coin out at an all time high though. There will be a line to cash out and coinbase doesn't have enough to pay everybody, they'll do a robinhood or worse just fold. You're better off selling 1% at a time on the way up once it gets "big enough" whatever that may be.

source: successfully traded 1% of a bitcoin. Currently not holding any or buying anymore until it takes a GIANT shit, whenever that may be.

Is it going to go up or down? Yes. Both.

[Utisz]

I don't understand fully how bitcoin is mined, and why running algorithms on a computer should translate into money. Who is getting value out of whatever the miners are doing? This is what has prevented me from ever buying any, recent runup not withstanding. And I think the lack of understanding on the part of most non-SWE folk is what prevents wider appeal and adoption.

That is the biggest problem I see with bitcoin specifically: that the work has no inherent value and is bad for the environment.

Not sure if you are looking for an explanation of cryptocurrency, but here's my unsolicited long-winded version from first principles (thought I'd manage to keep it short, but I was off by a smidge).

Back in the old days, we had commodity money, like gold or silver coins. Its value were based on the inherent value of the material is was made of. But then people started to mess around by debasing the currency, reducing the amount of gold and silver, maybe by folks taking tiny chips off of many coins, or by the government defining the standard gold coin to have less gold in it so they would have more coins to pay people with (causing inflation, as prices go up to adjust). Also it was presumably a pain in the ass to store or pay large sums of gold or silver.

So rather than deal with physical commodities, representative money came along, which is paper money (bank notes) backed by a fixed amount of a commodity in a way that the amount could not be reduced, and the money could not be debased (though of course the value of the underlying commodity might fluctuate, its quantity is fixed). Basically you could now trade money that entitled you to withdraw a certain amount of a commodity (backed by a government). This was the money used in the U.S. until 1971, where each U.S. dollar was backed by 1/35 of a troy ounce of gold (mostly held in Fort Knox).

Eventually, representative money gained sufficient traction that governments could do away with the need for some Fort Knox-style repository of gold to back the currency with, and just continue with the paper (dollars) itself. Superficially, little changed for the folk on the street; they go about business as usual paying with dollars, just that now, the money is not pegged to gold. The US dollar is now fiat money: a currency that has little or no inherent value (like the paper in a 100 dollar bill has no value), and that cannot be exchanged in a guaranteed way for a fixed amount of a commodity with inherent value (i.e., is not tied to the value of such a commodity), but is still backed and controlled by a government. The main change is that fiat money allows the government to generate as much money as they like (which again can lead to inflation) without having to stockpile more gold in Fort Knox.

Enter cryptocurrency: a new type of money that, like fiat money, has almost zero inherent value, but, unlike fiat money, it is not backed by a government or political union. But it has to follow some of the same principles as the other types of money (commodity, representative, fiat). Namely, money needs to be based on something scarce: it would be impractical to have money based on grains of sand as they are so common that only massive amounts could have value; durable: it would be impractical to have money based on some fruit or some radioactive stuff that will decay quickly as you could not store wealth; verifiable: you need, for example, to be able to verify that a dollar note is real, and to ensure it cannot be easily faked; portable: you need to be able to move it around; etc.

Actually, cryptocurrency easily fits most requirements for money (e.g., it's portable over the internet) except scarcity, and convincing people it's actually money without a history of commodity/representative money to instill faith, and/or a government telling them it's worth something.

For scarcity, cryptocurrency works off a similar principle to how the passwords work online: based on ideas from cryptography, and in turn on number theory. An analogy is that we can create digital locks and digital keys (just numbers basically) so that only a tiny fraction of keys will open a given lock. So it is very hard (in some cases impractical) to find the key for a given lock unless you already know it. The only option you have otherwise is to try lots and lots of keys until you find one that opens the lock.

Bitcoin was designed so as to view each new bitcoin (or some fixed amount of new bitcoins) as a lock that only a certain small fraction of keys can open. The process of mining is then to computationally try lots of different keys until you find one that fits and opens the lock. The first key that is found (with some minor exceptions) then becomes the new bitcoin. (In this case, the lock more precisely encodes a history or block of transactions, so once a key is found, it becomes part of the blockchain -- the official log of transactions -- and the miner who finds the key gets paid a certain number of bitcoins -- currently 12.5BTC -- for completing the block with its key.) You can then pass the key around to other people in the network to verify that it is the right one: it is easy for anyone to try the key in the lock and see that it works. All bitcoin have been generated in this way, based on finding a key in a haystack.

Why was Bitcoin designed this way? It creates artifical scarity based on the fact that it is hard to find a key and thus to create new bitcoin. Based on current values, we can say that it costs at least $50k on average -- in terms of the electricity and computational machinery -- needed to find a key first. The word "first" here is important. If you find the key second, it is (in most cases) worthless. If you could find a key first, on average, for less than $50k, then more folks will join the mining process as they pay less than $50k to gain $50k, which in turn makes it harder, and more expensive, to find a key first. So the mining process strikes a balance regarding market value. More value, more miners; less value, less miners.

But we don't want to tie value to the costs of mining. If bitcoin has a high value, more miners will come, but that doesn't mean that we necessarily want more bitcoin to be generated, as that would reduce scarcity and the value of bitcoin. On the other hand, it is important that mining continue as it is the process by which transactions are recorded. There are thus some valves that can be used to offset the costs/value of mining, like how much new bitcoin is paid out when someone finds a key to unlock a block of transactions; this value is constantly going down, and will eventually go to zero, at which point no new bitcoins will be created, and supply will be fixed at 21 million bitcoin in circulation. These values were defined at the outset of bitcoin and are not controlled by any government or whatever. Another factor is the number of valid keys that will unlock new bitcoins, which is adjusted based on how long it took all the miners to unlock the last block so as to keep the average time to around 10 minutes. These measures together mean that the market value determines the cost of mining, not the other way around; scarcity finds a balance to match the value of the market. (Also the difficulty is important to avoid there being lots of different competing blocks emerging at the same time, which would lead to competing forks of the blockchain, something that happens rarely given the current difficulty adjusted to generate a new block every 10 minutes.)

Just because something is scarce though, doesn't mean it's worth something. None of this is worth anything unless folks actually believe it is worth something. This is the intangible part and the volatile part. The actual mechanics of Bitcoin are well understood, transparent, and fairly predictable. What is not predictable is the market value in terms of how accepted Bitcoin will become as a currency you can buy things with. Bitcoin is believed to satisfy all the necessary criteria for money, but it's not money unless people believe it is.

Why might people believe it has value? Well, we believe the US dollar (or whatever traditional currency) has value, fundamentally, because not only is it useful for us to believe that, we have a vested interested in believing that. It is useful because it is better than having to cart around beaver hides everywhere we go to trade for something. We all have a vested interested in upholding that belief because our bank accounts use US dollars as their measure, and if we lose that belief, we (at least if we have a positive amount) lose our wealth. (Here, "we" is not an equal reference; those with more dollars have more of a vested interest and tend to wield more political power in terms of controlling regulation, monetary policy, etc.)

So is Bitcoin useful to believe in? Well it is not backed by any government, and can theoretically be used for anonymous payments across political borders, without interference or control from financial institutions (with a big caveat: this is subject to the laws and regulations of particular governments). Also we buy more and more things online, so the infrastructure and culture is already in place for bitcoin to prosper. So people see this as a useful feature that means that it will be useful for us to believe in Bitcoin as money. Aside from that, the whole idea of cryptocurrency has its roots in crypto-anarchism, meaning being able to operate individually beyond the government sphere of influence using cryptography to make oneself anonymous and beyond prosecution (so long as the system itself, e.g., Bitcoin, is allowed to continue). In terms of vested interest, this is why Bitcoin sees peaks when "established players" jump on board. As more and more powerful players come on board, vested interests increase, and so too does political power and influence in terms of potential regulation, as well as the credibility of the cryptocurrency in terms of actually being money.


Now the key problem with Bitcoin is the ecological effect, which may be unnecessary. In general, to mine a new block/set of bitcoins, quintillions of keys have to be tried (in the space of 10 minutes), which requires massive computational resources that are constantly operating, using up energy, and creating pollution. As bitcoin's market value goes up, the number of miners goes up, and the difficulty in terms of the number of keys that have to be tried also goes up, which worsens the effect. This process creates the (artificial) scarcity that Bitcoin depends on, but is a massive waste of computational power. It has increased the prices of GPUs, increased the prices of electricity, has a negative effect on the environment, and is potentially unnecessary. It would be like saying, hey, let's use orphan blood as the ink for our new US dollar bills.

So if there were a way to create a more ecologically-friendly cryptocurrency, with some other form of artificial scarcity that doesn't waste massive amounts of computational power, but doesn't require some political power to prop it up, that would be the next step. I know there have been proposals along those lines, but not sure if any have taken off.

[djm]

I pretty much agree with everything Starla wrote.

I don't understand enough about it, and the more I learn the less I trust it.

[starla]

That was helpful, Utisz. I understood most of it already except for what exactly bitcoin mining is. I knew bitcoin mining required computers and energy and so it tracked with energy and chip prices.

So I'm still thinking of it as a tech stock on steroids. Except that as chips get scarcer BTC goes up, whereas chip prices are more likely to be a drag on most tech stocks. And I think BTC has higher regulatory risk.

[Utisz]

That was helpful, Utisz. I understood most of it already except for what exactly bitcoin mining is.

There are some other computational details about mining, but I don't think they are important to understanding how cryptocurrencies work (rather they are important to understand how Bitcoin was implemented).

I knew bitcoin mining required computers and energy and so it tracked with energy and chip prices.

Yep, the overall system is not so intuitive, but it is kind of elegant, except for the wasted energy (you could power Argentina off the power wasted by Bitcoin, according to the BBC).

Maybe one correction to what you say: the value of bitcoin should not, in general, track computational costs and energy all that closely. Cheap power (and computers, etc.) doesn't mean more bitcoins are produced, so cheap power does not devalue bitcoin.

This is because when the costs of mining go down, and more miners come online, the system adjusts to makes it artificially harder to mine bitcoins, and vice versa. The difficulty and thus resources required to mine are adjusted so that there's an average of a 10 minute delay between blocks. So all that should be in balance. This all means that neither the supply nor market value of bitcoin depend much on computer or energy costs, nor on the number of miners in general. Supply of bitcoin has been defined transparently since its inception: one can predict, more or less, how much there will be at any point, and that it will max out at 21 million BTC.

This is unintuitive since if we think of gold, for example, cheaper or better methods to mine gold will lead to there being a greater supply of gold and thus cheaper gold. In the case of bitcoin, it is almost like when cheaper or better methods are achieved to mine bitcoin, the bitcoins automatically burrow themselves deeper into the ground to make it as hard as before to mine them, i.e., so that one kilo of bitcoin is able to be extracted from the ground every ten minutes no matter how shit or shit-hot the mining equipment is or how cheap/expensive it is to run. So we can say with certainty that after about 100 minutes, there will be about 10 kilos of bitcoin produced, and so on, independently of the efficiency of the mining equipment.

So it's better to think of the market value of bitcoin and the costs of mining (electricity/computers) as controlling the amount of miners participating, but not as the amount of miners participating or the costs of mining controlling the market value of bitcoin.

The value mostly comes from whatever the market decides it is worth, which in turn is tied to what it can be used to buy either now, or in the future.

So I'm still thinking of it as a tech stock on steroids. Except that as chips get scarcer BTC goes up, whereas chip prices are more likely to be a drag on most tech stocks. And I think BTC has higher regulatory risk.

Yep, the regulatory issues are still grey, and could pose an existential threat if policy makers do not have a vested interest in seeing bitcoin or cryptocurrency flourish. This is also an issue if they have a vested interest against bitcoin or cryptocurrency, an obvious one being that the government can actually control the dollar, so there's a built-in bias for government/regulators in favour of a fiat currency that they can control by definition over a cryptocurrency that they (by default) cannot control at all. If/when cryptocurrency begins to threaten fiat currencies as a real alternative, and thus government start to lose control over the currency used and over their own monetary policy, that will be an interesting time to witness.

Aside from regulation, other existential threats that I can see:

• some major development that undermines scarcity and makes mining too easy/cheap; unlimited energy, some quantum computing advances maybe, etc.; doesn't seem likely in the short-term, and bitcoin surviving this long has vetted the technology pretty well;
• the ability for one politically-coherent organization or group to gain the majority of computing power in the mining pol, by which they gain control over the currency (something that happened in the past and could happen again if China decides to coordinate its mining pools, which constitute 60% of the computational power underlying Bitcoin, in order to stage a "take-over" of the currency, but that would crash the currency and it's not clear what they would gain from that);
• the popularisation of a cleaner alternative that wastes less energy while filling the same niche, or that lowers transactions times or costs; i.e., an alternative cryptocurrency that is clearly and strictly better than bitcoin by a margin that justifies the costs of switching over, making Bitcoin obsolete.

Will quit with the wall of posts now, I promise. First day back to work after holidays, and this is fertile ground to procrastinate on.

[starla]

This is because when the costs of mining go down, and more miners come online, the system adjusts to makes it artificially harder to mine bitcoins, and vice versa. The difficulty and thus resources required to mine are adjusted so that there's an average of a 10 minute delay between blocks. So all that should be in balance. This all means that neither the supply nor market value of bitcoin depend much on computer or energy costs, nor on the number of miners in general. Supply of bitcoin has been defined transparently since its inception: one can predict, more or less, how much there will be at any point, and that it will max out at 21 million BTC.

Ok so there is a central bank. But who is in charge of it?

Also, what is to stop some rich person from buying a supercomputer and hoarding it? Or is this happening already?

[Utisz]

This is because when the costs of mining go down, and more miners come online, the system adjusts to makes it artificially harder to mine bitcoins, and vice versa. The difficulty and thus resources required to mine are adjusted so that there's an average of a 10 minute delay between blocks. So all that should be in balance. This all means that neither the supply nor market value of bitcoin depend much on computer or energy costs, nor on the number of miners in general. Supply of bitcoin has been defined transparently since its inception: one can predict, more or less, how much there will be at any point, and that it will max out at 21 million BTC.

Ok so there is a central bank. But who is in charge of it?

In terms of these criteria, nobody.

When they designed and implemented Bitcoin, Satoshi decided that each block that was mined would be rewarded with 50 bitcoins, and that a block could only be mined on average every 10 minutes. That was a design choice he made. To do that, if folks started mining blocks faster than every 10 minutes, the mining process would be artificially made more difficult (fewer "keys" work), thus controlling the rate no matter how many miners were active or how efficient they were. He also decided that the number of bitcoins rewarded would half every 210,000 blocks. So now we are currently on a 50/2 = 25/2 = 12.5/2 = 6.25 BTC reward for bitcoins. All of these rules were defined transparently since the outset, and have been followed since. We've known from the outset that this process will generate at most 21 million bitcoin, of which over 18 million already exist. So no central bank: everything (in terms of supply) was defined at the outset and made transparent.


In terms of who is in charge of which transactions become part of the block chain, and are thus accepted, that would be the network of miners. If you control more than 50% of the computing power, theoretically you could start to accept/reject particular transactions, for example, because you can find keys faster than the rest of the network. This is why it's a big problem for Bitcoin if there's one group with more than 50% of the computing power in terms of mining. They could control which transactions become park of the blockchain, and thus the official history. But they could not change Satoshi's criteria, at least not without it becoming something other than Bitcoin.